Digital Twin of CERN LHC for Intervention Simulation

Digital Twin of CERN LHC for Intervention Simulation Supervisor: Antoine Ansel Mentor: Pablo García Ledo Jorge Antonio Velásquez Belmonte Internship Report by: Saumy Sharma CERN Openlab Summer Student Final year Information Technology, SOS E&T GGV Acknowledgement Having long dreamt of contributing to a project of global significance, I am profoundly grateful for the opportunity to work on the Digital Twin of CERN's LHC for Intervention Simulation. This experience has been a dream come true, offering me the rare privilege of applying my skills to one of the most sophisticated research infrastructures in the world. The blend of innovation, collaboration, and intellectual challenge has truly enriched my professional journey. I extend my deepest gratitude to my supervisor, Antione, and my mentors, Pablo and Jorge, whose unwavering support and expert guidance have been the cornerstones of my success. Their mentorship transcended mere technical instruction, embodying a holistic approach to professional growth. They imparted not only the intricate details of the project but also instilled in me the critical importance of adhering to best coding practices and maintaining a disciplined, analytical mindset. Antione, Pablo and Jorge consistently demonstrated approachability and patience, offering their expertise regardless of the complexity of the challenges I encountered. Their constructive feedback and encouragement were pivotal in helping me navigate obstacles and refine my problem-solving abilities. I was always confident that any doubts or difficulties would be met with insightful and practical solutions. This experience has been immensely rewarding, and I will always cherish the knowledge and skills I gained under their supervision. I am thankful for the opportunity to learn from such talented and supportive supervisors and mentors, and I will carry the lessons they taught me throughout my career. 1 1. Introduction The project starts with the EN-IM-PLM team considering using a new technology in the ongoing development of a Digital Twin: Nvidia Omniverse. In this project, I develop functionality using Nvidia Omniverse as a way to gauge the usefulness, quality and ease of development of this technology. In particular, the project consists of features to aid in the planning of interventions. 1.1 Digital Twins “Digital twins” blur the line between the physical and the virtual. More than simulations, these virtual copies of reality incorporate real data to reshape the way we design and build, allowing us to predict real-world outcomes. A recent workshop at CERN explored the complexities of this technology and showcased its applications. At CERN, digital twins could play a key role in high-energy physics experiments: from detector development, to designing strategies for data analytics and comparing results with theoretical models. They can be used not only to analyze and improve CERN’s infrastructure, but also to find unidentified anomalies before they arise in real life. 1.2 Nvidia Omniverse Nvidia Omniverse is a computing platform built to enhance digital design and development by integrating 3D design, spatial computing and physics-based workflows across Nvidia tools, third-party apps and artificial intelligence (AI) services. It is used for building digital twins of products, factories, warehouses and infrastructure. It can also streamline the creation of 3D-related media for entertainment and product demonstrations, as well as enterprise media content rendered on computers, phones and extended reality (XR) devices. The platform, launched in 2022, is available as a cloud service or a private instance running on premises. Additionally, it supports plugins and integrations for deploying omniverse content, applications and autonomous control systems across cars, robots, building controls, equipment and medical devices. 1.3 USD Framework Omniverse uses the Universal Scene Description framework (USD). USD is an open source framework used for interchange of 3D computer graphics data which focuses on collaboration, non-destructive editing. USD is a high-performance extensible software platform for collaboratively constructing animated 3D scenes, designed to meet the needs of large-scale film and visual effects production. USD enables robust interchange between digital content creation tools with its expanding set of schemas, covering domains like geometry, shading, lighting, and physics. USD’s unique composition ability provides rich and varied ways to combine assets into larger assemblies, enables collaborative workflows so that many creators can work together with ease, and more. 1.4 Intervention Planning When changes are required in installations at CERN, from installing a new generation of magnets in the LHC to replacing power converters, there is a process of planning of the transportation and installation of the equipment. Errors in this planning can lead to expensive delays or work-arounds if, for example, the equipment to be installed doesn’t fit through a corridor or if two teams are scheduled to work on the same place at the same time. 3 Internship Experience 1.5 Learning Omniverse Fig: Catia model of CMS in USD format I dedicated a significant amount of time to learning to use Omniverse during my internship. Under the guidance of my supervisors, my primary responsibility was to enable interventions and simulations within Omniverse using the 3D model of the CMS cavern. I documented my workflow and everything I learned from the NVIDIA Omniverse documentation. In addition to explaining several topics based on my understanding, I also provided a detailed, step-by-step procedure for various processes. 1.6 Movement Animation During my internship, I developed a Cone Navigator Extension for NVIDIA Omniverse. The extension allows users to calculate and visualize the shortest path between two cones in a 3D environment and animates one cone along this path. It also includes an option to spawn spheres along the path for visual reference. Key Features: ●Pathfinding: Calculates the shortest path between two cones using a navigation mesh. ●Animation: Animates one cone along the calculated path using the timeline. ●Sphere Spawning: Optionally spawns spheres along the path to highlight key points. ●User Interface: A simple UI allows users to control pathfinding, animation, and sphere spawning. 5 Challenges: ●Navigation Mesh Configuration: Ensuring accurate pathfinding within the environment. ●Animation Timing: Coordinating smooth movement along the path. Outcomes: Working on these features enhanced my understanding of Omniverse's capabilities in pathfinding, animation, and UI design. I also gained valuable experience in writing clean, efficient code. 1.7 Lift Team Extension The user needs to know how many people are needed to move equipment. The primary goal of this extension is to present a user-friendly interface where users can input the weight of an item, and the system will output the number of people required to lift it safely. The extension assumes a maximum safe weight limit that one person can handle, which is set at 25 kg. The extension is designed to be easy to use and integrates seamlessly with Omniverse’s UI framework. The core logic of the extension is encapsulated in the calculate_people_required function. This function takes the weight of the item as input and calculates the number of people required to lift the item based on the maximum weight each person can handle. ●Single Person Handling: If the weight is less than or equal to 25 kg, the function returns that only one person is needed. ●Multiple Person Calculation: If the weight exceeds 25 kg, the function calculates the total number of people required using ceiling division to ensure that the item is safely lifted without exceeding the weight limit for any individual. 7 Challenges ●Handling Invalid Input: A common challenge was ensuring that the user input was valid. This was addressed by using a try-except block to catch and handle ValueError exceptions if the input could not be converted to a float. ●User Interface Design: Designing an intuitive and straightforward UI that could be easily understood by users. The use of Omniverse's omni.ui components facilitated this by providing pre-built, customizable elements. Outcome: The "Lift Team Calculator" extension successfully provides a tool for calculating the number of people required to lift an item based on its weight. It offers a straightforward interface and reliable functionality that can be applied in various simulation scenarios within the Omniverse platform. 1.8 VR Setup in Omniverse During my internship, I explored the potential of using Virtual Reality (VR) within NVIDIA Omniverse. Specifically, I documented the process of setting up VR using a Meta Quest 1 headset. I found out that Graphics quality of Omniverse VR scenes are better than Unity’s VR rendering and it is easy to set up but it does require SteamVR.