Edge IoT Industrial Immersive and Spatial Computing Applications
Abstract
This position paper on “Edge IoT Industrial Immersive and Spatial Computing Applications” aims to provide a comprehensive overview of the convergence between IoT, AI, edge, and spatial computing, and how they are applied to various industrial immersive applications across different industrial sectors. It details the transformative impact of these applications across a wide range of industrial sectors, including culture and heritage, manufacturing, automotive, energy, construction, mobility, logistics, healthcare, agriculture, tourism, education and training. For each sector, the paper presents a thorough analysis of specific application scenarios. It identifies the key users and stakeholders involved, describes how these applications are implemented within a virtual world, and outlines the significant added value they bring to the industry. Furthermore, it specifies the required immersive technology functionalities, details the underlying technology layer requirements, and examines how cross-cutting horizontal issues manifest in each specific context.
Full text
Alliance for IoT, AI and Edge Continuum Innovation 2025 AIOTI WG Research Edge IoT Industrial Immersive and Spatial Computing Applications
AIOTI. All rights reserved. Edge IoT Industrial Immersive and Spatial Computing Applications Release 1 AIOTI WG Research 18 September 2025
© AIOTI. All rights reserved. 2 1 Executive Summary Immersive technologies, including virtual reality (VR), augmented reality (AR), mixed reality (MR), and extended reality (XR), alongside advanced concepts such as digital twins (DT), immersive triplets (IMT), the metaverse, omniverse, and spatial computing, represent a significant shift in industrial applications across sectors like culture and heritage, manufacturing, automotive, energy, buildings/construction, mobility. transportation, logistics, healthcare, agriculture/farming, tourism, education and training. The convergence of these immersive technologies with edge IoT, artificial intelligence (AI), and advanced intelligent connectivity infrastructure is shaping an industrial real-digital-virtual continuum, termed the "Phygital" world. By combining real-world interactions and virtual simulations, industries achieve improved operational efficiency, reduced downtime, enhanced safety protocols, and superior decision-making capabilities [24]. Edge IoT industrial immersive technologies require extensive interdisciplinary collaboration and robust infrastructure, from advanced computing platforms to advanced sensors and haptic devices. Real-time, high-performance processing capabilities, alongside reliable and secure connectivity with very low latency, are fundamental. As these technologies mature, standardisation, interoperability, trustworthiness, ethics, and sustainability become critical considerations, shaping global regulatory frameworks and industry standards. Developing immersive applications requires managing risks associated with data protection, privacy, AI ethics, and technological convergence while fostering innovation and growth. This position paper on “Edge IoT Industrial Immersive and Spatial Computing Applications” aims to provide a comprehensive overview of the convergence between IoT, AI, edge, and spatial computing, and how they are applied to various industrial immersive applications across different industrial sectors. It details the transformative impact of these applications across a wide range of industrial sectors, including culture and heritage, manufacturing, automotive, energy, construction, mobility, logistics, healthcare, agriculture, tourism, education and training. For each sector, the paper presents a thorough analysis of specific application scenarios. It identifies the key users and stakeholders involved, describes how these applications are implemented within a virtual world, and outlines the significant added value they bring to the industry. Furthermore, it specifies the required immersive technology functionalities, details the underlying technology layer requirements, and examines how cross-cutting horizontal issues manifest in each specific context. Beyond the sector-specific applications, the paper dedicates chapters to critical horizontal topics such as trustworthiness, ethics, sustainability, standardisation, and interoperability. For each of these areas, it identifies the primary challenges to successful implementation and discusses future research trends and directions, offering a forward-looking perspective on the evolution of these technologies. The goal of this approach is to create a holistic document that showcases the current state and potential of industrial immersive applications and provides a strategic roadmap. It highlights the technological requirements and addresses the crucial non-technical challenges, serving as a foundational resource for researchers, innovators, and industry leaders navigating this rapidly advancing field.
© AIOTI. All rights reserved. 3 Table of Content 4.1 Immersive Cultural Events Platform with Holographic Presence ..................................... 25 4.1.1 Scenario .................................................................................................................................. 25 4.1.2 Users and stakeholders ......................................................................................................... 26 4.1.3 Implementation in a virtual world and added values ...................................................... 26 4.1.4 Required immersive technologies functionalities .............................................................. 26 4.1.5 Technology layers requirements .......................................................................................... 27 4.1.6 Horizontal issues and characteristics ................................................................................... 28 4.2 IoT-based Management of Tangible Cultural Heritage Assets ....................................... 28 4.2.1 Scenario .................................................................................................................................. 28 4.2.2 User and stakeholders ........................................................................................................... 30 4.2.3 Implementation in a virtual world and added values ...................................................... 30 4.2.4 Required immersive technologies functionalities .............................................................. 30 4.2.5 Technology layers requirements .......................................................................................... 30 4.2.6 Horizontal issues and characteristics ................................................................................... 30 5.1 Industrial Metaverse ........................................................................................................... 31 5.1.1 Scenario .................................................................................................................................. 31 5.1.2 Users and stakeholders ......................................................................................................... 32 5.1.3 Implementation in a virtual world and added values ...................................................... 32 5.1.4 Required immersive technologies functionalities .............................................................. 33 5.1.5 Technology layers requirements .......................................................................................... 34 5.1.6 Horizontal issues and characteristics ................................................................................... 34
© AIOTI. All rights reserved. 4 5.2 Robotic Welding ................................................................................................................. 35 5.2.1 Scenario .................................................................................................................................. 35 5.2.2 Users and stakeholders ......................................................................................................... 35 5.2.3 Implementation in a virtual world and added values ...................................................... 36 5.2.4 Required immersive technologies functionalities .............................................................. 36 5.2.5 Technology layers requirements .......................................................................................... 37 5.2.6 Horizontal issues and characteristics ................................................................................... 37 5.3 Assistance for Equipment Servicing and Maintenance ................................................... 39 5.3.1 Scenario .................................................................................................................................. 39 5.3.2 Users and stakeholders ......................................................................................................... 40 5.3.3 Implementation in a virtual world and added values ...................................................... 40 5.3.4 Required immersive technologies functionalities .............................................................. 40 5.3.5 Horizontal issues and characteristics ................................................................................... 41 5.4 Tele-repair and Remote Maintenance ............................................................................. 41 5.4.1 Scenario .................................................................................................................................. 41 5.4.2 Users and stakeholders ......................................................................................................... 42 5.4.3 Implementation in a virtual world and added values ...................................................... 42 5.4.4 Required immersive technologies functionalities .............................................................. 43 5.4.5 Technology layers requirements .......................................................................................... 43 5.4.6 Horizontal issues and characteristics ................................................................................... 44 5.5 Circularity in Manufacturing with XR and AI ..................................................................... 44 5.5.1 Scenario .................................................................................................................................. 44 5.5.2 Users and stakeholders ......................................................................................................... 45 5.5.3 Implementation in a virtual world and added values ...................................................... 45 5.5.4 Required immersive technologies functionalities .............................................................. 46 5.5.5 Technology layers requirements .......................................................................................... 46 5.5.6 Horizontal issues and characteristics ................................................................................... 46 5.6 XR-Based Human-Robot Collaboration Along a Conveyor Picking Line: The case of Construction and Demolition Waste Sorting ................................................................................ 47 5.6.1 Scenario .................................................................................................................................. 47
© AIOTI. All rights reserved. 5 5.6.2 User and stakeholders ........................................................................................................... 47 5.6.3 Implementation in a virtual world and added values ...................................................... 47 5.6.4 Required immersive technologies functionalities .............................................................. 48 5.6.4.1 User Feedback Mechanisms ............................................................................................ 48 5.6.5 Technology layers requirements .......................................................................................... 48 5.6.6 Horizontal issues and characteristics ................................................................................... 49 6.1 Meta-Factory Model for Electrified Braking System ......................................................... 50 6.1.1 Scenario .................................................................................................................................. 50 6.1.2 Users and stakeholders ......................................................................................................... 51 6.1.3 Implementation in a virtual world and added values ...................................................... 51 6.1.4 Required immersive technologies functionalities .............................................................. 51 6.1.5 Technology layers requirements .......................................................................................... 52 6.1.6 Horizontal issues and characteristics ................................................................................... 52 7.1 Immersive VR for Operation of Wind Turbines .................................................................. 53 7.1.1 Scenario .................................................................................................................................. 53 7.1.2 User and stakeholders ........................................................................................................... 54 7.1.3 Implementation in a virtual world and added values ...................................................... 54 7.1.4 Required immersive technologies functionalities .............................................................. 54 7.1.5 Technology layers requirements .......................................................................................... 54 7.1.6 Horizontal issues and characteristics ................................................................................... 54 8.1 Virtual Architectural Design ............................................................................................... 55 8.1.1 Scenario .................................................................................................................................. 55 8.1.2 User and stakeholders ........................................................................................................... 55 8.1.3 Implementation in a virtual world and added values ...................................................... 56 8.1.4 Required immersive technologies functionalities .............................................................. 56 8.1.5 Technology layers requirements .......................................................................................... 56 8.1.6 Horizontal issues and characteristics ................................................................................... 56
© AIOTI. All rights reserved. 6 9.1 Collaborative Aircraft Cockpit Design .............................................................................. 57 9.1.1 Scenario .................................................................................................................................. 57 9.1.2 Users and stakeholders ......................................................................................................... 58 9.1.3 Implementation in a virtual world and added values ...................................................... 58 9.1.4 Required immersive technologies functionalities .............................................................. 59 9.1.5 Technology layers requirements .......................................................................................... 59 9.1.6 Horizontal issues and characteristics ................................................................................... 60 10.1 XR-Based Healthcare Cybersecurity Digital Twin ............................................................ 61 10.1.1 Scenario .................................................................................................................................. 61 10.1.2 User and stakeholders ........................................................................................................... 62 10.1.3 Implementation in a virtual world and added values ...................................................... 62 10.1.4 Required immersive technologies functionalities .............................................................. 62 10.1.5 Technology layers requirements .......................................................................................... 62 10.1.6 Horizontal issues and characteristics ................................................................................... 63 11.1 XR-based Agricultural Digital Twin .................................................................................... 64 11.1.1 Scenario .................................................................................................................................. 64 11.1.2 Users and stakeholders ......................................................................................................... 65 11.1.3 Implementation in a virtual world and added values ...................................................... 65 11.1.4 Required immersive technologies functionalities .............................................................. 66 11.1.5 Technology layers requirements .......................................................................................... 67 11.1.6 Horizontal issues and characteristics ................................................................................... 67 11.2 Immersive Agricultural Digital Twin ................................................................................... 68 11.2.1 Scenario .................................................................................................................................. 68 11.2.2 User and stakeholders ........................................................................................................... 69 11.2.3 Implementation in a virtual world and added values ...................................................... 69 11.2.4 Required immersive technologies functionalities .............................................................. 69 11.2.5 Technology layers requirements .......................................................................................... 69 11.2.6 Horizontal issues and characteristics ................................................................................... 69
© AIOTI. All rights reserved. 7 11.3 Smart Agriculture: Precision Farming ................................................................................ 70 11.3.1 Scenario .................................................................................................................................. 70 11.3.2 User and stakeholders ........................................................................................................... 70 11.3.3 Implementation in a virtual world and added values ...................................................... 71 11.3.4 Required immersive technologies functionalities .............................................................. 71 11.3.5 Technology layers requirements .......................................................................................... 71 11.3.6 Horizontal issues and characteristics ................................................................................... 71 12.1 Integration of Digital Twins in the Natural Reserve Laukvikøyene .................................. 72 12.1.1 Scenario .................................................................................................................................. 72 12.1.2 User and stakeholders ........................................................................................................... 73 12.1.3 Implementation in a virtual world and added values ...................................................... 73 12.1.4 Required immersive technologies functionalities .............................................................. 74 12.1.5 Technology layers requirements .......................................................................................... 74 12.1.6 Horizontal issues and characteristics ................................................................................... 74 12.1.7 Required immersive technologies functionalities .............................................................. 74 12.1.8 Horizontal issues and characteristics ................................................................................... 74 13.1 Training for Compressor Assembly .................................................................................... 75 13.1.1 Scenario .................................................................................................................................. 75 13.1.2 Users and stakeholders ......................................................................................................... 75 13.1.3 Implementation in a virtual world and added values ...................................................... 75 13.1.4 Required immersive technologies functionalities .............................................................. 76 13.1.5 Technology layers requirements .......................................................................................... 76 13.1.6 Horizontal issues and characteristics ................................................................................... 77 13.2 Grundfos Machine Operation and Safety Training .......................................................... 77 13.2.1 Scenario .................................................................................................................................. 77 13.2.2 Users and stakeholders ......................................................................................................... 78 13.2.3 Implementation in a virtual world and added values ...................................................... 78 13.2.4 Required immersive technologies functionalities .............................................................. 78
© AIOTI. All rights reserved. 8 13.2.5 Technology layers requirements .......................................................................................... 79 13.2.6 Horizontal issues and characteristics ................................................................................... 79 13.3 DSB Train Operator Training ............................................................................................... 80 13.3.1 Scenario .................................................................................................................................. 80 13.3.2 Users and stakeholders ......................................................................................................... 80 13.3.3 Implementation in a virtual world and added values ...................................................... 80 13.3.4 Required immersive technologies functionalities .............................................................. 81 13.3.5 Technology layers requirements .......................................................................................... 81 13.3.6 Horizontal Issues and Characteristics .................................................................................. 81 13.4 MR Incident Simulator for immersive Command and Control Room Training ................ 82 13.4.1 Scenario .................................................................................................................................. 82 13.4.2 User and stakeholders ........................................................................................................... 82 13.4.3 Implementation in a virtual world and added values ...................................................... 82 13.4.4 Required immersive technologies functionalities .............................................................. 82 13.4.5 Technology layers requirements .......................................................................................... 83 13.4.6 Horizontal issues and characteristics ................................................................................... 83 13.5 XR-based Remote Collaboration with IoT Contextual Integration .................................. 83 13.5.1 Scenario .................................................................................................................................. 83 13.5.2 User and stakeholders ........................................................................................................... 84 13.5.3 Implementation in a virtual world and added values ...................................................... 84 13.5.4 Required immersive technologies functionalities .............................................................. 84 13.5.5 Technology layers requirements .......................................................................................... 84 13.5.6 Horizontal issues and characteristics ................................................................................... 84 13.6 Advanced VR simulator for Training Law Enforcement Officers ...................................... 84 13.6.1 Scenario .................................................................................................................................. 84 13.6.2 User and stakeholders ........................................................................................................... 84 13.6.3 Implementation in a virtual world and added values ...................................................... 85 13.6.4 Required immersive technologies functionalities .............................................................. 85 13.6.5 Technology layers requirements .......................................................................................... 85 13.6.6 Horizontal issues and characteristics ................................................................................... 86
© AIOTI. All rights reserved. 15 List of Keywords 3D 5G 6G Aeronautics Agriculture AI assistant AI object detection Aircraft design Artificial intelligence Asset location Augmented reality Automotive Braking system Circular economy Cockpit design Collaborative design Collaborative innovation Collaborative training Compliance Compressor assembly Computer vision Construction demolition waste Conveyor systems Cultural events Computer vision Cybersecurity Data visualisation Digital assets Digital twin Door operation Edge computing Employee training Equipment maintenance Ergonomics Ethical Explainable AI Extended reality Farm management Farming Haptic feedback Head tracking Healthcare Hololens Holographic presence HTC Vive Human-machine interaction Human-machine interface Human-robot collaboration Human-robot interaction Immersive virtual reality Indoor positioning Industrial metaverse Industrial training Industry 4.0 Industry 5.0 Intellectual property Internet of things Interoperability Large language model Live performance Low latency Machine operation Manufacturing Meta quest Meta-factory Metaverse Mixed reality Network slicing Oculus quest Pose tracking Predictive maintenance Procedural learning Production downtime Railway simulation Remote collaboration Remote maintenance Remote training Robotics Robotic welding Robot Operating System Safe pathfinding Safety procedures Sensor network Smart factory Smart farming Standardisation Sustainability Sustainability in manufacturing Tele-repair Telemetry Telerobotics Tourism Train operator training Ultra-wideband Unity 3D engine Virtual commissioning Virtual prototyping Virtual reality Virtual reality training Waste sorting WEAVR platform Worker assistance
© AIOTI. All rights reserved. 16 2 Goal Fusion and convergence of technologies sparks innovation, allowing for cross-pollination of ideas, the creation of novel approaches, and transforms industrial landscape by enabling new edge IoT devices, systems, services, and business models. Developing cutting-edge IoT industrial immersive technologies and spatial computing requires a holistic, interdisciplinary approach as a driver of knowledge creation, research, and innovation. Collaboration between the disciplines is thus a vital complement to the grow of the disciplines themselves [24]. IoT and edge computing research and innovation address IoT/edge continuum distributed architectures, intelligent connectivity. End-to-end (E2E) security, heterogenous IoT edge mesh, IoT DTs, AI, IoT swarm systems, Internet of Things Senses (IoTS), trustworthiness, verification, validation, and testing (VV&T), standardisation, and the convergence of all the above into the Internet of Intelligent Things. Immersive technology refers to any technology that blurs the line between the physical and digital worlds, creating a sense of presence and engagement for the user. Immersive technologies aim to transport users to virtual environments or enhance their real-world experiences by overlaying digital information onto their physical surroundings through real-time interactions in physical, digital, virtual, cyber, and spatial environments. Immersive technologies have emerged as a revolutionary approach to creating digital experiences that feel real to users. By incorporating various tools and systems, immersive technology encompasses real-time interactions in physical, digital, virtual, cyber, and spatial environments using a broad spectrum of experiences that blur the boundaries between the physical and digital worlds, providing innovative ways to interact, explore, and learn. IoT and edge computing enable innovation and broad adoption in immersive technologies and applications by bringing the novel elements of converging technologies to the edge and realtime interaction between the physical and virtual worlds. Web 4.0 embodies a new era of the Internet, conceived as a decentralised online ecosystem founded on blockchain technology. Unlike the present Internet version Web 2.0, dominated by centralised platforms and services owned by a handful of large corporations, Web 4.0 aims to return control and ownership to the users. The technological advances it brings, have the potential to profoundly change the way one interacts with the digital realm, creating a more open, transparent, and user-empowered internet. The future of technology, particularly with the convergence of IoT, edge computing, AI, and industrial immersive technologies, is poised for groundbreaking developments, especially when integrated with emerging concepts like the metaverse, the omniverse, the multiverse, and Internet Web 4.0. The future version of the Internet is expected to be more autonomous, intelligent, and seamlessly integrated into everyday objects. It could leverage blockchain for security and decentralisation, facilitate microtransactions within the various verses, and support sophisticated AI-driven interactions. The convergence of these technologies signifies a technological shift, as well as a cultural and economic one, potentially altering how we perceive and interact with the digital and physical worlds. This convergence promises a more integrated, immersive, and interactive future.
© AIOTI. All rights reserved. 17 The deployment of immersive technologies in real-world applications necessitates addressing the challenge of ensuring end-to-end security across diverse and interconnected systems, considering the heterogeneous nature of the IoT edge systems, and the goal of interoperable integration. Ensuring the trustworthiness and reliability of industrial immersive applications through traditional verification, validation, and testing poses a significant challenge. From an economic standpoint, the initial investment required to deploy immersive and edge IoT infrastructure can be a significant barrier for many organisations. The costs associated with scaling, maintaining, and upgrading the industrial immersive systems also present ongoing economic challenges. This leads to the need for carefully planning and choosing the IoT, AI, communication and platform technologies used by various industrial applications. In addition, industrial immersive applications require professionals with interdisciplinary expertise to develop and manage these converged technologies. Today's industrial immersive applications market is fragmented, with competition between various proprietary systems, which can prevent widespread, standardised adoption and create resistance from industries hesitant to disrupt their existing workflows. The challenges presented require a clear focus on standardisation, interoperability, and the design, development, implementation and deployment of various industrial immersive applications across different industrial sectors involving interdisciplinary ecosystems to facilitate collaboration and cooperation. Technically, the convergence of AI, IoT and edge computing allows for distributed architectures that support intelligent connectivity and real-time interactions that can be applied to industrial immersive applications. 5G and future 6G private networks (NPNs) are pivotal for advancing immersive industrial applications by providing large industries the autonomy to deploy and manage their own dedicated network infrastructure. This localised control ensures reliable, low-latency, and highbandwidth connectivity, which is essential for data-intensive applications like AR, VR, and the real-time control of autonomous streams. By operating a private network, industries can guarantee security, customise network performance for specific operational needs, and ensure that mission-critical immersive services are not compromised by the congestion of public networks, thereby accelerating the adoption of innovative solutions across industrial domains. The development of IoT digital twins, immersive triplets and swarm systems offers effective new ways to simulate, monitor, and optimise industrial operations and integrate immersive technologies in real-time processes, while AI-driven analytics enhance decision-making and create more responsive user experiences. Economically, the rise of Web 4.0 and blockchain technology opens the door to decentralised business models and secure microtransactions within immersive industrial environments like the metaverse. Significant cost savings can be realised through predictive maintenance, remote collaboration, and the automation of complex operations, while new revenue streams can be created from innovative, value-added immersive services. Across various markets, the industrial immersive applications unlock transformative approaches. In manufacturing, immersive training and remote assistance can improve safety and efficiency. Healthcare can leverage immersive tech for advanced diagnostics, therapeutic interventions, and surgical training. The emergence of the metaverse, omniverse and multiverse is creating new ecosystems for commerce, social interaction, and entertainment, fundamentally altering how people and machines engage with the physical, digital and cyber worlds.
© AIOTI. All rights reserved. 18 Industrial immersive applications, powered by the convergence of AI, IoT, edge, and spatial computing, are fundamental to the digitisation and automation of industry. The adoption of these integrated industrial immersive applications is a direct driver of increased competitiveness. By using immersive environments for remote assistance and training, companies can drastically reduce downtime and travel costs, bringing expert knowledge to any location instantly. The integration of immersive applications into manufacturing processes fosters agile and responsive operations, enabling manufacturers to reconfigure production lines virtually and test new products' feasibility without disrupting the actual factory floor. Industrial immersive technologies are serving as a critical experimental laboratory for the nextgeneration vision Internet, called Web 4.0, and the broader development of virtual worlds. The demands of industrial applications for high-fidelity digital twins, immersive triplets, real-time collaboration, and high reliability are pushing the boundaries of what is possible, setting a high bar for performance that will eventually transition to consumer experiences. The solutions engineered for these immersive environments are laying the groundwork for the next-generation interactive Internet. Industry is pioneering the concept of digital twins and immersive triplets that are continuously synchronised with their real-world counterparts. The technical expertise acquired in building and maintaining these complex systems of systems, always-on virtual environments, provides a reference for creating scalable and dynamic shared virtual worlds. These industrial metaverses, omniverses, and multiverses built for mission-critical purposes are becoming the foundational testbeds for the technologies and practices that will define more expansive and socially-focused virtual realms. The high trust and identity, including security, reliability, resilience, robustness, privacy and several other quality requirements of industrial settings, are accelerating solutions that will be vital for a trustworthy Web 4.0. When controlling critical infrastructure or handling sensitive content in an immersive environment, robust methods for verifying identity and securing data are paramount. These high-stakes security models, forged to prevent industrial interference or operational failure, can provide the secure framework needed to manage digital identity, asset ownership, and economic transactions in the broader virtual economies of the future. The collaborative potential of future concepts like the metaverse, omniverse and multiverse profoundly accelerates innovation. These shared virtual spaces serve as persistent, interactive platforms where globally dispersed teams of engineers, designers, and stakeholders can cocreate and innovate in real-time. The virtual collaboration and interaction using industrial immersive applications facilitates rapid prototyping, instant feedback loops, and extensive testing in simulated environments, dramatically reducing the time and cost required to bring novel products from concept to market. Looking broader, the development of Web 4.0, a decentralised and intelligent internet, provides the foundational trust and economic layer for this new industrial paradigm. Its decentralised nature, often built on blockchain, can ensure secure and transparent data exchange between countless IoT devices, autonomous systems, and different company platforms, fostering interoperability and trust in multi-stakeholder ecosystems. This can create a secure framework for a machine-to-machine economy, enhance supply chain transparency, and support a more sustainable future by enabling precise resource management. The future of work is also reshaped, empowering a distributed workforce with immersive tools that make remote collaboration as effective as being physically present, ensuring continuous innovation and economic growth.
© AIOTI. All rights reserved. 19 3 The Industrial Immersive Continuum Human behaviour occurs in the time and space of the physical world. Spatial-temporality ensures that human behaviour proceeds normally and limits human behaviour from deviating from its norms. The concept of hyper-spatial-temporality has been mentioned and introduced in Chinese mythology, Taoism, and some Western philosophical theories. Human imagination and exploration of virtual worlds have led to the gradual conceptualisation of hyper-spatialtemporality [6]. The convergence of disruptive technologies such as the Internet of Things (IoT), artificial intelligence (AI), digital twins (DT), and spatial computing is forging a complex and powerful industrial real-digital-virtual continuum. This continuum represents a new paradigm for digital transformation, where the boundaries between the physical and virtual worlds are increasingly blurred. It is realised through immersive environments, powered by augmented reality (AR), virtual reality (VR), mixed reality (MR), and extended reality (XR), which allow users to feel physically present and interact with digital content naturally. These technologies are the building blocks of future virtual worlds, including concepts like the metaverse and the nextgeneration spatial web (Web 4.0), and their application in industrial settings is poised to unlock unprecedented efficiencies and capabilities [24]. Industrial immersive solutions are moving beyond isolated applications to become integrated platforms that combine data from the real world with powerful simulations and interactive virtual environments. This enables a continuous flow of information between physical assets and their digital counterparts, allowing for more innovative design, more efficient operations, and more effective training. From the factory floor to the operating room, these technologies are fundamentally changing how we work, learn, and collaborate. This document provides a comprehensive overview of the advancements, challenges, and future research trends of immersive applications across key industrial sectors, highlighting the transformative potential of this new technological wave. Next generation virtual worlds will leverage recent and future developments in AI, XR, and IoT but also connectivity and infrastructure advances. While offering many open opportunities, future virtual worlds also come with many challenges including technical, societal, economic and legal ones. The opportunity vs risk trade-off must therefore be carefully addressed from the earliest stages of development and deployment. The opportunities offered by immersive applications for society and the economy are significant in several sectors [68]. Several key advancements have enabled the widespread adoption of immersive technologies across industries. The development of high-fidelity, lightweight, and untethered XR headsets has significantly improved user comfort and mobility, making them practical for use in dynamic industrial environments [1]. Advances in computer vision (CV) and AI-powered spatial mapping enable these devices to understand and interact with the physical world in real-time, allowing for context-aware AR overlays and seamless MR interactions. The integration of digital twin technology is a particularly impactful advancement, allowing companies to create highly detailed, data-rich virtual replicas of physical assets, processes, and systems. These digital twins can be used in immersive environments for simulation, monitoring, and predictive maintenance, providing immense value [64]. The maturation of cloud and edge computing infrastructure has provided the necessary computational power to render complex scenes and process large datasets with low latency, which is critical for collaborative and responsive immersive experiences. An example of applications of edge intelligence for the Metaverse is presented in [61]. The infrastructure layer leverages edge intelligence to support AI for the intelligent Metaverse (e.g., edge for AI) and utilise AI to realize the resource efficient collaborative edge paradigm (e.g., AI for edge) [61].
© AIOTI. All rights reserved. 20 Figure 3-1 Applications of Edge Intelligence for the Metaverse [61]. Despite this progress, several cross-cutting challenges impede the full realisation of industrial immersive applications. A primary hurdle is the need for robust standardisation and interoperability. Many current solutions exist in proprietary, siloed ecosystems, which limit the ability to share data and assets between different platforms and applications [29]. Another significant challenge is data integration. Industrial environments are complex and involve numerous legacy systems, sensors, and data formats. Integrating this disparate data into a cohesive and usable format for immersive applications requires significant effort and expertise. Network reliability, low latency and bandwidth are also critical concerns, especially for applications that require the real-time streaming of high-fidelity 3D data in remote or mobile settings [17]. Finally, user acceptance and workforce training remain crucial factors. Ensuring that immersive interfaces are intuitive and comfortable for long-term use, and that workers are adequately trained to use them effectively, are key to successful implementation. Immersive technologies are evolving and are better adapted to meet the needs of immersive experiences, which fosters a paradigm shift illustrated by three trends [51]: • The convergence of technologies like broader sensory spaces to integrate senses such as visual, auditory, kinesthetics, tactile, smell and taste, to ensure better sensory coverage and multisensory interaction, actuation, haptic evolving into a set of MR solutions, enabling both AR and VR. • The diversification of uses spans fields such as maintenance, health, safety, industrial production, assembly, and broader domains, including immersive leisure, immersive training, and immersive AI. • The emergence of environments that are sometimes realistic and concrete, or metaphorical and abstract, can facilitate access to and interaction with AI tools. The progress of AI is evident with the use of textual/vocal AI (chatbot, vocal assistant) and further generative AI, AI agents and agentic AI to support analysis or decision functionalities. Among the different immersive technologies, the Metaverse has evolved in the last years with applications in different industrial sectors [67].
© AIOTI. All rights reserved. 21 The Metaverse is composed of the words "meta" and "verse" (meta comes from Greek, meaning "transcending," and "verse" means "universe"). It is a parallel world closely connected to the real world, the product of the development and integration of various technologies, the next stage of Internet development, and a virtual living space with social attributes. The Metaverse's development represents a certain extent of the virtual world's development. Five stages in the evolution of virtual worlds are proposed in [30], from those that initially existed only in literature and games to today's immersive 3D virtual worlds, where users can create content independently [6]. An example of the Metaverse architecture [52] is illustrated in Figure 3-2, where human society is centred around users who interact with digital avatars through smart wearable devices and technologies like human-computer interaction and XR. Figure 3-2 Metaverse Architecture – Combination Human, Physical, and Digital Realms (Source: Adapted from [52]). The physical infrastructure facilitates data perception, transmission, processing, and physical control through smart objects, sensors, and diverse networks. These infrastructures assist the interaction between the digital and human worlds. The digital world comprises interconnected sub-metaverses, offering users a range of virtual goods/services and environments. The metaverse engine leverages this interactivity to generate, maintain, and update the virtual world using data from the real world, AI, digital twins, and blockchain (BC) technologies to ensure the richness and sustainability of the metaverse ecosystem. In the metaverse, information flows freely across each world, whether human, physical, digital, virtual, or cyber, driven by social networks, IoT infrastructure, and the metaverse engine. The IoT devices bridge these worlds, facilitating the interaction between the physical and digital realms and allowing seamless information flow [52]. The Industrial Metaverse refers to the establishment of a shared virtual space empowered by Metaverse technologies, to support multi-user, multi-device industrial scenarios for 3D modelling and immersive interaction, and used for product design, production operations, industrial quality inspection, and product testing for industrial production [53] driven by key enabling technologies, including blockchain, digital twins, immersive triplets, IoT, 5G/6G, immersive technologies (AR, VR, XR, MR, Web 4.0) and AI.
© AIOTI. All rights reserved. 22 The reference architecture of the industrial Metaverse in industrial scenarios described in [53] is presented in Figure 3-3. The architecture includes three layers: the data input layer, the enabling layer, and the industrial application layer. The enabling layer comprises six components: AI, DT, BC, XR, the Metaverse management centre, and the data processing system. Figure 3-3 Industrial Metaverse Architecture [53]. Building information modelling (BIM) is used in the architecture and building industry to enhance the quality of documentation produced, as well as improve constructability. New technological developments combine BIM with immersive technologies like AR to enable the physical context of each construction activity or task to be visualised in real-time, using AR ubiquitously (including context awareness) and thus operate in conjunction with tracking and sensing technologies [37].
© AIOTI. All rights reserved. 23 The Metaverse presents a compelling and transformative vision for the future of communication systems, marked by greater immersion, interactivity, and inclusivity. It is crucial to develop a robust understanding of the Metaverse’s technical requirements, societal implications, and economic potential. This understanding is key to ensuring its successful integration into daily lives and leveraging it for the broader benefits of society [54]. Initially, the Metaverse has emerged from the convergence of three major digital technologies: gaming, AR/VR, and Web3. The technological developments among these major digital technologies have resulted in platform ecosystems consisting of connected but unrelated stakeholders from different backgrounds, and complex systems comprising components from various sectors that are converging. The new wave of immersive technologies addresses the integration of 6G-enabled edge AI and Metaverse, different types of edge-Metaverse architectures that use 6G-enabled edge AI to solve resource and computing constraints in Metaverse [63]. 6G edge intelligence has the advantages of low latency, computing offload, and high performance while bringing highaccuracy positioning supporting immersive solutions. The application of 6G-oriented edge intelligence offers several benefits, including balanced data storage, efficient data transmission, and high reliability with very low latency. Figure 3-4 shows an example of an architecture of the Metaverse, which includes physical layer, virtual layer, and technical layer that supports the real-time interaction of users in the physical-virtual world [63]. Figure 3-4 Metaverse Architecture - Real-Time Physical-Virtual World Interaction Supported by Low Bandwidth, Low Latency, Ubiquitous Access, and Trustworthiness [63]. The future immersive applications are expected to use the 6G mobile communication technology has several advantages such as high performance, global coverage, real-time processing, high reliability, and energy efficiency. Table 3-1shows a comparison between 5G and 6G technologies [63] with specific performance indicators that are key for immersive applications.
© AIOTI. All rights reserved. 24 Table 3-1 Performance indicators comparison between 6G and 5G [63]. Performance Indicators 5G 6G Peak transmission rate 10 ∼ 20 Gb/s 100 Gb/s ∼ 1 Tb/s User experience rate 0.1 ∼ 1 Gb/s 30 ∼ 50 Gb/s Time delay 10 ∼ 50 ms 0.1 ∼ 1 ms Reliability 10−5 10−9 Flow density 10 Tbps/km2 100 ∼ 1000 Tbps/km2 Positioning precision 1 ∼ 10 m 0.1 ∼ 1 m Connection density 1 million/km3 10 ∼ 100 million/km3 Network efficiency 100 bit/s 200 bit/s Mobility 500 km/h 1000 km/h Spectrum bandwidth 30 ∼ 100 bps/Hz 200 ∼ 500 bps/Hz Base station computing power 100 ∼ 200 Tops 1000 Tops Coverage Partial Global Security Patchy security Endogenous security Information timeliness High Extremely high As the technologies have developed, it has become evident that the interplay of these technologies, along with the popularity and significance of platform ecosystems, necessitates the development of technology standards at multiple levels, from data to functional interfaces and protocols, by building on a foundation of open standards. The technical standards applicable in the Metaverse are established norms or rubrics for the common and repeated use of rules, conditions, guidelines, or characteristics for products or related processes and production methods, as well as related management system practices. Responsible standards development is a required component of the commercial research and innovation process. The concept of responsible innovation has been described as an inclusive and risk-mitigating approach to research and innovation. It aims to ensure that unintended negative impacts are avoided, that barriers to dissemination, adoption and diffusion of research and innovation are reduced, and that the positive societal and economic benefits of research and innovation are realised [55][56]. Detailed discussions of standardisation activities related to immersive technologies, including the Metaverse, are presented in [57][59][29][24] and in the chapter covering standardisation and interoperability. The future of industrial immersive applications will be defined by greater intelligence, deeper integration, and seamless interoperability. The next generation of these systems will be powered by increasingly sophisticated AI, enabling environments that are not just interactive but are truly responsive and adaptive to user behaviour and context. The concept of the immersive triplet, which links the physical asset, its real-time digital twin, and an AI-driven predictive model, will become more prevalent, allowing for proactive maintenance and operational optimisation. The rise of the IoT of Senses (IoTS) will enable richer multi-sensory experiences, incorporating haptics, and even smell and taste, to create a more profound sense of presence. A primary focus of future research is on developing open, interoperable standards necessary for a true industrial metaverse, enabling seamless transfer of data, assets, and avatars between platforms. The development of the next-generation spatial web (Web 4.0) will provide the decentralised infrastructure needed for this vision. As these technologies become increasingly integrated into critical industrial processes, ensuring their trustworthiness, security, and ethical use will be of paramount importance. The industrial real-digital-virtual continuum is still in its early stages, but its potential to enhance human capability, optimise complex systems, and drive the next wave of digital transformation is undeniable.
© AIOTI. All rights reserved. 31 5 Manufacturing The manufacturing sector is utilising immersive technologies to create the "factory of the future." Digital twins of entire production lines are visualised and manipulated in VR to optimise layouts, simulate production flows, and identify potential bottlenecks before any physical equipment is installed or altered [37]. The "virtual commissioning" drastically reduces setup times and production risks. For worker training, VR offers a safe and repeatable environment in which to learn how to operate complex machinery or perform hazardous tasks without risk to the individual or the equipment. Remote assistance via AR is another key application, where an expert located anywhere in the world can see what a field technician sees and provide real-time guidance by annotating the technician's view, significantly reducing machine downtime and travel costs [38]. The inclusion of VR/AR technologies in manufacturing processes has the potential to benefit significantly adopters, especially in terms of increased efficiency and productivity, thanks to traditional processes being carried out in a cheaper and faster way. Employees can also benefit from VR/AR adoption, including higher-quality training, increased workplace security, and opportunities to upskill their digital knowledge. VR/AR can revolutionise many manufacturing processes. Assembly and maintenance enhanced by AR information can minimise mistakes while reducing the need for on-site support for experts, especially in combination with IoT and AI technologies. Personnel getting at-a-glance details on a specific product or material could also be more efficient in raw material preparation, and in setting up production processes [65]. Product development already highly benefits from VR, since industrial product design allows for a more intuitive design phase and an easier customer−manufacturer interaction. The utilisation of 3D models or DT during the prototyping phase brings numerous benefits. The creation of refined virtual product models is instrumental in bridging the gap between design and manufacturing and reducing material waste, which is noticeable in cost and time savings. Collaborative working in manufacturing is an emerging application where several actors are usually engaged during the first stages of prototyping and product design. Meetings conducted in a 3D and interactive environment allow design teams from different physical locations to come together in photorealistic virtual spaces, minimising the reliance on physical models and prototypes. Remote collaboration is evolving in applications such as remote guidance and activity supervision across various manufacturing sectors [65]. 5.1 Industrial Metaverse 5.1.1 Scenario The Industrial Metaverse represents a paradigm shift in the domain of industrial operations and maintenance. It is a concept that involves the deep integration of the IoT, AI, and a spectrum of immersive technologies including VR/AR/MR. The primary objective is to revolutionize how industrial environments are managed and operated.
© AIOTI. All rights reserved. 32 Figure 5-1 –Industrial Metaverse The application enables the creation of comprehensive virtual simulations of products, processes, and entire factories. It allows for real-time monitoring of physical assets through their digital twins, facilitates advanced predictive maintenance to anticipate failures, and supports seamless remote collaboration between teams. By providing a rich, data-driven virtual environment, the Industrial Metaverse supports enhanced decision-making, helps create more efficient production workflows, and allows for real-time troubleshooting of complex issues. 5.1.2 Users and stakeholders The users of the Industrial Metaverse are diverse, spanning the entire lifecycle of industrial operations. This includes design engineers who can prototype and test in a virtual space, production managers who can simulate and optimise workflows, and maintenance technicians who can receive remote expert guidance or practice complex repairs in a safe, simulated environment. The key stakeholders are the industrial enterprises themselves. By adopting this technology, they can achieve significant improvements in operational efficiency, reduce costly downtime, and enhance overall product quality. The innovations fostered by the Industrial Metaverse are intended to create more sustainable practices and cultivate a highly adaptive and competitive industrial landscape for these organisations. 5.1.3 Implementation in a virtual world and added values The implementation of this concept is a Metaverse, a persistent, shared, and interactive virtual space that mirrors the physical industrial world. This is not a single application but an ecosystem of interconnected digital twins and simulations. The user interface within this world is built on detailed 3D elements, holographic displays, and interactive navigation systems that provide intuitive pathways to explore virtual factories and industrial layouts.
© AIOTI. All rights reserved. 33 The added value is transformative. The system directly contributes to sustainability by enabling virtual prototyping that reduces physical waste, optimizing resource usage through AI-driven analytics, and supporting remote collaboration to minimize travel-related emissions. It enhances safety by allowing training and problem-solving in a risk-free environment. Ultimately, these innovations lead to reduced downtime, improved product quality, and a more resilient and agile industrial operation. 5.1.4 Required immersive technologies functionalities The functional and technological foundation of the Industrial Metaverse is extensive and multilayered. User interaction is designed to be highly intuitive, utilising a range of modes including hand gestures for manipulating virtual tools, voice commands for hands-free operation, and eye tracking for gaze-based control. VR controllers offer precision for specific tasks, while emerging interfaces like Brain-Computer Interfaces (BCIs) are on the horizon. User feedback is delivered through realistic haptic feedback and AR displays that provide real-time data overlays. The system requires seamless HMI and IoT integration, leveraging data from motion, depth, and proximity sensors, along with environmental actuators for precise data collection. To process this vast amount of data and render complex simulations in real time, the architecture relies on lowlatency edge computing with high-performance CPUs and GPUs. This is supported by highspeed 5G or 6G networks to ensure seamless data transmission and real-time updates. Data management must support complex 3D and spatio-temporal data, with scalable local and cloud storage solutions. A core component is the integration of AI algorithms for predictive analytics, computer vision, and natural language processing. The entire system is built to be platform-agnostic, with support for hardware like Oculus Rift and Microsoft HoloLens, using SDKs that ensure flexibility and broad usability. The use case is using Metaverse as immersive technology. Hand gestures that facilitate intuitive manipulation of virtual tools and equipment in industrial processes. Voice commands enabling hands-free operation of complex machinery and systems, improving workflow efficiency. Eye tracking, which enhances safety and efficiency by enabling gaze-based control and monitoring in critical industrial tasks. Motion tracking, which accurately replicates worker movements for virtual training and ergonomic analysis in industrial settings. Brain-Computer Interfaces (BCIs) as an emerging interface for enabling control of machinery and systems in high-stakes environments with minimal physical input. VR controllers used to offer precise control for training simulations and interaction with virtual replicas of industrial setups. 3D elements used to enable detailed visualization of industrial assets, processes, and environments for enhanced design and decision-making. Interactive navigation systems that provide intuitive pathways to explore virtual factories and industrial layouts, improving user experience and efficiency. Real-time overlays, which deliver critical operational data and instructions overlaid on physical or virtual equipment for precise and timely actions. Holographic displays used to create immersive 3D projections of industrial systems, allowing collaborative planning and problem-solving in virtual spaces. Haptic feedback that provides realistic tactile feedback for remote operation and precision tasks in simulated environments. AR displays, which delivers real-time operational data and stepby-step instructions directly within the user's field of view for maintenance and assembly.
© AIOTI. All rights reserved. 34 5.1.5 Technology layers requirements Seamless interfaces integrating AR/VR tools and IoT-enabled devices for real-time interaction. Motion, depth, and proximity sensors, along with environmental actuators, for precise data collection and interaction. Low-latency processing at the edge with high-performance CPUs/GPUs to support real-time immersive experiences. High-speed 5G/6G networks and intelligent connectivity to ensure seamless data transmission and real-time updates. Support for 3D and spatial-temporal data with scalable local/cloud storage for lifecycle management and quick retrieval. Integration of AI algorithms for predictive analytics, computer vision, and natural language processing to enhance workflows. Support for platforms like Oculus Rift and Microsoft HoloLens with SDKs ensuring usability and flexibility. 5.1.6 Horizontal issues and characteristics Trustworthiness is a foundational requirement, demanding robust security, strict user data privacy, and full compliance with industry regulations to maintain operational integrity. The system's ethical framework is designed to prioritize user privacy and data security, safeguard against bias, and ensure it does not discriminate, guaranteeing equitable operation. The system is designed for high interactivity, with real-time responsiveness to user actions creating a dynamic and immersive experience that seamlessly blends physical and digital realities. The system perceives a wide range of human inputs, including gestures, voice commands, biometric data, and spatial positioning, to enable its rich interactivity. A commitment to standardisation and interoperability is essential to ensure that different systems and platforms within the metaverse can communicate and work together, allowing for a truly integrated and collaborative environment. This fosters a competitive landscape where different technologies can connect and enhance the overall ecosystem. Robust security, user data privacy, and compliance with industry regulations to maintain operational integrity. Ensures ethical use by prioritizing user privacy, data security, and compliance with regulatory standards. It safeguards against bias and does not discriminate against individuals or groups, ensuring equitable operation. The inputs the system perceives from humans are gestures, voice commands, biometric data, and spatial positioning. The interactivity of the system is materialised through real-time responsiveness to user actions, providing a dynamic and immersive experience. The immersiveness of the system is represented by seamless blending of physical and digital realities, fostering a sense of presence in the industrial metaverse. The use case promotes sustainability by enabling virtual prototyping, reducing physical waste, optimizing resource usage through AI-driven analytics, and supporting remote collaboration to minimize travel-related emissions. The use case complies with industry standards to ensure system compatibility and interoperability across various platforms.
© AIOTI. All rights reserved. 35 5.2 Robotic Welding 5.2.1 Scenario WELD-E targets the manufacturing sector and develops a highly interactive, responsive and immersive, bidirectional, XR-based, Human-Machine-Interaction application facilitating realtime cooperation between a welding expert and a specialised robot that performs welding operations. Welding experts shall be able to provide both voice and visual support to the robotic welder enabling it to execute precise welding tasks following experts’ guidelines and feedback. More specifically, the application uses a combination of AI models such as Automatic Speech Recognition, Neural Machine Translation, Visual Language models and a Conversation Agent (components developed and enhanced from the VOXREALITY project). After conducting each step of the welding process, the robotic welder sends real-time feedback to the welding expert, enabling comprehensive, end-to-end support. Figure 5-2 – Immersive Welding 5.2.2 Users and stakeholders The primary user of the WELD-E system is the welding expert, who leverages their specialized knowledge to guide the robotic arm remotely. This allows a single expert to oversee multiple operations or to provide guidance in environments that are hazardous or difficult to access. Other key stakeholders include the manufacturing companies that own and operate the robotic welders, as their operational efficiency and quality control are directly improved. The development team, building upon the foundation of the VOXREALITY project, is also a stakeholder, as are the providers of the hardware components such as the robotic arms and the mixed reality headsets.
© AIOTI. All rights reserved. 36 5.2.3 Implementation in a virtual world and added values The implementation of WELD-E is grounded in Mixed Reality (MR) technology, experienced through a HoloLens headset. As Industry 5.0 technologies transform manufacturing, XR applications are becoming essential for teleoperation of intricate machinery. WELD-E capitalizes on these advancements to create a system where complex human-machine interactions feel intuitive and seamless. The system fuses virtual and physical data to generate 3D holograms, creating an accurate representation of the welding environment that is superimposed onto the user's real-world view. The primary added value of WELD-E is its ability to efficiently monitor and guide robotic welding tasks in real-time by combining human expertise with advanced AI in an immersive, multi-modal XR environment. This enhances safety, improves precision, and allows for the democratization of expert knowledge across the factory floor. 5.2.4 Required immersive technologies functionalities The system is built on a foundation of Mixed Reality, with user interaction facilitated through a combination of voice commands, hand tracking, and controller inputs. The user interface incorporates both 2D elements for displaying data and menus, and 3D elements for navigating the spatial environment and interacting with the digital twin of the robot. Feedback is provided to the user through both audio cues and visual overlays within the MR environment. A significant portion of the system's functionality is driven by a sophisticated AI pipeline. WELD-E offers an end-to-end, model-based, and voice-driven system that utilises pre-existing models from the VOXReality project, including an Automatic Speech Recognition model based on OpenAI's Whisper and a context-aware Neural Machine Translation model. WELD-E enhances this foundation with two new components: an open-source Coqui-TTS model for natural, multilingual voice synthesis, and the Welding Large Language Model (WeLLM). WeLLM is a finetuned LLM that improves the accuracy of translated commands and corrects potential errors, ensuring instructions are rational and comprehensible. Furthermore, WELD-E customizes the VOXReality Vision-Language model by introducing an auto-labelling and knowledge distillation mechanism. This process uses foundation models to refine bounding boxes into precise segmentation masks, which are then used to finetune a YOLO-NAS model. This allows the system to recognize a broader range of objects with greater accuracy. The overall goal is to extend the capabilities of the VOXReality ecosystem for realworld industrial applications by enhancing technical reasoning. The platform architecture integrates these AI models within the XR environment. It consists of ROS2 (Robot Operating System) for robotic communication, which is connected with a Unity3D application. The XR environment is deployed on a Microsoft HoloLens. A central orchestrator, Welde-Connect, manages the data flow between the AI models, ROS2 nodes, and the XR application via REST API interfaces, ensuring synchronized and efficient operations. The type of immersive technology used is MR (Hololens). The user interaction modes are based on VR Controllers, Hand Tracking, Voice Commands. The user interface design employs 2D Elements, 3D Elements (Navigation). The user feedback mechanisms comprise of audio and Visual feedback.
© AIOTI. All rights reserved. 37 5.2.5 Technology layers requirements The AI technologies used are NLP, ML models, AI integration points, voice interaction, visual recognition, training data requirements, and inference capabilities. WELD-E offers an end-to-end model-based and voice-driven pipeline to support welding operations performed by robotic arms. To accomplish this, WELD-E utilises pre-existing VOXReality models in a black-box manner —including Automatic Speech Recognition (ASR), Neural Machine Translation (NMT) — but also provides two new architectural components: an open-source Text-to-Speech (TTS) model and the Welding Large Language Model (WeLLM). The TTS component adopts Coqui-TTS, an open-source solution known for its flexibility and highquality output, enabling natural and multilingual voice synthesis. The Welding Large Language Model (WeLLM) consists of a finetuned LLM model capable of improving the accuracy of translated welding commands but also correcting potential errors (misspellings, irrational phrases, hallucinations produced from incorrect or noisy translations, etc.). This ensures that all technical instructions are rational and comprehensible. By incorporating TTS and WeLLM, WELD-E enhances the overall user experience and the operational efficiency of the welding tasks. The Automatic Speech Recognition (ASR) model, reused from the VOXReality project, leverages OpenAI's Whisper architecture, finetuned with Adapter modules to improve performance in lowresource languages such as Greek. Meanwhile, the VOXReality Neural Machine Translation (NMT) model focuses on context-aware, robust, and simultaneous translation, utilising techniques such as multi-encoder architectures and data augmentation. WELD-E customises the VOXReality Vision-Language (VL) model by introducing an auto-labelling and knowledge distillation mechanism to improve object detection capabilities. This process uses foundation models for refining bounding boxes into precise segmentation masks. These high-quality masks are then used to finetune a YOLO-NAS model through knowledge distillation techniques, thus allowing it to recognise a broader range of objects with enhanced accuracy and efficiency. WELD-E aims to extend the scope and capabilities of the VOXReality model ecosystem in real-world industrial applications by enhancing their technical reasoning and capabilities. WELD-E is designed to integrate advanced AI models within an XR environment to improve welding operations. The platform consists of ROS2 (Robot Operating System) connected with Unity3D, while the XR environment employs Mixed Reality (MR) using Microsoft HoloLens. The WELD-E model ecosystem includes a central orchestrator (Welde-Connect), while interactions with the XR environment are conducted through REST API (OpenAPI) interfaces. Microsoft HoloLens provides the necessary MR capabilities for operators to interact with virtual elements overlaid in the physical world. Unity3D serves as the development engine that creates the interactive XR application, offering powerful tools for 3D content creation. ROS2 acts as the middleware for robotic communication, handling real-time data exchange and ensuring seamless coordination between hardware and software components. The central orchestrator manages the flow of data between the AI models, ROS2 nodes, and the XR application, ensuring synchronised operations and efficient processing. This architecture ensures interoperability with other platforms and devices, providing flexibility and robustness in real-world industrial applications. 5.2.6 Horizontal issues and characteristics The system is designed to perceive a variety of human inputs to enable effective and intuitive control. These include spoken commands in multiple languages, hand gestures to define actions like a welding trajectory, and touch inputs on holographic interfaces for basic commands. The
© AIOTI. All rights reserved. 38 system's interactivity is a core feature, providing rich feedback to the operator. This includes a visual MR environment acting as a digital twin of the welding area, offering real-time insights and visual alerts. It also delivers audio feedback with natural speech to provide status updates and safety warnings. The immersiveness of the system is central to its design. By using MR to superimpose 3D holograms and data onto the physical world, it allows for a seamless fusion of virtual information and realworld context. This high level of immersion facilitates the intuitive interaction, visualization, and manipulation of the digital twin, which is critical for the precise guidance of the welding robot. The system ensures that all interactions are natural, enhancing situational awareness and decision-making for the expert operator. Trustworthiness is embedded through the AI's ability to rationalize commands and the system's robust feedback loops, ensuring safety and reliability in an industrial setting. The XR Solution perceives various human inputs to enable effective interaction and control. These include spoken commands in multiple languages, which allow users to issue instructions naturally, and visual cues, such as gestures or contextual signals within the workspace. The system perceives several types of inputs from humans to enable intuitive interaction and control: • Voice Inputs: Spoken commands or verbal instructions in various languages, that the system processes to execute tasks or provide feedback. • Gestures: Hand movements or other physical gestures recognised by the system to trigger actions, such as the welding trajectory. • Touch Inputs: Interaction with virtual buttons or sliders through touch gestures on holographic interfaces to initiate basic actions such “Start Welding”, “Cancel Welding” and so on. The system also offers several feedback channels to the operators: • Visual Interaction: A MR (Mixed Reality) environment acting as the Digital Twin representation of the Welding Area. This visualisation provides operators with real-time insights into the welding process, including spatial layouts, tool positioning, and progress updates, enabling enhanced situational awareness and precise decision-making. Additionally, it incorporates visual alarms and alerts, such as colour-coded warnings or flashing indicators, to notify operators of potential safety hazards, system malfunctions, or deviations in the welding process. • Audio feedback: Delivers natural, human-like speech to provide operators with updates on the welding cycle and issues alerts when safety regulations are not followed. Virtual Reality (VR), Mixed Reality (MR), and Augmented Reality (AR) are example XR technologies that are essential in the development of teleoperation mechanisms for intricate machinery and processes. As manufacturing in Industry 5.0 has been transformed by XR technologies, it progressively depends more on complex and intelligent human-machine interactions, especially in specialised industrial fields such as welding. Accurate representations of intended surroundings are now possible thanks to recent developments in XR, which in turn simplifies the interaction, visualisation and manipulation of 3D objects superimposed to the real world. Interactions with these immersive representations are facilitated by XR systems, which offer 3D holograms by fusing virtual and physical data. The primary objective of WELD-E is to efficiently monitor and guide robotic welding tasks in real time by combining human expertise with advanced AI in immersive multi-modal XR environments.
© AIOTI. All rights reserved. 39 5.3 Assistance for Equipment Servicing and Maintenance 5.3.1 Scenario This application is a proof of concept developed by the Aalto Factory of the Future at Aalto University. It addresses a common challenge in complex industrial environments: the efficient servicing and maintenance of equipment. The use case is focussing on providing real-time assistance to technicians and workers on the factory floor. Figure 5-3 – Assistance for Servicing Figure 5-4 – Immersive Servicing The system is designed to help a user locate a specific piece of equipment within a large and potentially cluttered facility. It then generates and displays a safe path from the user's current location to the target equipment, projecting this path directly into the user's field of view via an Augmented Reality headset. This guidance system is dynamic, updating the path in real-time as the user moves through the environment. Once the user reaches the equipment, the application provides a display overlay showing relevant information about that asset.
© AIOTI. All rights reserved. 40 5.3.2 Users and stakeholders The primary users of this application are service technicians and maintenance workers who are responsible for the upkeep and repair of industrial machinery. Their workflow is directly impacted by the efficiency and safety of locating and identifying equipment. Key stakeholders include factory floor managers, who are concerned with overall operational efficiency, workforce productivity, and safety compliance. The Aalto Factory of the Future research group at Aalto University, are also central stakeholders as the developers and innovators behind this proof of concept. 5.3.3 Implementation in a virtual world and added values The application is implemented using Augmented Reality (AR) technology, which overlays digital information onto the user's view of the physical world. This creates an extended reality experience where virtual guidance is seamlessly integrated with the real environment. The implementation does not create a fully virtual world but rather enhances the existing physical one. The added value of this system is substantial. It significantly reduces the time technicians spend searching for equipment, directly improving productivity. By calculating and displaying the safest possible route, the application enhances worker safety, helping them navigate around potential hazards on the factory floor. Furthermore, by providing immediate access to equipment-specific information on the AR display, it can reduce errors and improve the quality of maintenance tasks. 5.3.4 Required immersive technologies functionalities The functionality of the system is built upon a specific set of technologies designed for robust performance in an industrial setting. The immersive technology employed is Augmented Reality, experienced through a Microsoft HoloLens 2 Enterprise headset. User interaction is managed through gesture tracking and an AR controller, allowing for intuitive control of the application. The user interface is focused on navigation and the display of 3D elements, with user feedback provided visually. The technological foundation relies on Ultra-Wide Band (UWB) positioning modules for precise location tracking. This setup includes fixed anchors throughout the environment, a tag carried by the user, and tags attached to the equipment of interest. This allows for highly accurate, realtime spatial awareness. All computational processing is handled at the edge, utilising the onboard processing power of the HoloLens 2 and an Nvidia Jetson device. This on-edge approach ensures low latency and real-time responsiveness. For connectivity, the system operates on a 5,4GHz Wi-Fi network, providing the necessary bandwidth for stable communication. Data storage is minimal, with the application file being stored locally on the device. It is important to note that this proof of concept does not currently incorporate any AI algorithms; its logic is based on the real-time processing of positioning data. The chosen platform, Microsoft HoloLens 2, provides the necessary enterprise-grade features and SDKs for this type of industrial application.
© AIOTI. All rights reserved. 47 5.6 XR-Based Human-Robot Collaboration Along a Conveyor Picking Line: The case of Construction and Demolition Waste Sorting 5.6.1 Scenario The application aims to leverage on human-robot collaboration (HRC) improve the efficiency of current waste sorting practices. By introducing human dexterity in the loop, HRC can improve sorting efficiency, measured by speed and accuracy, appropriately combining human and robot capabilities. The XR technology provides the necessary feedback mechanism between the conveyor-cobot sorting system and the humans aiding and supervising the process. Figure 5-8 – XR-Based Human-Robot Collaboration 5.6.2 User and stakeholders Primary users include waste sorting operators who interact directly with the XR system and cobots on the conveyor line to perform sorting tasks. Stakeholders encompass waste sorting plant managers focused on efficiency gains, environmental regulators ensuring compliance with waste management standards, and technology providers supplying XR hardware and software. Additionally, workers' unions may be involved to address job impacts and safety concerns arising from HRC integration. 5.6.3 Implementation in a virtual world and added values In a virtual world, the system simulates the conveyor line environment using digital twins of humans, cobots and waste materials, allowing operators to train in risk-free scenarios and optimise workflows. Added values include reduced physical strain on humans through virtual rehearsals, improved accuracy in waste categorization via immersive previews, and scalable testing of new sorting algorithms without disrupting real operations.
© AIOTI. All rights reserved. 48 5.6.4 Required immersive technologies functionalities The type of immersive technologies used are XR HMD and projection-based AR. These systems overlay real-time sorting data and cobot intent onto HMD enhancing situational awareness. Immersive projections on the physical conveyor line support collaborative spaces for multiple users, improving team-based sorting decisions. The user interaction modes are hand tracking for system control and for direct manipulation of virtual objects on the conveyor, enables the update of the current state of the virtual objects to assure task completion. Voice commands allow hands-free control, such as directing cobots to pick specific items or pausing the line for inspection. These modes ensure seamless integration in fast-paced environments, reducing cognitive load and enhancing collaborative efficiency between humans and robots. The UI design combines 2D overlays for real-time data, like sorting statistics and alerts, with 3D holographic representations of waste items that are interactive and allow the user to update the state of an item, e.g. change its sorting classification. Intuitive icons and color-coded elements guide user actions, ensuring accessibility for diverse skill levels. Adaptive layouts adjust based on user preferences, promoting ergonomic interaction in prolonged sessions. 5.6.4.1 User Feedback Mechanisms The user feedback mechanisms include visual feedback includes augmented overlays highlighting detected objects, error alerts, progress indicators on the conveyor line to confirm actions, and cobot paths for spatial awareness. 5.6.5 Technology layers requirements The requirements for the different layers are as following: Sensing and perception layers • Motion sensors: Track operator body posture and hand movements for natural interaction. • Depth sensors: Enable accurate 3D mapping of conveyor environment and waste items. • Proximity sensors: Ensure safe distances between humans and cobots. • RGB cameras: Provide visual feeds for computer vision pipelines. • X-ray cameras: Support material composition analysis (e.g., detecting nails or metals in wood pieces). • Depth cameras: Allow real-time 3D detection and localization of irregular waste objects. • Wearables: XR HMDs, haptic gloves, and biometric monitors for operator feedback. Actuation layer • Cobots: Collaborative robotic arms with compliant force sensors for safe object manipulation. • Alternatives: Compressed air nozzles or delta robots Processing and computing layers • Edge computing nodes: Localized processing to minimize latency in visual feeds and AI inference. • High-speed computer vision pipelines: Require <50ms latency for object detection and tracking (YOLOv8 or similar models).
© AIOTI. All rights reserved. 49 • AI model hosting: Support for real-time inference, retraining, and continuous dataset updates. Communication and networking layer • Low-latency connectivity: 5G or industrial Ethernet ensuring <10ms communication delay between XR devices, cobots, and edge nodes. • Protocols: OPC UA, ROS2 middleware for interoperability and secure data exchange. • Cloud integration: For long-term data storage, digital twin synchronization, and performance analytics. XR and feedback Layer • XR devices: Provide tactile cues for object classification or error alerts. • Augmented overlays: Real-time visualization of sorting paths, cobot trajectories, and progress indicators. AI for near-real-time inference models like YOLO for object detection of waste items or dimensionality reduction of hyperspectral data for material recognition; integra. Machine learning algorithms adapt to varying waste compositions, improving accuracy over time through supervised training on labelled datasets. The system leverages ROS2 for cobot control and coordination, offering enhanced real-time capabilities, improved security, and better support for distributed systems in human-robot collaboration. For XR visualization, Unity or UnrealEngine provide robust rendering for immersive interactions in virtual environments. NVIDIA Omniverse enables advanced 3D simulation and collaborative workflows using USD for digital twins and physics-based robotics testing. 5.6.6 Horizontal issues and characteristics The use case incorporates transparent AI decision logs and fail-safe mechanisms that allow humans to override of cobot actions. Regular audits and certification against industrial standards like ISO 10218 for collaborative robots ensure reliability and safety. User training programs emphasize system predictability, reducing apprehension and fostering acceptance in HRC workflows. Ethical considerations include protecting worker privacy through GDPR-compliant data processing and ensuring equitable job distribution without displacing human roles. Human rights are upheld by designing inclusive interfaces accessible to diverse abilities and preventing overreliance on automation that could lead to skill erosion. Bias mitigation in AI models avoids disproportionate task allocation to human sorters that are performing well. The application promotes sustainability by optimizing waste sorting to increase recycling rates and reducing landfill contributions from construction debris. Energy-efficient edge computing and low-power XR devices minimize environmental footprint. Long-term benefits include resource conservation through better material recovery, aligning with circular economy principles in manufacturing. Adoption of open standards in the context of Digital Twins enhances scalability and collaboration in multi-vendor ecosystems. Standardisation following the ISO23247 reference architectures for digital twins in manufacturing and relevant component implementation from the Eclipse or FIWARE ecosystem and protocols like MQTT or OPC-UA for seamless integration between cobots, sensors, and XR systems across vendors ensures interoperability and compatibility with existing conveyor infrastructures.
© AIOTI. All rights reserved. 50 6 Automotive The automotive industry has been an early adopter of immersive technologies, leveraging them across the entire vehicle lifecycle. In the design phase, VR allows engineers and designers from around the world to collaborate in shared virtual spaces, interacting with full-scale digital prototypes of vehicles long before physical models are built. This accelerates the design process, reduces errors, and saves costs associated with physical prototyping. For manufacturing, AR provides technicians on the assembly line with interactive, step-by-step instructions overlaid directly onto their field of view, improving accuracy and reducing assembly time. These systems can also be used for quality assurance, automatically highlighting defects or deviations from the design specification. In sales and marketing, immersive showrooms allow customers to explore and customise vehicles in a highly realistic virtual environment. 6.1 Meta-Factory Model for Electrified Braking System 6.1.1 Scenario The METABRAKE application is situated within the automotive sector, specifically focusing on the collaborative innovation of advanced automotive products and their associated manufacturing processes. Led by Brembo S.p.A., the project aims to develop a fully electrified braking system that is synergistically integrated into the modern ecosystem of electric, digital, and connected cars. Figure 6-1 – METABRAKE use case. Figure 6-2 – Immersive “Meta-Factory” To achieve this, METABRAKE utilises a "meta-factory" model for the pre-industrial evaluation of the innovative new product and its assembly process. This meta-factory is a comprehensive and interactive virtual environment that represents the entire manufacturing system.
© AIOTI. All rights reserved. 51 Within this virtual space, product modifications can be introduced and analysed, allowing the team to assess the impact on factory layout, production costs, and final product quality before any physical resources are committed. 6.1.2 Users and stakeholders The primary users of the METABRAKE meta-factory are the engineers, product designers, and process planners within Brembo and its partner organisations. They will interact with the virtual environment to test hypotheses, simulate changes, and collaboratively refine both the product and the production line. The stakeholders for this project extend beyond the immediate development team. The collaborative nature of the platform is designed to integrate downstream supply chain actors and even customers directly into the innovation process. This allows for a more holistic approach to product development, where feedback from suppliers and end-users can be incorporated and evaluated early in the design cycle. 6.1.3 Implementation in a virtual world and added values The core of this application is implemented as a Metaverse, a persistent, shared virtual space where users can interact with a digital twin of the factory. This meta-factory is not just a static model but a dynamic simulation environment where processes can be run, data can be generated, and changes can be tested in real time. The added value of this approach is multifaceted. It significantly enhances both economic and environmental sustainability by allowing for extensive testing and optimization in the preindustrial phase, reducing waste and costly physical prototypes. It fosters a new paradigm of collaborative innovation by breaking down silos between the manufacturer, its supply chain, and its customers. Furthermore, by simulating processes in detail, the system will be used to evaluate and optimise the ergonomics of workstations, improving human well-being on the future production line. 6.1.4 Required immersive technologies functionalities The system's functionality is built upon a sophisticated stack of industrial simulation and virtual reality technologies. User interaction within the Metaverse is primarily through VR controllers and hand tracking, allowing for intuitive manipulation of the 3D elements that constitute the virtual factory. Feedback is provided to the user through both audio and visual cues to create a responsive experience. The underlying technology layers are designed for industrial-grade simulation. The field-level IoT devices and actuators are not physically present but are modelled using virtual commissioning solutions. A supporting IT infrastructure has been designed to simulate the entire edge layer, including devices, applications, and management. This requires significant CPU and GPU power to ensure real-time rendering and low-latency interactions during complex simulations. A lowlatency network is also crucial for providing real-time feedback from these demanding simulations. The platform manages complex data types, including 3D data, life-cycle object data, and the digital twin of the assembly line, complete with virtual sensors. All data storage and backup are handled on-premises. The technology platform is a composite of several industrial and gaming solutions, including the Siemens simulation suite (NX, PlantSimulation, PLCSimAdvanced), the Unity Real-Time Development Platform for the virtual environment, SIMATIC WinCC for the user dashboard, and Unity Relay for collaboration. Artificial Intelligence is not a primary focus of the current project phase.
© AIOTI. All rights reserved. 52 The type of immersive technology used is Metaverse. The user interaction modes are VR controllers, hand tracking with user interface based on 3D Elements and user feedback mechanisms relying on audio and visual feedback. 6.1.5 Technology layers requirements The field-level layer is modelled with virtual commissioning solutions, and a support IT infrastructure has been designed and will be implemented to simulate edge devices, edge apps, and the edge management layer. The edge computing processing uses CPU/GPU specifications to ensure real-time rendering and low-latency interactions. The high-speed networks and intelligent connectivity infrastructure requires low-latency to ensure real-time feedback. This is particularly challenging due to the complex simulations involved. The data types and storage must consider 3D data, life-cycle object data, DT of the assembly line, virtual sensors and storage and backup on-premises. The AI is not in the focus of the use case. Industrial platforms for simulation and digital representation are used (e.g., Siemens platform: NX mechatronic concept designer, PlantSimulation Standard, PLCSimAdvanced) and virtual environment platform (e.g., Unity RealTime Development Platform | 3D, 2D, VR & AR Engine), user dashboard in the metaverse (e.g., SIMATIC WinCC Unified) and collaboration platforms (e.g., Unity Relay). 6.1.6 Horizontal issues and characteristics Trustworthiness is a key consideration, and the system is designed to guarantee user privacy and compliance with all relevant sector regulations. The reliability and robustness of the platform are critical, as is the security of the sensitive product and process data it contains. From an ethical standpoint, while there are no major implications, the project actively contributes to human well-being by using the simulations to improve the ergonomics of future workstations. The system is designed to be highly interactive, enabling users to dynamically manipulate the virtual environment and simulate the impact of their changes in real time. Human inputs are captured via VR controllers and hand tracking, and users can also upload new design data to evaluate its effects. The immersiveness of the system is ensured by the detailed IT infrastructure and virtual sensors that create a complete and responsive digital replica. This approach inherently supports sustainability and requires a strong commitment to standardisation and interoperability to facilitate the seamless integration of external partners. User privacy and compliance with sector regulations should be guaranteed. The system should be reliable and robust, and the security of sensitive data must be ensured. This application does not feature significant ethical implications. However, concerning human well-being, it should be noted that the simulations in the virtual environment will also be used to evaluate and optimise the ergonomics of the working stations. Inputs through the VR controller and tracking of hand motions to interact with the virtual environment. Moreover, the user will be able to upload design modifications (e.g., 3D data) to evaluate the impact in the meta-factory. The system is highly interactive to enable dynamic manipulation and impact simulation. IT infrastructure and virtual sensors have been implemented to ensure the system's immersiveness and completeness. Meta-factories can enhance the economic and environmental sustainability of the preindustrial phase. Standard and interoperability should be guaranteed to allow integration with downstream supply chain actors and customers.
© AIOTI. All rights reserved. 53 7 Energy The energy sector covers all the stakeholders in its entire value chain such as the Energy organisations (Oil& Gas production units, Refineries, Thermal Power Plants, Nuclear Power Plants, Wind Power Farms, Solar Power Plants, Power Transmission & Distribution Systems etc), Energy Technology / Process Licensors, Engineering organisations, EPC organisations, Manufacturers and Fabricators, Construction companies, Mining Companies (for Coal) as well as several service providers covering HSE, logistics, inspection and other services [39]. The energy sector, from traditional oil and gas to renewable sources, is leveraging immersive applications to improve safety and operational efficiency. VR training modules are used to simulate high-stakes emergency procedures, such as responding to a blowout on an offshore oil rig or performing maintenance on a wind turbine, in a completely safe environment. AR and MR are utilised for asset management and maintenance on complex sites, such as power plants or substations. A technician wearing an MR headset can see real-time operational data, historical maintenance records, and schematics overlaid directly onto the physical equipment they are servicing. This improves accuracy, reduces the chance of human error, and enhances worker safety by providing critical information at the point of need. VR can be applied in the energy sector to optimise renewable energy systems, enhancing energy efficiency, and facilitating workforce training. One of the primary areas in which VR is utilised in sustainable energy is in the design and testing of renewable energy sources. Virtual environments enable the simulation of solar panels, wind farms, and energy storage systems under various environmental conditions, allowing engineers to analyse performance, identify potential inefficiencies, and optimise layouts [49]. 7.1 Immersive VR for Operation of Wind Turbines 7.1.1 Scenario The training application has the scope of providing an easy to transport, safe, and repeatable VR-based training solution for operators in the fields. It helps staff members to learn or improve procedural skills and familiarise them with the work environment and tools. For instance, process training steps on how to assemble parts of the wind turbine, how to identify areas where objects like screws are placed and how to operate a telescopic crane. Figure 7-1 – IVR training solution in Siemens-Gamesa
© AIOTI. All rights reserved. 54 Figure 7-2 – IVR training solution 7.1.2 User and stakeholders Main users are unexperienced staff members of companies that have to deploy in -the-filed some equipment (e.g., a wind turbine). 7.1.3 Implementation in a virtual world and added values The added value is to reduce the cost of training and provide an in-situ capability of better addressing construction problems thanks to the use of VR. 7.1.4 Required immersive technologies functionalities The type of immersive technology used is VR with VR controllers as user interaction modes and a user interface design built around 3D elements, gamification and multi-user interactivity. The user feedback mechanisms are haptic feedback is used in the simulator to give vibrations when trainees do something wrong, e.g., when they walk into virtual walls. No haptics when interacting with objects. 7.1.5 Technology layers requirements The HTC Vive is used as platform for the currently deployed VR training. For most parts of the training, the interactions involve two hands (using the VR controllers). The workspace is roomscale, as the trainees must kneel, walk around and turn. 7.1.6 Horizontal issues and characteristics No specific horizontal issues defined.
© AIOTI. All rights reserved. 55 8 Buildings The Architecture, Engineering, and Construction (AEC) sector uses immersive technology to revolutionise how buildings are designed, constructed, and managed. Architects utilise VR to create immersive walkthroughs of their designs, enabling clients to experience a space and provide feedback well before construction begins. This helps to align expectations and reduce costly late-stage design changes. On the construction site, AR can be used to overlay Building Information Models (BIM) onto the physical environment, allowing workers to verify that construction is proceeding according to plan and to identify clashes between different systems (e.g., plumbing and electrical) before they become problems [46]. Post-construction, digital twins of buildings, accessed via immersive interfaces, are used for facilities management, space planning, and optimising energy consumption. Construction design began with the drafting board, moved through the age of ComputerAided Design (CAD), and now utilises BIM technology, which employs unified data models and immersive technologies to support designing 3D structures in 2D space. As part of the ongoing evolution of the construction industry, immersive technologies such as AR have a significant impact on the AEC sector, visualisations, information retrieval, and interaction [47]. BIM and VR can be integrated to enhance indoor lighting design in construction by leveraging gaming engine technologies, the approach allows for the creation of 3D environments to experiment with various lighting designs, addressing the critical role of lighting in both aesthetics and functionality [48]. The authors in [48] provide an overview and several examples of the adoption of virtual and augmented reality in the architecture, engineering, construction, and facilities management. 8.1 Virtual Architectural Design 8.1.1 Scenario This immersive application allows city planners and architects to visualize and interact with digital twins of urban environments. Through this platform, users can simulate a variety of scenarios, from infrastructure changes to the implementation of green spaces, and assess their impact on urban dynamics. By integrating real-time data, users can explore how traffic patterns shift or how new developments affect pedestrian flow. The application also includes predictive analytics to evaluate environmental impacts, such as air quality improvements or potential noise pollution changes. Figure 8-1 – Virtual Architectural Design 8.1.2 User and stakeholders Using immersive technology, stakeholders can collaborate in a virtual space, bringing together architects, engineers, and city officials to make informed decisions.
© AIOTI. All rights reserved. 56 8.1.3 Implementation in a virtual world and added values N/A. 8.1.4 Required immersive technologies functionalities The type of immersive technologies used are Metaverse and XR (AR/VR) environments with user interaction modes based on hand tracking, voice commands, VR controllers, eye tracking. The user interface design combines 3D elements for realistic city models and interactive design layers with 2D elements for HUDs and data display, providing intuitive navigation through virtual urban spaces. The user feedback mechanisms Incorporate haptic feedback for tactile interactions, Audio feedback for guided exploration and alerts, and visual feedback for real-time changes and notifications. 8.1.5 Technology layers requirements Integrates with IoT for real-time data collection and environmental monitoring, enhancing HMI through seamless interactions with virtual representations of physical spaces. Utilises motion sensors, depth sensors, cameras for capturing environmental data, and haptic devices for realistic feedback. Environmental actuators simulate elements like temperature or wind, enhancing realism. Requires high-performance CPU/GPU capabilities for real-time rendering and low-latency interactions. Combines distributed computing paradigms to manage computational loads between edge and cloud-based resources effectively. Employs robust communication protocols to ensure low-latency, real-time data transfer and processing, essential for synchronising virtual and real-world data streams. Manages large volumes of audio/video, 3D data types, and spatial-temporal data. Utilises hybrid local/cloud storage solutions for capacity scalability, with backup/recovery and data life cycle management protocols ensuring data integrity. AI leverages computer vision for spatial analysis, natural language processing for voice commands, and machine learning for predictive analytics. AI algorithms enhance user interactions and visualise complex data, informed by robust training datasets. The platforms used are compatible with Oculus Rift, HTC Vive, Microsoft HoloLens, supporting cross-platform interoperability through comprehensive SDKs and targeted OS optimisations. 8.1.6 Horizontal issues and characteristics Trustworthiness ensures user data anonymisation and compliance with urban planning standards and regulations. The system is designed for reliability, fault tolerance, and security, safeguarding sensitive developmental data and user interactions. The use case prioritises privacy, ensuring that simulated scenarios do not infringe on individual rights. Emphasizes ethical AI deployment by eliminating biases in planning simulations and promoting inclusive urban designs. The use case facilitates sustainable urban planning by allowing users to visualize and evaluate the environmental impact of various design choices, promoting eco-friendly urban development practices. The use case adheres to industry standards for data formats and communication protocols to ensure interoperability with other urban planning tools, facilitating easier integration into broader city development initiatives. This application leverages immersive technologies to transform traditional urban planning into an interactive, data-driven process, enabling more informed, inclusive, and sustainable city development.
© AIOTI. All rights reserved. 63 The platforms used for XR visualization are Unity or Unreal Engine provide robust rendering for immersive interactions in virtual environments. VR HMD devices will be used by the VR client to control the VR cybersecurity assessment. An AR tablet(iPad) could be used for the client to control and track any AR learning and assessment elements of the application. Any supporting e-learning content could be linked to a WebView editor and a Moodle API interface to any training course content. Supporting tools such as shapes XR, Revit, twin Motion, blender, Unity would be leveraged to complete the digital twin creation of the hospital environment and simulated experiences. 10.1.6 Horizontal issues and characteristics Such training programs provide a healthcare system that promotes patient safety as core to its organisation, is transparent and is proactive in its cybersecurity measures. Cybersecurity threats in a hospital environment raise major ethical concerns and challenges as they can impact the patient’s safety and privacy. There is an obligation of healthcare workers and professionals to protect patients’ confidentiality and not expose sensitive data. Such immersive training scenarios with such defined cybersecurity training scenarios help train such professionals and build public trust in healthcare systems and their duty of preparedness. The application promotes sustainability by educating the individual to be more knowledgeable about potential cybersecurity threats and how to actively avoid. This in turn ensures the hospital operates sustainably, maintain high quality care without being hindered by cyber threats. The use case uses consistent protocols and best practice. Streamlines risk assessment and aligns with regulatory requirements and standards. With regard interoperability its builds trust between systems and providers and allows for more secure data exchange.
© AIOTI. All rights reserved. 64 11 Agriculture Immersive technologies are bringing precision agriculture to a new level. Farmers in the field can use AR applications to visualise data from drones and sensors directly over their crops. This can include information on soil moisture, nutrient levels, or the presence of pests and diseases, allowing for highly targeted interventions. This data-driven approach, often connected to a digital twin of the farm, helps to optimise resource use, such as water and fertiliser, and increase crop yields sustainably [50]. VR is also being used for training agricultural workers on the operation of sophisticated farming equipment and for simulating the potential impact of different farming strategies or climate change scenarios. Using DT as a central means for farming and control can offer significant advantages as DTs remove limitations concerning place, time, and human observation, and agriculture would no longer require physical proximity, which enables remote and automated execution, monitoring, control, and coordination of farm operations, allowing for the decoupling of physical flows from information aspects of farm processes. DT can also be enriched with information that cannot be observed by the human senses (e.g. sensor and satellite data) or data that other information owners provide. DT can add intelligence using advanced analytics that not just represent actual states, but can also analyse historical states and simulate future behaviour, which enables farmers and other stakeholders to act immediately in case of (expected) deviations and bring smart farming to new levels of farming productivity and sustainability [50]. 11.1 XR-based Agricultural Digital Twin 11.1.1 Scenario The use case proposes a platform hosting digital twins of agricultural farms, which utilises extended reality for both data collection and visualisation. The application domain is an XRbased Digital Twin of a farm, containing multiple data layers fed by an Augmented Reality application and made visible through a Virtual Reality interface. Figure 11-1 - Immersive XR-based Agricultural Digital Twin Concept
© AIOTI. All rights reserved. 65 Figure 11-2 - Immersive XR-based Agricultural Digital Twin By combining the real-world farm's real-time IoT data with the ability to review, analyse, and audit farm processes, we enable the farmer of the future to optimise their operations. This allows users to monitor crop growth, ensure animal welfare, check soil health, and maintain compliance with health and safety requirements on the farm. Augmented Reality is used to capture live data and map farm locations, while a Virtual Reality headset is harnessed to visualise the various data layers of the farm. 11.1.2 Users and stakeholders The primary users of this application are farmers and farm managers who seek to modernise and optimise their agricultural processes. They will use the system for day-to-day monitoring, longterm planning, and ensuring operational safety and efficiency. Secondary stakeholders include agricultural consultants, compliance auditors, and technology providers. Consultants can use the twin to provide data-driven advice, while auditors can remotely verify adherence to safety and environmental regulations. Technology providers, such as digital twin platform hosts and sensor manufacturers, are also key stakeholders in the development and maintenance of the ecosystem. 11.1.3 Implementation in a virtual world and added values The implementation of this use case revolves around extended reality (XR) technology, which encompasses both Augmented Reality for data collection and Virtual Reality for data visualisation and interaction. The AR component, likely on a mobile device, captures real-world data and scans environments, while the VR component provides an immersive, scaled representation of the entire farm. The added value is significant, providing a comprehensive tool for modern farm management. It moves beyond simple data collection to offer actionable insights through intuitive visualisation. This enhances decision-making for crop rotation, resource allocation, and animal herd management. It also provides a robust framework for ensuring and documenting compliance with health, safety, and environmental standards.
© AIOTI. All rights reserved. 66 11.1.4 Required immersive technologies functionalities The system's functionality is built upon several layers of advanced technology. The core immersive technology is XR, leveraging the unique strengths of both AR and VR. User interaction is designed to be natural and intuitive, primarily using hand tracking to interact with the user interface. This is augmented by eye tracking and voice commands to highlight and engage with the various layers of the digital twin, allowing for precise interaction with specific processes and virtual farm machinery. The user interface design thoughtfully integrates both 2D and 3D elements. The AR application utilises a 2D UI for efficient data display and capture in the field. Within the VR headset, a fully 3D UI allows users to directly manipulate the farm model with their hands. Navigation through the AR experience is simply the user's movement across the physical farm, whereas the captured twin in VR can be navigated and scaled intuitively. To enhance the sense of presence and control, the system relies on precise audio and visual feedback as a response to user interactions, compensating for the lack of physical haptic devices. The technology layers required to support this are extensive. For Human-Machine Interaction and IoT, the twin integrates data from various sensors. This includes low-energy proximity sensors and advanced mmWave radar for analysing footfall, occupancy, leak detection, and even human or animal gait recognition for welfare and safety. Hyperspectral cameras mounted throughout the farm will monitor soil health and crop growth. To process this vast amount of data, the system will utilise a distributed computing paradigm, relying on a digital twin solution provider like Bentley Systems. This combines high-performance cloud computing with edge computing for real-time processing. To ensure seamless data transfer across the large landmass of a farm, a high-speed 5G network is proposed. This intelligent connectivity is crucial for providing the low-latency, real-time data feeds necessary for safety and operational monitoring. Data management involves handling complex types, including 3D spatial data, which will be stored in the highly interoperable GLTF format. Artificial Intelligence is essential, with algorithms like Gaussian splatting used to create photorealistic replications of interior farm environments from scans. The target platforms are chosen for their robust features and ease of development; the Meta Quest 3 for the VR experience and an iPhone with LiDAR capabilities using ARKit for the AR data capture component. The use case utilises as immersive technologies XR which encompasses AR for data collection and capture and VR for data visualisation. The user interaction modes use of hand tracking to interact with the UI whilst utilising eye tracking and voice commands to highlight and interact with the varying layers of the digital twin and precisely pinpoint specific processes and farm machinery. The user interface design uses of both 2DUI in AR applications and 3DUI in VR headsets to convey the information required efficiently. Navigation throughout AR experience is the landmass of the twinned farm whilst the captured twin in VR can be modified manipulating the model with the user’s hands using the 3DUI. The user feedback mechanisms considered are hand tracking in the VR visualisation to create a more naturalistic input mechanism. As such we rely on precisely designed interaction methods to provide audio & visual feedback to the user as a means of haptic feedback and enhance sensation within the application.
© AIOTI. All rights reserved. 67 11.1.5 Technology layers requirements As a digital twin of an agricultural farm, the use case separates the twin into representative layers to show case different aspects of the twinned farm. To analyse footfall and occupancy the use case integrates both low energy external solutions for proximity sensing as well as newer technologies such as mmWave radar for leak detection, storage tank inspection and human/animal gait recognition. The use case plans to include soil monitoring and crop growth sensors using hyper spectral image analysis through mounted cameras through the farm area. To deploy such a system, a digital twin solution provider such as Bentley, who provide the resources required to deploy twins of this scale will be used. To facilitate data transfer throughout the agricultural environment is proposed to use of 5G networking. This allows connectivity across the larger landmass of the twinned area. Since the visualisation of the twin for health and safety would require real time data low latency is a high priority. The model gathered from the scanning using AR would be stored as GLTF as it a highly interoperable exchange format. The use of Ai algorithms across the digital twin are essential to its data capture and visualisation needs. Gaussian splatting would be used to create replication of interior environments. For ease of use and interoperability of components the Meta Quest 3 provides a stable software environment and easy to integration SDKs to develop the VR platform. For the AR component of this application the use of an iPhone with LiDAR capabilities and ARKit SDK would serve as a powerful AR visualisation device. In addition is plan the development of the twin using Bentley’s digital twin platforms. 11.1.6 Horizontal issues and characteristics Trustworthiness is paramount, especially as the platform gathers information on human activity. Strict adherence to GDPR will be required, with user data anonymised and associated only with a unique ID. The system will also account for ISO standards regarding AR/VR usage safety and data security. From an ethical perspective, data anonymisation is critical to preserving the privacy of any human participant, upholding their right to data removal. While sensor data is used to ensure farm safety, it is managed in a way that protects the privacy of individuals. The system is designed to be highly immersive and interactive, allowing users to explore the farm's data layers at a real-world scale. For sustainability, the system is designed to be extensible, allowing for the easy inclusion of additional sensors or the updating of scanned environments through the AR application. Finally, for standardisation and interoperability, the platform will adhere to the ISO/IEC 30173:2023 standard for digital twins, ensuring it can integrate with other systems and technologies in the future. As the platform would gather information on human activity around the farm strict adherence GDPR would be required. Users would be associated only via a UUID controlled by an IDM. Adherence to ISO standards on both use of AR/VR usage safety& data security would need to be accounted for.
© AIOTI. All rights reserved. 68 The inputs the system perceives from humans requires data anonymisation to preserve the privacy of any human participant as well as upholding their right to data removal. The gait analysis/presence would ensure farm gathered from the sensors would enable farm safety whilst ensuring the privacy of the humans present on the farm. The user of the system interacts through XR. Both in AR by capturing data and recording locations and through VR as twin visualisation mechanism. The platform is highly immersive allowing the user to navigate and explore the data layers of the digital twin in real world scale. The system is extensible via the inclusion of additional sensors or updating of additional scanned components through the AR application. The platform would seek to adhere to ISO/IEC 30173:2023 standard for digital twins. 11.2 Immersive Agricultural Digital Twin 11.2.1 Scenario The application proposes a XR-based DT of an agricultural farm containing multiple data layers fed by AR applications and visible through a VR interface. XR is used for both data collection and visualisation and combines the real world of farm real-time IoT devices with the ability to review, analyse and audit farm processes, thus optimizing the management of all processes and aspects of a farm. For instance, to monitor crop growth, ensure animal welfare, check soil health and allow for compliance with health & safety requirements on the farm. AR captures live data and map farm locations, whilst the VR headset is harnessed to visualise farm data layers. Figure 11-3 – XR based Agricultural Digital Twin
© AIOTI. All rights reserved. 69 11.2.2 User and stakeholders The main users of the applications are farmers that want to have a digitized and more efficient management of their property. 11.2.3 Implementation in a virtual world and added values No specific implementation in a virtual world. 11.2.4 Required immersive technologies functionalities The application utilises immersive technologies like XR, which encompasses AR for data collection and capture and VR for data visualisation. The user interaction modes comprise of hand tracking is used to interact with the UI whilst utilising eye tracking and voice commands to highlight and interact with the varying layers of the DT and precisely pinpoint specific processes and farm machinery. The application uses 2DUI in AR and 3DUI in VR headsets to efficiently convey the required information. Navigation throughout AR experience is the landmass of the twinned farm whilst the captured twin in VR can be modified manipulating the model with the user’s hands using the 3DUI. The user feedback mechanisms used are hand tracking in the VR visualisation process is implemented, so to create a more naturalistic input mechanism. Precisely designed interaction methods can provide audio and visual feedback to the user as a means of haptic feedback and enhance sensation within the application. 11.2.5 Technology layers requirements The application is highly immersive allowing users to navigate and explore the data layers of the DT. As the application gathers information on human activity around the farm, strict adherence to GDPR is required. Regarding the data storage, the model gathered by using AR is stored in the highly interoperable GL Transmission Format (GLTF), designed for the efficient transmission and loading of 3D models. To facilitate data transfer throughout the agricultural environment 5G is used, thanks to its capability of ensuring real time data exchanged at low latency. For ease of use and interoperability of components the Meta Quest 3 provides a stable SW environment and easy to integration SDKs to develop the VR platform. For the AR component of this application the use of an iPhone with LiDAR capabilities and ARKit SDK serves as a powerful AR visualisation device. Development of the twin is done using Bentley’s DT platforms. 11.2.6 Horizontal issues and characteristics Data anonymisation would be required to preserve the privacy of any human participant as well as upholding their right to data removal. The gait analysis/presence would ensure farm gathered from the sensors would enable farm safety whilst ensuring the privacy of the humans present on the farm. Adherence to the ISO/IEC 30173:2023 standard for DT is consider for this use case.
© AIOTI. All rights reserved. 70 11.3 Smart Agriculture: Precision Farming 11.3.1 Scenario The application employs eXtended Reality (XR) and Digital Twin (DT) technologies to enhance sustainable agricultural practices. By creating a digital replica of a farm ecosystem, it allows real-time monitoring, analysis, and decision-making. Farmers and agricultural experts can immerse themselves in a 3D representation of their fields, visualize crop health, simulate irrigation strategies, and predict yields. The virtual world component enables collaboration across stakeholders (farmers, scientists, and policymakers) in a shared, virtual environment to co-create solutions, test interventions, and train in advanced farming techniques. Imagine you are a farmer and would like to understand when it would be best to irrigate your fields and using how much water. Using a DT of your field you would be able to run scenario comparison and see the impact of low-, medium-, and high-irrigation on your crop moving forward in time. Figure 11-4 – XR–DT Integration for Immersive Monitoring and Irrigation Scenario Analysis in Precision Agriculture. 11.3.2 User and stakeholders The primary users of this application are farmers, agricultural consultants, goods and appliances delivery firms in the agricultural world. Stakeholders are the customer of farmers, farmers themselves, and more in general all the ecosystem around farming and production of goods.
© AIOTI. All rights reserved. 71 11.3.3 Implementation in a virtual world and added values The application is implemented within an XR world where different aspects of farming are taken care of and seamlessly arranged into a unified 3D layout. This virtual environment allows farmers and other stakeholders to interact with virtual replica of the farming domain (fields, cattle, instrument for irrigation, machine to spread fertilizers, etc), fostering real-time assessment of the status of crops and fields as well as allowing for scenario playing (change water, fertilizer levels, etc). Among the added values one can list reduction in cost to run a factory, energy-efficient use of resources, foster small farmers, scenario creation of different technologies/choices. 11.3.4 Required immersive technologies functionalities The immersive technologies used are AR for on-field guidance and data overlay, VR for immersive training and strategy simulation and virtual worlds for stakeholder collaboration and education. The user interaction modes are using AR for smartphone/tablet interface, AR glasses, voice commands, and hand gestures, VR for VR headsets, hand tracking, and voice commands and virtual worlds for PC and VR headsets for accessing the shared virtual platform. The user interface design is based on 2D dashboards for performance metrics that can be integrated with interactive 3D models of fields and crop stages. In addition, navigation tools for exploring the DT will be made available. The user feedback mechanisms are visual feedback (real-time analytics and alerts), together with audio feedback for actionable insights and haptic feedback in training simulations for equipment handling. 11.3.5 Technology layers requirements HMI are used for IoT devices that offer an intuitive and multilingual interfaces for farmers. Drones, weather stations, smart irrigation systems and sensors/actuators that can provide a set of key measurements (e.g., soil moisture). Real-time data processing is applied for quick insights through the implementation of edge computing. Low-latency connectivity is used for remote field areas using satellite or 5G connection to guarantee high-speed networks and an intelligent connectivity infrastructure. Data types and storage consider spatial and temporal data, crop health images, and IoT sensor data stored locally and on the cloud. The AI developments include the development of predictive models for crop health and yield. Computer vision algorithms for pest and disease identification. Natural Language Programming (NLP) for conversational AI in native languages. The platforms used are mobile AR/VR platforms (Oculus, HTC Vive, etc.), metaverse-compatible ecosystems (Decentraland, Meta Horizon). 11.3.6 Horizontal issues and characteristics Data anonymization and compliance with agricultural regulations are used and added to a secure data storage and transfer protocols. The use case ensures accessibility for smallholder farmers and provides transparency in AI decision-making processes. The sustainability is considered by the use case and reflected in reduction in resource wastage (e.g., water, fertilizers) and deployment of energy-efficient computing solutions. The standardisation issues address the integration with global agricultural databases and IoT standards.
© AIOTI. All rights reserved. 72 12 Tourism The tourism industry is leveraging immersive tech as both a marketing tool and a way to enhance the visitor experience. VR "try-before-you-buy" experiences enable potential tourists to experience a destination, from exploring a hotel room to visiting a famous landmark, before booking a trip. Once at a destination, AR applications can act as interactive tour guides, providing information about points of interest, translating signs in real-time, and offering gamified scavenger hunts to make exploration more engaging. In the context of the metaverse, there is potential for a new form of virtual tourism, where users can visit and socialise in realistic digital replicas of destinations, creating new revenue streams for the industry [44]. The future of industrial immersive applications will be defined by greater intelligence, deeper integration, and seamless interoperability. The next generation of these systems will be powered by increasingly sophisticated AI, enabling environments that are not just interactive but are truly responsive and adaptive to user behaviour and context. The concept of the "Immersive Triplet," which links the physical asset, its real-time digital twin, and an AI-driven predictive model, will become more prevalent, allowing for proactive maintenance and operational optimisation. The rise of the IoT of Senses (IoTS) will enable richer multi-sensory experiences, incorporating haptics, and even smell and taste, to create a more profound sense of presence. A primary focus of future research is on developing open, interoperable standards necessary for a true industrial metaverse, enabling seamless transfer of data, assets, and avatars between platforms. The development of the next-generation spatial web (Web 4.0) will provide the decentralised infrastructure needed for this vision. As these technologies become increasingly integrated into critical industrial processes, ensuring their trustworthiness, security, and ethical use will be of paramount importance. The industrial real-digital-virtual continuum is still in its early stages, but its potential to enhance human capability, optimise complex systems, and drive the next wave of digital transformation is undeniable. 12.1 Integration of Digital Twins in the Natural Reserve Laukvikøyene 12.1.1 Scenario The Laukvikøyene Nature Reserve is in Vågan municipality, Nordland County, Norway. The reserve covers an area of 10,888 decares, of which 7,372 decares are marine areas. It was established to preserve this valuable coastal area and its associated plant and animal life. The reserve is characterized by shallow marine areas, islands, marshlands, and several freshwater ponds. Numerous bays and coves cut into the landscape, creating a diverse and dynamic environment. Laukvikøyene is renowned for its rich biodiversity, particularly as a breeding ground for wetland birds and a wintering area for seabirds. In addition to birdlife, the reserve is home to the European Otter (Lutra lutra), a species of conservation concern. The botanical value of the area is also significant, especially in the coastal zone, which hosts several rare and endangered plant species. The variety of habitats, from marine to freshwater ecosystems, contributes to the ecological importance of Laukvikøyene. The nature reserve serves as a critical reference point for monitoring environmental changes, providing a robust foundation for sustainable management and research. It stands out as a vital location for both conservation and biodiversity studies, highlighting the interplay between land and aquatic ecosystems. The application creates a DT for Laukvikøyene that integrates data from a diverse sensor network. This digital representation provides real-time visualization of the current state of the ecosystem and supports decision-making, research, and public engagement.
© AIOTI. All rights reserved. 79 The user interface design considers that the IVR scenarios included a realistic representation of the machines and environment, and visual cues to guide the user to perform the training tasks. • Haptics – No haptic feedback was reported. • User Adaptation – No adaptation for specific users was identified There is no systematic validation of the IVR training. An informal contest-type assessment between IVR trained and non-IVR trained employees provided equal performance metrics, depicting the validity of IVR as a training tool. No haptic feedback was reported and no adaptation for specific users was identified. There is no systematic validation of the IVR training. An informal contest-type assessment between IVR trained and non-IVR trained employees provided equal performance metrics, depicting the validity of IVR as a training tool. 13.2.5 Technology layers requirements Interactions are through the Vive controllers using both their tracked positions and the controller buttons. The trainees get audio, visual, and text cues guiding them to complete the training steps. 13.2.6 Horizontal issues and characteristics The system is designed to be highly interactive and immersive. The realistic graphical representation of the machinery and the factory environment, combined with clear visual cues, creates a compelling and engaging training experience for the user. The IVR training is in active use within the company. From a human-centric perspective, the system perceives user input through their physical movements and interactions with the VR controllers. While largely successful, a notable side effect of cybersickness was reported among a group of workers who had no previous experience with video games, highlighting an important consideration for deploying VR to diverse user groups. There are no other significant ethical implications noted for this application. The system's primary contribution to sustainability is economic, through the reduction of costly machine downtime. For standardisation, the platform excels by enabling the deployment of a uniform training standard across Grundfos's global operations. The IVR training is in use and stakeholders testify that with the use of IVR training, they observed an increase in the motivation and decrease of the time required to train employees. In addition, the employment of IVR provided the possibility of a) increased collaboration, as several VR users could train together on the same content regardless of time and space and b) deployment of a uniform training method and scenarios on a global scale [3]. Nevertheless, some side effects like cybersickness, in a group of workers with no previous game experience, was also reported. The IVR scenarios included a realistic representation of the machines and environment, and visual cues to guide the user to perform the training tasks.
© AIOTI. All rights reserved. 80 13.3 DSB Train Operator Training 13.3.1 Scenario Danish State Railways (DSB) faces an operational challenge in training their train operators on critical, time-sensitive procedures such as door and whistle operation. This is a crucial part of the operator's job, both for routine station stops and for handling emergency situations. Figure 13-3 – Immersive Train Operator Training The traditional training method requires decommissioning a functional train, making it unavailable for passenger service. This process causes unwanted downtime and associated costs. Furthermore, this method often fails to provide a realistic training scenario, as it cannot easily replicate the noisy, bustling, and unpredictable environment of a live train station. To address these issues, an immersive virtual reality experience was created to simulate these scenarios effectively. 13.3.2 Users and stakeholders The primary users of this application are the DSB train operator trainees. These individuals need to learn and master specific operational procedures in a safe, repeatable, and realistic environment before they perform them on active train lines. The key stakeholders are the DSB organisation, which benefits from reduced operational downtime and lower training costs, and the expert trainers who manage the learning experience. The trainers use the system not just for direct instruction but also to facilitate reflection and feedback among the trainees, enhancing the overall learning outcome. 13.3.3 Implementation in a virtual world and added values The solution is implemented using Immersive Virtual Reality (IVR), creating a detailed and realistic virtual representation of a train station platform, the train doors, and the immediate interior of the train coach. Trainees can navigate this 3D environment via teleportation while remaining stationary in the physical world. The added value of this virtualized approach is significant. It eliminates the downtime of functional trains, providing substantial cost savings. It also offers a more realistic simulation than traditional methods can achieve, immersing the trainee in a dynamic station environment. The system allows for efficient training in batches and introduces a collaborative element, where one trainee can learn in VR while another observes and provides feedback, fostering peer-topeer learning.
© AIOTI. All rights reserved. 81 13.3.4 Required immersive technologies functionalities The technology platform for this training utilises Virtual Reality hardware, specifically mentioning the HTC Vive and Oculus Quest headsets. The training setup is collaborative, involving one trainee in the IVR headset while a second trainee observes the user's progress on an iPad, a setup managed by expert trainers. Interactions within the virtual world are primarily conducted using two-handed VR controllers. User feedback is multi-modal, incorporating visual cues that highlight points of interest and display information like timers. The system also provides haptic feedback, such as a controller vibration, which acts as a simple nudge to confirm that an action, like opening the doors, has been successfully completed. The system does not currently feature any form of adaptation for specific user skill levels. The types of immersive technologies used are VR and IVR with user interface modes based on VR Controllers, Visual cues. The virtual environment is a detailed representation of the train station platform, the train doors, and the immediate interior of the train coach behind the doors. The trainees navigate the virtual environment through teleportation while standing in one physical location. Visual cues are present to aid the trainee by displaying time and highlight points-of-interest. The trainee uses an Oculus Quest and an iPad, where one trainee experiences IVR training in the Quest while the other observes the progression, and provides feedback. Haptic feedback is present as an indicator of the trainee that certain actions were performed. For example, when a trainee opens the train doors gets vibration feedback to indicate successful completion. This is meant more like a nudge than to represent a physically realistic interaction. No adaptation for specific users was identified. 13.3.5 Technology layers requirements The HTC Vive is used for the currently deployed VR training. For most parts of the training, the interactions involve two hands (using the VR controllers). 13.3.6 Horizontal Issues and Characteristics The system is designed to be immersive and interactive. The training is delivered to groups of trainees, who are split into pairs with one IVR headset and one tablet per pair, creating a unique interactive learning dynamic. The expert trainers play a crucial role in managing the experience and helping the trainees reflect on what they have learned. From a human-centric perspective, the system perceives the user's actions through the tracked VR controllers. Interviews with participants indicate that they perceive the IVR training as a beneficial method and a step in the right direction for modernizing training, even though they acknowledge it still requires further development. The primary contribution to sustainability is economic, through the reduction of costs and the elimination of production train downtime. No significant ethical issues are noted for this application. The interviewees perceived the IVR as a beneficial training method, still requiring development but due to the ever-increasing demand for fast training at a lower cost a step towards the right direction. Training is delivered at the headquarters in batches of 12 trainees, who are split into groups of two. Each group is assigned an IVR headset and a tablet. There are two expert trainers who manage the training experience and aid the trainees reflect on what they learned.
© AIOTI. All rights reserved. 82 13.4 MR Incident Simulator for immersive Command and Control Room Training 13.4.1 Scenario The mixed-reality application called Incident Simulator for Command-and-Control Rooms allows for training purposes to operate a control room that has been ported into a virtual metaverse, enabling the simulation of critical events leveraging on customized DTs. The main aim is to allow dispatchers and control centre staff to practice real-world scenarios in a familiar environment without interrupting the operations of the actual control room. The simulated scenarios for training are drawn from a meta-database, that is populated with domain-specific procedures and guidelines. Trainees interact within a XR environment, featuring virtual multi-monitor panels that replicate the control room setup, and respond to simulated incidents using real-world devices, such as physical phone systems, smartphones, and computers, enhancing the realism and effectiveness of the training experience. Figure 13-4 – MR Incident Simulator. 13.4.2 User and stakeholders Among the users of the application one can mention the staff operating a control room, e.g., dispatchers, control centre staff members, and trainees. 13.4.3 Implementation in a virtual world and added values Among the added values one can list offering the capability of operating a control room in a MR environment, without interrupting the operation of the real control room. Moreover, training purposes and complex scenario management before real problems come into the play are also two important additional added values. 13.4.4 Required immersive technologies functionalities The types of immersive technologies used are VR and XR.
© AIOTI. All rights reserved. 83 To replicate the control room environment, the application immerses trainees in a MR by accurately recreating the control room in a virtual space, allowing them to work with identical systems and interfaces. Rather than relying on game controllers or VR controllers, the simulation integrates actual phones, laptops, and communication tools, reinforcing real-world immersion. So, the user interaction modes are a mix of: VR Controllers, Hand Tracking, Voice Commands, Eye Tracking, Smartphones, real control room equipment. The user interface design uses both 2D and 3D elements with user feedback mechanisms like haptic feedback, audio feedback, and visual Feedback. 13.4.5 Technology layers requirements To monitor and incorporate trainees' stress levels into the training, smartwatches and other wearables can be integrated with the incident simulation. This data enables stress levels based adaptive training adjustments, creating a more immersive and personalized experience. An incident simulation is partially empowered by AI-based modules in multiple use-cases. Large Language models (LLM) are be used to initiate phone-calls to the trainee for immersive communication scenarios within a training setup. Pattern-Recognition models understand the stress-level of the trainee by incorporating data from wearables as well as voice analytics. Measurements of the trainee’s reaction time can help to further streamline the training-program by intelligently adapting the simulation to the trainee’s capabilities. The leading platforms for MR applications are Meta Quest and Vision Pro, as both provide developer SDKs that provide access to a front-facing camera and augmentation of virtual elements. This enables to, e.g., implement multi-panel control-room environments in combination with “see-through” elements in the real-world. 13.4.6 Horizontal issues and characteristics N/A. 13.5 XR-based Remote Collaboration with IoT Contextual Integration 13.5.1 Scenario The application develops an open, versatile, inclusive and scalable digital workplace to facilitate work and social activities between many simultaneous users. It creates an immersive experience by integrating rich contextual information into video streams. Figure 13-5 – XR-based Remote Collaboration
© AIOTI. All rights reserved. 84 13.5.2 User and stakeholders N/A. 13.5.3 Implementation in a virtual world and added values N/A. 13.5.4 Required immersive technologies functionalities The types of immersive technologies used are Metaverse and XR (AR/VR) environments. The user interaction modes are voice recognition/commands, hand/motion tracking, avatars, VR controllers e.g. for natural gestures recognition, virtual Assistants, chatbots. (The user interface design considers 2D/3D elements, and multi-scene navigation. 13.5.5 Technology layers requirements Cloud-native IoT platform for distributed deployment over edge-cloud environments, real-time sensing and rendering, audio/video/text, 3D data types, local/cloud storage, IoT 2D/3D objects. AI-powered meeting assistant with extended capabilities such as natural speech interaction, meeting summarisation or translation, rich semantic annotations for task-oriented dialogue modelling. HTC Vive Focus 3, Pico 4, Meta Quest 2/3, Windows, Unity/WebXR, Android, ALCATEL Rainbow teleconferencing solution, IoT platform/services provided by Intracom Telecom and relevant Javascript/C# SDKs). 13.5.6 Horizontal issues and characteristics User data anonymisation for personal privacy preservation and security protection, compliance with privacy protection regulations (e.g., the EU AI Act), authentication and authorization services for authorised participation to conferences, IoT access and user role management are several horizontal issues and characteristics of these types of use cases. 13.6 Advanced VR simulator for Training Law Enforcement Officers 13.6.1 Scenario The application is a VR-based simulator designed specifically for training police and law enforcement staff. Trainees are immersed in lifelike 3D scenarios that range from complex crime scenes to vehicle accidents and terrorist incidents. They can examine evidence, question virtual witnesses, and make rapid, critical decisions in high-pressure scenarios. The simulator provides real-time feedback and performance analytics, enabling trainees to understand their strengths and target areas for improvement, while engaging in a safe and controlled learning environment. Each environment is carefully crafted with high-resolution textures, accurate physics, and dynamically adjustable conditions, such as varying weather and time-of-day settings. 13.6.2 User and stakeholders Users are staff members of law enforcement officers that are to be trained.
© AIOTI. All rights reserved. 85 Figure 13-6 – Advanced VR Simulator for Training Law Enforcement Officers 13.6.3 Implementation in a virtual world and added values The application uses haptic, audio, and visual feedback inputs from users, allowing for real-time interaction and environmental manipulation based on trainee actions. Officers can utilise voice commands, gestures, and physical navigation tools for enhanced training experience. The simulator supports multiplayer modes, enabling officers to train collaboratively in scenarios that require teamwork and coordination. Scenarios can be customized to reflect local legal frameworks, specific crime trends, and targeted training goals, providing highly relevant and adaptable training experiences 13.6.4 Required immersive technologies functionalities The types of immersive technologies used are VR, XR, and MR. The user interaction modes are based on VR controllers, hand tracking, and voice commands, using a user interface design based on both 2D and 3D navigation. The user feedback mechanisms are haptic feedback, audio Feedback, visual Feedback for enhanced situational awareness. 13.6.5 Technology layers requirements The application assumes a network of devices equipped with cameras, haptics, motion, depth, and proximity sensors. The edge computing has to support CPU/GPU with latency optimization, capable of real-time and 3D data types, spatial and spatial-temporal data processing. Regarding the storage, both local and cloud options should be made available for scalable data handling, in compliancy with GDPR standards. The application has attachable AI modules embedded that complies to strict ethical guidelines. ML for scenario optimization. The platforms used are Oculus Rift and Microsoft HoloLens.
© AIOTI. All rights reserved. 86 13.6.6 Horizontal issues and characteristics The application incorporates modules that are respecting fundamental rights during investigations including respecting the dignity of individuals involved in accidents or criminal incidents, ensuring fair treatment, and understanding the ethical implications of their actions. It generates scenarios that are unbiased and inclusive, providing a wide range of situations without any prejudice or stereotyping N/A. Standard technologies that are relevant for the application are JSON and OpenXR for VR, ensuring compatibility with a variety of HW and SW systems, which allows flexibility in deployment and integration. The application is built on well-known platforms like Oculus Rift with SDKs for compatibility across devices and compliance with OpenXR standards.
© AIOTI. All rights reserved. 87 14 Trustworthiness 14.1 Introduction The convergence of advanced technologies such as edge IoT, AI, DT, immersive triplets, and spatial computing is forging a complex industrial real-digital-virtual continuum. This continuum is realized through immersive environments where users can feel physically present within a computer-generated perceptual context. These industrial immersive solutions are advancing the integration of AR), virtual reality VR, MR, and XR with next-generation concepts like metaverses and the spatial web (Web 4.0). As these technologies become more integrated into critical sectors, ensuring their trustworthiness is a primary concern for widespread adoption and safe implementation [1][2]. Trustworthiness in this context is a multifaceted concept, encompassing the system's security, privacy, reliability, and the ethical integrity of the virtual worlds being created [3]. Significant strides have been made in enhancing the trustworthiness of immersive applications. The development of secure edge computing frameworks has been a key advancement, allowing for the local processing of sensitive data generated by immersive devices. This approach not only reduces latency for a more seamless user experience but also minimizes the exposure of data to potential cyber-attacks during transmission [4]. In parallel, advanced cryptographic techniques, such as homomorphic encryption, are being actively researched to enable secure data analysis in cloud environments, which would allow for computation on user data without decrypting it, thus preserving privacy. Another area of progress is the establishment of interoperability standards. These standards are crucial for creating a cohesive and trusted digital ecosystem, enabling secure and seamless interaction between various immersive platforms and emerging metaverses. Furthermore, AI is being increasingly leveraged for real-time threat intelligence and anomaly detection within these immersive environments. This proactive approach helps to ensure the integrity of the system and the safety of its users. Within the burgeoning metaverse, the application of blockchain technology has shown considerable promise for securing digital asset transactions, verifying ownership, and managing user identities in a decentralised and transparent manner, which fosters greater user trust [3]. The rapid expansion of immersive technologies like VR, AR, and MR is introducing complex challenges at the intersection of cybersecurity, privacy, and user trust. For these technologies to be widely accepted and successful, users must trust them, a feeling rooted in factors like dependability, including safety, security, privacy, reliability, maintainability, resilience, etc. This trust relationship is fundamental, as the very nature of immersive experiences requires users to surrender a significant amount of personal data and sensory control, making the need for robust security and privacy measures more critical than ever before. The core of the immersive ecosystem is built upon the collection and processing of vast amounts of XR data. This data goes far beyond traditional personal information, encompassing sensitive biometric data inferred from eye-tracking and body movements, as well as sensor data capturing a user's physical surroundings. This continuous stream of information is essential for creating the sense of presence and immersion that defines XR. However, the sheer volume and intimate nature of this data create unprecedented privacy risks and security vulnerabilities, making these systems attractive targets for cyberattacks that could lead to identity theft, fraud, and other forms of data misuse. The unique operational requirements of immersive applications render traditional privacy controls largely ineffective.
© AIOTI. All rights reserved. 88 For example, privacy measures like simple camera access indicators are inadequate for devices that must constantly sense and map the environment to function correctly, inadvertently exposing not just the user but also bystanders to surveillance. This highlights a critical challenge: security and privacy solutions must be strong and highly usable. If privacy controls are cumbersome, confusing, or interfere with the immersive experience, they will likely be ignored or disabled, leaving users exposed. Trust and usability of the industrial immersive applications are deeply interdependent in the context of immersive security and privacy. A system that is technically secure but difficult for a user to understand or control will not be trusted. Conversely, a system that prioritises a seamless user experience at the expense of robust security will inevitably lose user trust through breaches and data misuse. Therefore, building safe and secure industrial immersive applications requires a design philosophy that integrates usable end-to-end privacy and security measures from the ground up, ensuring that they are integrated into the system, intuitive, and place minimal cognitive load on the user, thereby fostering a trustworthy relationship between individuals, the technology and the immersive application. 14.2 Key Challenges in Trustworthiness for Immersive Applications Despite these advancements, formidable challenges to achieving robust trustworthiness in immersive applications persist. A primary concern is user privacy, stemming from the vast quantities of sensitive biometric, behavioural, and even neurological data that next-generation XR devices are capable of collecting [5]. Protecting this data from misuse or unauthorized access is a complex technical and ethical problem that has yet to be fully resolved. The sheer diversity of hardware, software, and network conditions presents another substantial hurdle, making it difficult to guarantee a consistent and secure user experience across a fragmented ecosystem of devices and platforms. The potential for perceptual manipulation within immersive environments represents a critical and unique challenge. Malicious actors could theoretically alter a user's perception of reality, which could lead to significant physical or psychological harm. This underscores the importance of ensuring the authenticity and integrity of all digital content and interactions within these virtual worlds. The absence of universally accepted standards for security, data privacy, and content verification continues to impede the development of a fully interoperable and trustworthy immersive web. 14.3 Future Research Trends Future research must pursue a holistic and interdisciplinary approach to building trust in immersive systems. This involves developing comprehensive trust models that account for the intricate interplay between technical, social, and ethical factors. There is a pressing need for the creation of adaptive security frameworks that can dynamically evolve to identify and neutralize emerging threats in real time. Advancing research into explainable AI (XAI) will also be crucial for making the decision-making processes of these complex systems transparent, thereby fostering greater understanding and trust from users. The integration of digital twins with immersive environments for industrial applications, such as predictive maintenance and the simulation of complex operations, offers a significant opportunity to build highly reliable and trusted solutions. The exploration of decentralised architectures, particularly those based on blockchain and distributed ledger technologies, will likely play a central role in the future of the metaverse, enhancing security and empowering users with greater control over their personal data and digital identities [3]. Ultimately, the establishment of standardised protocols for secure data exchange and device interoperability will be the foundational pillar upon which the trustworthy next-generation spatial web, or Web 4.0, is built.
© AIOTI. All rights reserved. 95 Gaining consensus on such complex, multi-faceted issues among diverse stakeholders with competing interests is a slow and arduous process. Furthermore, the rapid pace of innovation in immersive technology means that standards can risk becoming obsolete before they are even widely adopted. The process of developing and ratifying a formal standard is often deliberate and time-consuming, while the technology itself is evolving at an exponential rate. This creates a difficult balancing act: standards must be stable and robust enough to provide a reliable foundation, yet flexible enough to accommodate future technological breakthroughs. Ensuring that standards can evolve and adapt without breaking backward compatibility is a critical and ongoing challenge for the entire industry. 17.3 Future Research Trends in Standardisation and Interoperability for Immersive Applications Looking forward, research in immersive interoperability is moving beyond basic asset formats and device APIs to address more complex and nuanced challenges. A key area of future research is semantic interoperability. This goes beyond simply being able to exchange data; it means ensuring that the meaning of that data is understood consistently across different systems. This involves developing common ontologies and data models that can describe the properties, behaviours, and relationships of objects and users in a virtual world. For example, a semantic standard would ensure that a "chair" object is recognized as a sittable object with consistent properties in any virtual environment that adheres to the standard [30]. Another critical research trend is the development of frameworks for cross-platform identity and data management. For a truly interoperable metaverse, users need a persistent digital identity that they control and can use across different virtual worlds, along with the ability to manage their personal data, assets, and social graphs in a secure and private manner. Research into decentralised identity systems, leveraging technologies like blockchain and verifiable credentials, is exploring ways to give users sovereignty over their digital selves, breaking down the data silos created by today's platform-centric models [31]. Future research will also focus on the integration of AI with standardisation efforts. AI can be used to automate the process of converting assets between different formats and to dynamically broker communication between systems that do not share a common standard. Furthermore, research is needed to develop standards for the ethical and responsible use of AI within immersive environments, ensuring that AI-driven avatars and systems behave in a predictable and trustworthy manner across different platforms. The goal is to create a dynamic, integrated ecosystem where standardisation enables not just technical connectivity, but a truly seamless and meaningful exchange of experiences across a vast network of virtual worlds. Standardisation and interoperability are the foundational pillars upon which the future of the industrial real-digital-virtual continuum will be built. They are the invisible infrastructure that will enable the creation of a persistent, open, and interconnected metaverse, rather than a fragmented collection of disconnected virtual experiences. While significant progress has been made in developing open standards for hardware, assets, and communication, formidable challenges related to market fragmentation, technological complexity, and corporate competition remain. Overcoming these challenges will require a sustained and collaborative effort from across the industry, academia, and the public sector. The future of research lies in tackling higher-level problems like semantic interoperability, decentralised identity, and the standardisation of AI behaviours. By investing in these areas, we can build the common language needed to unlock the full potential of immersive technologies, fostering an environment rich with innovation, economic opportunity, and shared human experience. The journey is complex, but the creation of a truly open and interoperable spatial web is a goal worthy of the effort.
© AIOTI. All rights reserved. 96 18 Conclusions The fusion of disruptive technologies, including the IoT, edge computing, AI, and spatial computing, is driving a period of unprecedented innovation. This convergence sparks the crosspollination of ideas, creating novel approaches and transforming the industrial landscape. The result is a new generation of smart devices, integrated systems, and innovative services that are redefining business models and operational efficiencies across sectors. Immersive technology stands at the heart of this transformation, offering a revolutionary approach to creating digital experiences that feel tangibly real. By blurring the line between the physical and digital worlds, these technologies cultivate a profound sense of presence and engagement. They transport users into rich virtual environments or augment their real-world surroundings with interactive digital information, enabling seamless interactions across physical, digital, and spatial domains. The broad adoption and successful implementation of immersive experiences are fundamentally enabled by advancements in IoT and edge computing. These technologies provide the critical infrastructure for real-time data processing and distributed intelligence. By bringing computational power closer to the source of data, IoT and edge computing facilitate the low-latency interactions required for a fluid and responsive connection between the physical and virtual worlds. Developing such technologies requires a holistic and interdisciplinary approach. The complexity of integrating hardware, software, connectivity, and user experience demands collaboration across diverse fields of expertise, from network engineering to human-computer interaction. This collaborative spirit is a vital complement to the growth of the individual disciplines, serving as a powerful driver of knowledge creation, research, and genuine innovation. Key research and innovation efforts within the IoT, AI and edge computing continuum are focused on critical challenges. These include designing resilient distributed architectures, ensuring intelligent connectivity, and providing end-to-end security for heterogeneous systems. Furthermore, progress in areas like IoT-enabled digital twins, immersive triplets, swarm intelligence, and establishing trustworthiness through rigorous verification and validation is paving the way for a true Internet of Intelligent Things. Looking ahead, the evolution of the internet is pointing towards Web 4.0, a new era conceived as a decentralised online ecosystem. Built upon blockchain technology, this next-generation web aims to shift control and data ownership from centralised corporations back to the users. The principles of Web 4.0 promise to create a more open, transparent, and user-empowered digital realm, profoundly changing how we interact with data and virtual environments. The convergence of IoT, AI, digital twins, and spatial computing is forging a robust industrial immersive continuum. This continuum represents a new paradigm for digital transformation, where the distinctions between the real, digital, and virtual worlds become increasingly fluid. It allows human interaction and industrial processes to transcend traditional physical limitations, opening new possibilities for innovation and efficiency. Within this continuum, industrial immersive solutions are evolving from standalone applications into highly integrated platforms. These platforms enable a continuous and bidirectional flow of data between physical assets and their dynamic digital counterparts. This synergy allows for more innovative design processes, highly efficient operations, and more effective, contextualised training programs that fundamentally alter how we work, learn, and collaborate. Several key technological advancements have been instrumental in enabling the widespread adoption of these immersive solutions.
© AIOTI. All rights reserved. 97 The development of high-fidelity, lightweight, and untethered XR headsets has dramatically improved user comfort and mobility, making them practical for dynamic industrial settings. Simultaneously, progress in computer vision and AI-powered spatial mapping allows devices to understand and interact with the physical world in real time. The integration of digital twin and immersive triplets’ technologies is a particularly impactful advancement. It allows users to create detailed, data-rich virtual replicas of physical assets, processes, and entire systems. When combined with immersive environments, these twins and triplets become practical tools for simulation, remote monitoring, and predictive maintenance, unlocking ample value and operational insight. Supporting these applications, the maturation of cloud and edge computing has provided the necessary computational power to render complex virtual scenes and process vast datasets with minimal latency. This infrastructure is critical for creating the responsive, collaborative experiences that are essential for industrial use cases, from the factory floor to the operating room. Immersive industrial applications, such as AR, VR, real-time digital twins and immersive triplets, are transformative for industrial environments and place extreme demands on network infrastructure. They require a combination of high data rates for video streams, ultra-low latency for real-time interaction and control, and very high reliability to prevent operational failures. 5G/6G NPNs are uniquely designed to address these stringent, end-to-end requirements through a combination of key technological features such as network slicing and QoS control, where NPNs enable the creation of multiple virtual networks, or "slices," on a single physical infrastructure. Each slice can be configured with its own guaranteed quality of service (QoS) parameters. For an immersive AR application, a dedicated slice can be provisioned with high bandwidth and low latency, ensuring that the video feed remains smooth and interactive, completely isolated from other network traffic like email or sensor data uploads. The synergy between NPNs and multi-access edge computing (MEC) is critical. By deploying compute and data storage resources at the edge of the private network, physically close to the devices and users, data processing occurs on-site. For remote-controlled devices, this minimises the round-trip time for data, reducing latency from hundreds of milliseconds (on a typical cloud setup) to just a few milliseconds. This immediate processing is what makes realtime control and immersive overlays possible. In contrast to public mobile networks, which can suffer from unpredictable congestion, an NPN provides a "controlled radio environment. By operating on dedicated or licensed spectrum, the network is immune to performance degradation from public user traffic. It ensures consistent, predictable performance, which is essential for mission-critical industrial processes where a dropped connection or lag spike could lead to production halts or safety incidents. 5G standards specifically define URLLC to support applications requiring less than 1ms latency and 99.999% reliability. NPNs enable organisations to fine-tune network parameters, prioritising URLLC traffic, which is crucial for applications such as autonomous systems and collaborative robotic systems that rely on instantaneous communication for safe and efficient operation. In an NPN, the data can be kept within the enterprise's private domain. This on-premises data handling is fundamental for industrial applications dealing with sensitive intellectual property, proprietary operational data, or regulated information. This robust security model builds trust and ensures that immersive applications do not create new vulnerabilities. NPNs can be designed to provide uniform, high-performance coverage across complex industrial environments like factories, ports, or mines. Using technologies like beamforming and optimisation of the wireless cells’ placement, the network can be tailored to eliminate dead zones and deliver consistent signal strength to both stationary and mobile devices, ensuring uninterrupted QoE for users and devices moving across a large site.
© AIOTI. All rights reserved. 98 All these features create a continuous cycle that can guarantee bandwidth and low latency, eliminate motion conditions in VR and lag in AR, making the experience more natural and effective. High reliability ensures that critical operations for the immersive applications are never compromised. By providing secure, customisable, and high-performance networking environments, 5G/6G NPNs move immersive technologies from a novelty to a reliable and indispensable industrial tool, dramatically improving both operational efficiency and the human quality of experience. As these technologies mature, they lay the groundwork for next-generation virtual worlds, including concepts like the metaverse and omniverse. While these future environments offer immense opportunities, their development is accompanied by significant technical, societal, and economic challenges. A careful and proactive approach is required to balance the potential rewards against the inherent risks from the earliest stages of design and deployment. Ultimately, the convergence of IoT and industrial immersive applications heralds a significant technological, cultural, and economic shift. It promises a more integrated, interactive, and intelligent future, altering how we perceive and interact with both the digital and physical worlds. The opportunities for unlocking unprecedented capabilities and driving progress across nearly every sector of the economy and society are truly profound. This position paper gives an overview of edge IoT industrial immersive applications across industrial sectors such as culture and heritage, manufacturing, automotive, energy, buildings, mobility and transportation, healthcare, agriculture, and tourism, highlighting how immersive technologies are catalysing a profound transformation across these diverse range of industries, moving beyond niche applications to become a cornerstone of modern digital strategy. By blending the physical and digital worlds, these technologies are unlocking unprecedented efficiencies, enhancing safety, and creating entirely new ways to design, operate, and interact with complex systems. From the factory floor to the operating room, immersive solutions are redefining the boundaries of what is possible. The automotive industry has been a vanguard in this technological shift, embedding immersive tools throughout the entire vehicle lifecycle. In the design phase, virtual reality enables global teams to collaborate on full-scale digital prototypes, significantly accelerating development and reducing reliance on costly physical models. On the assembly line, augmented reality provides technicians with interactive, in-situ instructions, dramatically improving accuracy and efficiency. For consumers, immersive showrooms offer engaging and personalised ways to explore and customise vehicles. In manufacturing, immersive technologies are the engine of the "factory of the future." Companies are leveraging digital twins of their production lines to simulate and optimise layouts in virtual reality before committing to physical changes, a process known as virtual commissioning. This drastically cuts down on setup times and mitigates production risks. VR also provides a safe, repeatable environment for training workers on complex or hazardous machinery, while AR-powered remote assistance connects on-site technicians with global experts, minimising downtime. The agricultural sector is leveraging immersive tech to advance precision farming. By using augmented reality, farmers can visualise real-time data from drones and sensors directly over their fields, enabling highly targeted applications of water, fertilisers, and pesticides. This datadriven approach, often linked to a comprehensive digital twin of the farm, optimises resource use and boosts crop yields sustainably. In the energy sector, immersive applications are critical for enhancing safety and operational efficiency. VR is used for high-stakes emergency training, allowing workers to practice procedures for scenarios like oil rig blowouts or wind turbine maintenance in a risk-free environment. On-site, technicians use AR and MR to overlay real-time data and maintenance
© AIOTI. All rights reserved. 99 records onto complex equipment, improving accuracy and worker safety in power plants and substations. Healthcare is one of the most impactful domains for these technologies. Virtual reality is revolutionising surgical training by allowing surgeons to rehearse complex procedures in hyperrealistic simulations. In the operating room, augmented reality provides surgeons with "X-ray vision," overlaying 3D patient scans directly onto the body to guide instruments with incredible precision. Furthermore, VR is proving to be a powerful tool for pain management and mental health therapy. The Architecture, Engineering, and Construction (AEC) sector has revolutionised its workflows with immersive technology. Architects offer clients immersive VR walkthroughs to experience and refine designs before construction begins. On-site, AR is used to overlay building information models (BIM) onto the physical structure, ensuring accuracy and preventing costly clashes between systems like plumbing and electrical. The transportation and logistics sectors are also realising significant gains. Urban planners use VR to simulate traffic flows and test new infrastructure, while AR applications guide warehouse workers through picking orders with hands-free, visual cues, boosting speed and accuracy. In heritage and tourism, immersive tech offers powerful new ways to experience culture, from bringing museum artefacts to life with AR to creating virtual reconstructions of historical sites, making our shared history more accessible and engaging than ever before.
© AIOTI. All rights reserved. 100 References [1] R. T. Azuma, “A Survey of Augmented Reality,” Presence: Teleoperators and Virtual Environments, vol. 6, no. 4, pp. 355–385, 1997, https://doi.org/10.1162/pres.1997.6.4.355. [2] P. Milgram, and F. Kishino. “A Taxonomy of Mixed Reality Visual Displays”. IEICE Transactions on Information and Systems, E77-D (12), pp. 1321-1329. 1994. https://cs.gmu.edu/~zduric/cs499/Readings/r76JBoMilgram_IEICE_1994.pdf. [3] L. C. Lee et al., “All One Needs to Know about Metaverse: A Complete Survey on Technological Singularity, Virtual Ecosystem, and Research Agenda,” arXiv (Cornell University), Oct. 2021, https://doi.org/10.48550/arxiv.2110.05352. [4] H. Ning et al., “A Survey on Metaverse: the State-of-the-art, Technologies, Applications, and Challenges,” arXiv (Cornell University), Nov. 2021, doi: https://doi.org/10.48550/arxiv.2111.09673. [5] M. Zyda, "Let’s Rename Everything “the Metaverse!”," in Computer, vol. 55, no. 3, pp. 124-129, March 2022, https://doi.org/10.1109/MC.2021.3130480. [6] H. Wang et al., "A Survey on the Metaverse: The State-of-the-Art, Technologies, Applications, and Challenges," in IEEE Internet of Things Journal, vol. 10, no. 16, pp. 14671-14688, 15 Aug.15, 2023, https://doi.org/10.1109/JIOT.2023.3278329. [7] M. Slater et al., “The Ethics of Realism in Virtual and Augmented Reality,” Frontiers in Virtual Reality, vol. 1, no. 1, Mar. 2020, https://doi.org/10.3389/frvir.2020.00001. [8] J. Lanier, Dawn of the new everything: encounters with reality and virtual reality. New York: Picador/Henry Holt and Company, 2018. https://us.macmillan.com/books/9781250097408/dawnoftheneweverything/. [9] J. Bailenson, “Experience on Demand: What Virtual Reality Is, How It Works, and What It Can Do | Program in Science, Technology & Society,” sts.stanford.edu, 2018. https://sts.stanford.edu/publications/experience-demandwhat-virtual-reality-how-it-works-and-what-it-can-do. [10] IEEE. "Ethically Aligned Design - A Vision for Prioritizing Human Well-being with Autonomous and Intelligent Systems," in Ethically Aligned Design - A Vision for Prioritizing Human Well-being with Autonomous and Intelligent Systems, vol., no., pp.1-263, Version 2. https://standards.ieee.org/wp-content/uploads/import/documents/other/ead_v2.pdf. [11] Y. K. Dwivedi et al., “Metaverse marketing: How the Metaverse Will Shape the Future of Consumer Research and Practice,” Psychology & Marketing, vol. 40, no. 4, pp. 750–776, Dec. 2022, https://doi.org/10.1002/mar.21767. [12] M. Madary and T. K. Metzinger, “Real Virtuality: A Code of Ethical Conduct. Recommendations for Good Scientific Practice and the Consumers of VR-Technology,” Frontiers in Robotics and AI, vol. 3, Feb. 2016, https://doi.org/10.3389/frobt.2016.00003. [13] D. J. Chalmers, “The Virtual and the Real,” Disputatio, vol. 9, no. 46, pp. 309–352, Nov. 2017, https://doi.org/10.1515/disp-2017-0009. [14] S. Zuboff, The Age of Surveillance Capitalism. PublicAffairs, 2017. https://www.hachettebookgroup.com/titles/shoshana-zuboff/the-age-of-surveillancecapitalism/9781610395694/?lens=publicaffairs. [15] Z. Allam, A. Sharifi, S. E. Bibri, D. S. Jones, and J. Krogstie, “The Metaverse as a Virtual Form of Smart Cities: Opportunities and Challenges for Environmental, Economic, and Social Sustainability in Urban Futures,” Smart Cities, vol. 5, no. 3, pp. 771–801, Jul. 2022. https://www.mdpi.com/2624-6511/5/3/40. [16] A. Patney et al., “Towards foveated rendering for gaze-tracked virtual reality,” ACM Transactions on Graphics, vol. 35, no. 6, pp. 1–12, Nov. 2016, https://doi.org/10.1145/2980179.2980246. [17] M. S. Elbamby, C. Perfecto, M. Bennis and K. Doppler, "Toward Low-Latency and Ultra-Reliable Virtual Reality," in IEEE Network, vol. 32, no. 2, pp. 78-84, March-April 2018, https://doi.org/10.1109/MNET.2018.1700268. [18] L. Lannelongue, J. Grealey, and M. Inouye, “Green Algorithms: Quantifying the Carbon Footprint of Computation,” Advanced Science, vol. 8, no. 12, p. 2100707, May 2021, https://doi.org/10.1002/advs.202100707. [19] E. Williams, “Environmental effects of information and communications technologies,” Nature, vol. 479, no. 7373, pp. 354–358, Nov. 2011, https://doi.org/10.1038/nature10682. [20] R. Schwartz, J. Dodge, N. A. Smith, and O. Etzioni, “Green AI,” Communications of the ACM, vol. 63, no. 12, pp. 54– 63, Nov. 2020, https://doi.org/10.1145/3381831. [21] M. Hatami, Q. Qu, Y. Chen, Hisham Kholidy, and E. Ardiles-Cruz, “A Survey on Real-Time Metaverse: Challenges and Opportunities,” Sep. 10, 2024. https://www.researchgate.net/publication/384098636_A_Survey_on_RealTime_Metaverse_Challenges_and_Opportunities. [22] T. Baidya and S. Moh, “Comprehensive survey on resource allocation for edge-computing-enabled metaverse,” Computer Science Review, vol. 54, p. 100680, Nov. 2024, https://doi.org/10.1016/j.cosrev.2024.100680. [23] R. Raman, P. Kautish, A. Siddiqui, M. Siddiqui, and P. Nedungadi, “The role of metaverse technologies in energy systems towards sustainable development goals,” Energy Reports, vol. 13, pp. 4459–4476, Jun. 2025, https://doi.org/10.1016/j.egyr.2025.04.012.
© AIOTI. All rights reserved. 101 [24] O. Vermesan and V. Frascolla (Eds,). Edge IoT Industrial Immersive Technologies and Spatial Computing Continuum. 2024. AIOTI. https://aioti.eu/wp-content/uploads/AIOTI-Paper-Edge-AI-IoT-Immersive-TechnologiesPublished.pdf. [25] R. Chatila and J. C. Havens, “The IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems,” Robotics and Well-Being, vol. 95, pp. 11–16, 2019, https://doi.org/10.1007/978-3-030-12524-0_2. [26] S. Mystakidis, “Metaverse,” Encyclopedia, vol. 2, no. 1, pp. 486–497, Feb. 2022, https://doi.org/10.3390/encyclopedia2010031. [27] OpenXR. (2021). The OpenXR Specification. Khronos Group. “The OpenXRTM Specification The Khronos ® OpenXR Working Group.” Accessed: Aug. 30, 2025. [Online]. Available: https://registry.khronos.org/OpenXR/specs/1.0khr/pdf/xrspec.pdf. [28] A.-M. Halacheva, Y. Miao, J.-N. Zaech, X. Wang, V. Gool, and P. D. Pani, “Holistic Understanding of 3D Scenes as Universal Scene Description,” arXiv (Cornell University), Dec. 2024, https://doi.org/10.48550/arxiv.2412.01398. [29] Metaverse Standards Forum. (2022). Building the Foundation for an Open Metaverse. [30] J. D. N. Dionisio, W. G. B. III, and R. Gilbert, “3D Virtual worlds and the metaverse,” ACM Computing Surveys, vol. 45, no. 3, pp. 1–38, Jun. 2013, dhttps://doi.org/10.1145/2480741.2480751. [31] L. H. Lee, et al. (2021). All One Needs to Know about Metaverse: A Complete Survey on Technological Singularity, Virtual Ecosystem, and Research Agenda. arXiv preprint arXiv:2110.05352. https://doi.org/10.48550/arXiv.2110.05352. [32] J. Radoff, “The Metaverse Value-Chain,” Building the Metaverse, Jun. 03, 2021. https://medium.com/building-themetaverse/the-metaverse-value-chain-afcf9e09e3a7. [33] J. Radoff, “Market Map of the Metaverse,” Medium, Jul. 29, 2021. https://medium.com/building-themetaverse/market-map-of-the-metaverse-8ae0cde89696. [34] M. Koziol, “The Metaverse Needs Standards, Too,” IEEE Spectrum, Aug. 31, 2022. https://spectrum.ieee.org/metaverse-standards-forum. [35] The Metaverse Standards Forum. https://metaverse-standards.org/. [36] S. Mahdavi and S. Sojoodi, “Impact of ICT on Environment,” Oct. 2021, https://doi.org/10.21203/rs.3.rs-1020622/v1. [37] F. Tao, M. Zhang, Y. Liu, and A. Y. C. Nee, “Digital twin driven prognostics and health management for complex equipment,” CIRP Annals, vol. 67, no. 1, pp. 169–172, 2018, https://doi.org/10.1016/j.cirp.2018.04.055. [38] R. Palmarini, J. A. Erkoyuncu, R. Roy, and H. Torabmostaedi, “A systematic review of augmented reality applications in maintenance,” Robotics and Computer-Integrated Manufacturing, vol. 49, pp. 215–228, Feb. 2018, https://doi.org/10.1016/j.rcim.2017.06.002. [39] S. P. Bhagwat et. al. “VR/AR in the Energy Sector. VR/AR Association White Paper, April 2021. https://www.oberontech.com/wp-content/uploads/2021/05/VRARA-ENERGY-WHITEPAPER-FINAL-April2021.pdf. [40] A. G. Gallagher and C. U. Cates, “Virtual reality training for the operating room and cardiac catheterisation laboratory,” The Lancet, vol. 364, no. 9444, pp. 1538–1540, Oct. 2004, https://doi.org/10.1016/s0140-6736(04)17278-4. [41] P. Pratt et al., “Through the HoloLensTM looking glass: augmented reality for extremity reconstruction surgery using 3D vascular models with perforating vessels,” European Radiology Experimental, vol. 2, no. 1, Jan. 2018, https://doi.org/10.1186/s41747-017-0033-2. [42] B. Schwerdtfeger et al., “Pick-by-Vision: A first stress test,” 2009 8th IEEE International Symposium on Mixed and Augmented Reality, Oct. 2009, https://doi.org/10.1109/ismar.2009.5336484. [43] V. Vlahakis et al., “Archeoguide: an augmented reality guide for archaeological sites,” IEEE Computer Graphics and Applications, vol. 22, no. 5, pp. 52–60, Sep. 2002, https://doi.org/10.1109/mcg.2002.1028726. [44] D. A. Guttentag, “Virtual reality: Applications and Implications for Tourism,” Tourism Management, vol. 31, no. 5, pp. 637–651, Oct. 2010, https://doi.org/10.1016/j.tourman.2009.07.003. [45] X. Wang, P. E. D. Love, M. J. Kim, C.-S. Park, C.-P. Sing, and L. Hou, “A conceptual framework for integrating building information modeling with augmented reality,” Automation in Construction, vol. 34, pp. 37–44, Sep. 2013, https://doi.org/10.1016/j.autcon.2012.10.012. [46] X. Wang, M. J. Kim, P. E. D. Love, and S.-C. Kang, “Augmented Reality in built environment: Classification and implications for future research,” Automation in Construction, vol. 32, pp. 1–13, Jul. 2013, https://doi.org/10.1016/j.autcon.2012.11.021. [47] H. Nassereddine, A. S. Hanna, D. Veeramani, and W. Lotfallah, “Augmented Reality in the Construction Industry: Use-Cases, Benefits, Obstacles, and Future Trends,” Frontiers in Built Environment, vol. 8, Apr. 2022, https://doi.org/10.3389/fbuil.2022.730094. [48] S. Fathi, S. Sabeti, Omidreza Shoghli, Arsalan Heydarian, and Vahid Balali, “Adoption of virtual and augmented reality in the architecture, engineering, construction, and facilities management (AEC-FM): mixed method analysis of trends, gaps, and solutions,” Frontiers in Built Environment, vol. 11, Jun. 2025, https://doi.org/10.3389/fbuil.2025.1580639.
© AIOTI. All rights reserved. 102 [49] W. Y. Leong, Y. Z. Leong, and R. Kumar, “Virtual Reality for Sustainable Energy,” pp. 54–58, Jan. 2025, https://doi.org/10.1109/ic3ecsbhi63591.2025.10991138. [50] C. Verdouw, B. Tekinerdogan, A. Beulens, and S. Wolfert, “Digital twins in smart farming,” Agricultural Systems, vol. 189, p. 103046, Apr. 2021, https://doi.org/10.1016/j.agsy.2020.103046. [51] F. Benaben, A. Congès, and A. Fertier, “A prospective vision of the evolution of immersive technologies: Towards a definition of metaverse,” Technovation, vol. 140, p. 103154, Feb. 2025, https://doi.org/10.1016/j.technovation.2024.103154. [52] M. Uddin, S. Manickam, H. Ullah, M. A. Obaidat, and Abdulhalim Dandoush, “Unveiling the Metaverse: Exploring Emerging Trends, Multifaceted Perspectives, and Future Challenges,” IEEE Access, vol. 11, pp. 1–1, Jan. 2023, https://doi.org/10.1109/access.2023.3281303. [53] S. Zhang et al., "Industrial Metaverse: Enabling Technologies, Open Problems, and Future Trends," in IEEE Communications Surveys & Tutorials, https://doi.org/10.1109/COMST.2025.3563919. [54] M. Dohler, Y.-G. Kim, Mar Gonzalez Franco, and T. West, “Making the Metaverse A Reality: Challenges and Opportunities,” IEEE communications magazine, vol. 61, no. 9, pp. 14–15, Sep. 2023, https://doi.org/10.1109/mcom.2023.10268119. [55] T. A. Hemphill, “The ‘Metaverse’ and the challenge of responsible standards development,” Journal of Responsible Innovation, vol. 10, no. 1, Jan. 2023, https://doi.org/10.1080/23299460.2023.2243121. [56] M. Koziol, “The Metaverse Needs Standards, Too,” IEEE Spectrum, Aug. 31, 2022. https://spectrum.ieee.org/metaverse-standards-forum. [57] GS1 US. Building the Metaverse: A Foundation of Standards. Release 1.0 December 2022. https://documents.gs1us.org/adobe/assets/deliver/urn:aaid:aem:62e09bb7-370c-4938-ab86-9f60218fd6ee/GS1-USBuilding-the-Metaverse.pdf. [58] Y. Wu, Y. Chen, Y. Zhang, and Y. Liu, “Building Metaverse: Propose a design framework for the metaverse based on the requirements,” pp. 107–113, Sep. 2024, https://doi.org/10.1145/3697355.3697375. [59] Standards Australia. The Metaverse and Standards. Whitepaper. May 2023. https://doi.org/10.13140/RG.2.2.14650.77763. [60] Y. Xu, D. Feng, M. Zhao, Y. Sun, and X.-G. Xia, “Edge Intelligence Empowered Metaverse: Architecture, Technologies, and Open Issues,” IEEE network, pp. 1–1, Jan. 2024, https://doi.org/10.1109/mnet.2023.3317477. [61] W. Y. B. Lim et al., "Realizing the Metaverse with Edge Intelligence: A Match Made in Heaven," in IEEE Wireless Communications, vol. 30, no. 4, pp. 64-71, August 2023, https://doi.org/10.1109/MWC.018.2100716. [62] M. Aloqaily, O. Bouachir, I. A. Ridhawi and M. Guizani, "Realizing the Metaverse in the 6G Era with AI-Enabled Network Orchestration," in IEEE Network, vol. 37, no. 2, pp. 78-85, March/April 2023, https://doi.org/10.1109/MNET.002.2200271. [63] L. Chang et al., “6G-enabled Edge AI for Metaverse: Challenges, Methods, and Future Research Directions,” arXiv (Cornell University), Apr. 2022, https://doi.org/10.48550/arxiv.2204.06192. [64] M. Grieves. Digital Twin: Manufacturing Excellence through Virtual Factory Replication. https://www.3ds.com/fileadmin/PRODUCTS-SERVICES/DELMIA/PDF/Whitepaper/DELMIA-APRISO-Digital-TwinWhitepaper.pdf. [65] A. Vigkos, et. Al. 2022. Virtual and Augmented Reality Industrial Coalition. Strategic Paper. https://op.europa.eu/en/publication-detail/-/publication/9aaef6fd-28db-11ed-8fa0-01aa75ed71a1/language-en. [66] Y. K. Dwivedi, “Metaverse beyond the hype: Multidisciplinary Perspectives on Emerging challenges, opportunities, and Agenda for research, Practice and Policy,” International Journal of Information Management, vol. 66, no. 66, 2022, https://www.sciencedirect.com/science/article/pii/S0268401222000767. [67] A. Meige, et.al. The Industrial Metaverse. Making the invisible visible to drive sustainable growth. July 2023. Arthur D. Little. https://www.adlittle.com/sites/default/files/reports/ADL_BLUE%20SHIFT_Industrial_metaverse_2023_0.pdf. [68] I. Hupont Torres, et. Al. Joint Research Centre, “Next generation virtual worlds - Publications Office of the EU,” Publications Office of the EU, 2023. JRC133757. https://op.europa.eu/en/publication-detail/- /publication/775e6eec-16f5-11ee-806b-01aa75ed71a1/language-en. [69] 5G Americas Report, “3GPP Technology Trends”. https://www.5gamericas.org/wp-content/uploads/2024/01/3GPPTechnology-Trends-WP.pdf [70] J. Aberger, S. Shami, B. Häcker, J. Pestana, K. Khodier, and R. Sarc, “Prototype of AI-powered assistance system for digitalisation of manual waste sorting,” Waste Management, vol. 194, pp. 366–378, Jan. 2025, https://doi.org/10.1016/j.wasman.2025.01.027. [71] J. Chen, Y. Fu, W. Lu, and Y. Pan, “Augmented reality-enabled human-robot collaboration to balance construction waste sorting efficiency and occupational safety and health,” Journal of Environmental Management, vol. 348, p. 119341, Dec. 2023, https://doi.org/10.1016/j.jenvman.2023.119341. [72] DDS-MAP. https://ddsmap.easpd.eu/.
© AIOTI. All rights reserved. 103 Contributors Editors: Ovidiu Vermesan, SINTEF, Norway Valerio Frascolla, Intel, Germany Reviewer: Damir Filipovic, AIOTI Secretary General Contributors (alphabetic order): Alain Pagani, German Research Center for Artificial Intelligence, Germany Albena Mihovska, Research and Development and Innovation Consortium, Bulgaria Andreas El Saer, KONNECTA Systems, Greece Ángel Martín Navas, VICOMTECH, Spain Aristea Zafeiropoulou, KONNECTA Systems, Greece Artur Krukowski, Intracom Telecom, Greece Asbjørn Hovstø, Hafenstrom, Norway Björn Debaillie, imec, Belgium Cian O Murchu, Tyndall National Institute, Ireland Daniela Buleandra, SIMAVI, Romania Eridy Lukau, Fokus Fraunhofer, Germany Frances Cleary, Walton institute, SETU, Ireland Francesco Chinello, Aarhus University, Denmark François Fischer, FSCOM, France George Suciu, BEIA, Romania George Tsakiris, KONNECTA Systems, Greece Guido Perricone, Brembo N.V., Italy Hazel Peavoy, Walton Institute, Ireland Ignacio Lacalle, Universitat Politecnica de Valencia, Spain Ilia Pietri, Intracom Telecom, Greece Jesus Angel Garcia Sanchez, Indra Sistemas, Spain Joachim Hillebrand, Virtual Vehicle Research, Austria Joao Sousa, CCG, Portugal Konstantinos Koumaditis, Aarhus University, Denmark
© AIOTI. All rights reserved. 104 Konstantinos Loupos, INLECOM Innovation, Greece Leonidas Valavanis, KONNECTA Systems, Greece Maria Xezonaki, Intracom Telecom, Greece Martin Serrano, Insight SFI research Centre for Data Analytics, Ireland Matthias Hartmann, imec, Belgium Mirko Presser, Aarhus University, Denmark Monica Florea, SIMAVI, Romania Natalie Samovich, Enercoutim, Portugal Ovidiu Vermesan, SINTEF, Norway Pierre Yves Danet, 48deg79min-Consulting, France Pietri Ilia, Intracom Telecom, Greece Ranga Rao Venkatesha Prasad, Technical University Delf, Netherlands Ronald Maandonks, Signify, The Netherlands Roumen Nikolov, Virtech, Bulgaria Roy Bahr, SINTEF, Norway Roy-Inge Eilertsen, Livland AS, Norway Sergio Gusmeroli, Politecnico di Milano, Italy Silvia Romana Ottaviani, SIAD Macchine Impianti S.p.A., Italy Tina Katika, ICCS, Greece Kostas Naskou, ICCS, Greece Udayanto Dwi Atmojo, Aalto University, Finland Valerio Frascolla, Intel, Germany Vasileios Karagiannis, Austrian Institute of Technology, Austria Veronica Antonello, TXT e-tech, Italy Veronica Quintuna Rodriguez, Orange, France Vladimir Poulkov, Technical University of Sofia, Bulgaria