Towards Intelligent Costruction Machines: the Gigabot

Towards Intelligent Construction Machines: the Gigabot Gabriele Ribolla1,2†, Alexandr Rucodainii3†, Francesca Negrello1†, and Manuel G. Catalano1,2† Abstract—The shortage of skilled operators and increasing safety requirements in harsh environments like construction sites, forests, and mines demand innovation in construction machinery. Simplifying control and enhancing machine intelligence are essential to address these challenges. Inspired by robot-avatar concept, which enables remote human activity via intuitive interfaces, we explored how to improve construction machine usability, making them safer and accessible to untrained users. Motivated by this, we developed an industrial Gigabot architecture composed by cartesian control and teleoperation technologies. The system was validated operating a crane remotely at the Bauma 2025, the world’s leading construction machinery fair. Thanks to the positive feedback received by crane operators and interest for potential application, the industrial partner is currently industrializing it. Index Terms—Construction Robotics, Remote Control, Intuitive Interfaces, Human-Robot-Interface, Hydraulic Systems I. INTRODUCTION Currently, the construction industry faces a growing shortage of experienced machine operators, along with persistent safety concerns [1]. This is particularly relevant for heavy-duty equipment such as loader cranes, which are widely used for lifting operations on building sites. These machines generally require precise, joint-level control of hydraulic actuators, a process that calls for extensive training and continuous focus from the operator. To overcome these difficulties, there is increasing interest behind the automation of construction machinery, particularly through the use of cartesian control systems for crane boom tips. This approach aims to streamline machine operation and make it more user-friendly, having the operator focusing on tip movement and delegating the management of joints to the crane controller. The concept of boom tip control first emerged in 2013, with the introduction of John Deere Forestry’s Intelligent Boom Control (IBC) system [2]. Research has demonstrated that this form of control can significantly reduce crane operating times [3]. Comparable innovations have also been explored for other hydraulic equipment, such as excavators [4]. Another trend gaining increasing attention in the field is operator remotization to enhance safety and improve the working conditions of operators. Examples include Caterpillar’s Command remote station [5] and the teleoperation cabin developed for excavators at ETH Zurich [6]. These systems typically involve relocating the operator to †All the authors are part of the JOiiNT LAB project. 1Soft robotics for Human Cooperation and Rehabilitation, Istituto Italiano di Tecnologia, via Morego, 30, 16163 Genova, Italia. 2Dipartimento di Ingegneria Gestionale, dell’Informazione e della Produzione, Universit` a di Bergamo, Bergamo, Italia. 3Fassi Innovation Center, Via Tobia Ferrari, 1, 24027 Nembro BG, Italy. Fig. 1: The heavy-duty crane F1150, thanks to the cartesian control becomes the largest robot in the world (peak torque of 1000 kNm, reach 32 m, 4 Dof). a safer distance from the machine, using traditional control interfaces such as joysticks and pedals. However, these solutions largely retain the traditional control paradigm, preserving both the operational complexity of the machine and the cognitive demands placed on the operator. Advances in teleoperation and avatar-based robotics have made remote control systems more intuitive and immersive, as demonstrated in works such as [7]. This paper presents an architecture to transform a typical crane in a smarter construction machine (Fig. 2) by leveraging user-friendly interfaces and cartesian control strategies, that we called Gigabot. This is the last milestone of a series of activities started in 2021 within the JOiiNT LAB project, where we explored the potential of robot avatars innovations in construction sites demonstrating the feasibility of remote crane operation using a compact M20 crane at Bauma 2023 [8]. Thanks to the positive feedback from end-users and cartesian control validation and the increasing interest of Fassi to industrialize the concept of modularity and scalability on different models, we developed an industrial system for remote crane control. The system was implemented on a heavy duty Fassi crane model (F1150, Fig. 1) and tested under real-world conditions at the Bauma 2025 construction exhibition, where the F1150 was successfully operated remotely from Munich over a distance of 350 km via an internet connection by experienced crane operators and first-time users. This result was first step for Fassi to initiate the industrialization of the concept. This underlines the potential for collaboration between industry and research institutes, where industrial needs can meet research expertise to develop innovative solutions, applying new approaches to real-world problems. 2025 I-RIM Conference October 17-19, Rome, Italy ISBN: 9788894580570 10.5281/zenodo.17629760 131 Fig. 2: The Gigabot architecture (I), stages of live tests with the Gigabot (II), layout of target symbols on the floor (III). II. INDUSTRIAL GIGABOT ARCHITECTURE The proposed architecture for the Gigabot consists of: •Fassi F1150 Crane: A heavy-duty loader crane featuring a maximum reach of 31.8 meters and a lifting moment of 1000 kNm, referring to the base (system origin). It provides 4 degrees of freedom via hydraulic actuators powered by a pump that consumes 78 kW at 1500 rpm and 350 bar under nominal operating conditions (Fig. 1). •JOiiNT LAB Boom Tip Control: Model-based Cartesian velocity tracking using inverse kinematics, feedforward/feedback control, and servo-valve compensation [8]. •Industrial Network: Stable, low latency (20-30 ms) and private (under VPN) communication between the crane and the remote operator, enabled by satellite internet connection. This choice allows having in every construction site (usually without wired connections and internet coverage) a dedicated industrial grade connection to perform the crane remote control. •Industrial Human-Robot Interface (HRI): Composed by a radio controller with joysticks to control the crane tip, a Virtual Reality headset (VR) to perceive the environment by multipoint of view such as the crane tip (via industrial camera) and by third person view (via a camera on a robotic neck) and a monitor to control real time data of the network connection (average and actual data for latency, bandwidth and packet loss). •Computational Unit: Its main specifications to run all the software needed for the Gigabot system are an Intel core i7-11800H, 16 GB RAM, NVIDIA GeForce RTX 3080 and Ubuntu 20.04. III. PRELIMINARY RESULTS The remote Gigabot system, in Bauma 2025, was active 10 days, 8 hours per day, with users taking turns to operate the crane remotely. Over the fair, around 50 individuals, including both experienced crane operators and first-time users, used the system. Operators, after a short explanation about the radio controller configuration had the objective to follow a path created on the ground, invoking linear and diagonal maneuvers (which are complex for traditional control) of the crane tip over a large area (about 6x4 m) (see Fig. 2). The latency of the network connection was on average 20-30 ms, with a bandwidth of approximately 2 Mb/s. The system over the time that was used proved to be robust and reliable, with no major issues or crashes. Anyway, to handle communication losses a safety feature was implemented that: 1) deactivates the remote control, 2) holds the current state of the crane, and 3) displays a warning message to the operator interface. We were able to maintain the crane control stable (control stability below 4 ms), manage the pilot commands at 100 hz and send the camera streaming with an average of 30 fps, with peaks of 60 fps. This confirms the effectiveness of the chosen industrial components and the overall system architecture, making possible to operate the crane remotely in a real-world scenario. IV. CONCLUSION We presented an industrial system architecture designed to transform a conventional crane into an intelligent, user-friendly system with robust remote control capabilities. These features were really appreciated by crane operators and Fassi is interested to further develop the concept. In fact, it has requested the license to JOiiNT LAB to industrialize the Gigabot system transforming it in a real product based on the results obtained in Bauma 2025. Gigabot demonstrates significant potential for broader adoption in the construction industry, offering improvements in safety, operational efficiency, and the emergence of new business opportunities. Looking ahead, future developments will aim to integrate enhanced sensory feedback and shared autonomy, further boosting operator awareness and overall system performance. ACKNOWLEDGMENTS This activity was developed within the JOiiNT LAB in collaboration with Fassi Gru and Phoenix Informatica. REFERENCES [1] B. Brucker Juricic et al. Review of the construction labour demand and shortages in the eu. Buildings, 11(1), 2021. [2] A. Sokolov et al. Forest machine automation and ergonomics. In E3S Web of Conferences, volume 389, page 03002. EDP Sciences, 2023. [3] J. Manner et al. Comparing forwarder boom-control systems based on an automatically recorded follow-up dataset. Silva Fennica, 53:15, 11 2019. [4] M. Hutter et al. Force control for active chassis balancing. IEEE/ASME Transactions on Mechatronics, 22(2):613–622, 2017. [5] J. Rubenstone. Remote operation coming to more caterpillar equipment, 2021. [6] M. Hutter et al. Ibex — a tele-operation and training device for walking excavators. pages 48–53, 10 2016. [7] S. Behnke et al. The 10 million ana avatar xprize competition advanced immersive telepresence systems. 2023. [8] A. Duz et al. From construction machines to remote construction robots: control, interfaces, and usability of the cranebot. Frontiers in Robotics and AI, Volume 11 - 2024, 2025. 132