Hexagon Unveils AEON: AI-Powered Humanoid Robot to Address Global Labor Shortage in Manufacturing and Logistics

Global labor shortages have left millions of job vacancies unfilled, particularly in manufacturing and logistics. To address this issue, Hexagon, a leader in measurement technologies, is developing AEON, a new humanoid robot designed to handle various industrial tasks. The unveiling of AEON took place at the Hexagon LIVE Global conference, highlighting its capabilities in manipulation, asset inspection, reality capture, and operator support. Hexagon plans to deploy AEON across multiple sectors, including automotive, transportation, aerospace, manufacturing, warehousing, and logistics. Deepu Talla, NVIDIA’s vice president of robotics and edge AI, emphasized that advancements in simulation and physical AI have ushered in the era of general-purpose robotics. AEON exemplifies this trend by integrating NVIDIA's three-computer robotics platform. This includes AI supercomputers for training and fine-tuning models, the NVIDIA Omniverse platform for simulation and optimization, and NVIDIA IGX Thor for real-time model execution and functional safety. The development of AEON leveraged NVIDIA’s comprehensive stack, enabling rapid skill acquisition. NVIDIA Isaac Sim, a key component of the Omniverse platform, was used to simulate complex robot actions such as navigation, locomotion, and manipulation. These skills were further refined using reinforcement learning in NVIDIA Isaac Lab, an open-source framework. The simulation-first approach allowed AEON to learn core locomotion skills in just 2-3 weeks, compared to the usual 5-6 months required for real-world deployment. Hexagon’s President of the Robotics Division, Arnaud Robert, stated that the goal with AEON was to create an autonomous humanoid that can effectively tackle industrial challenges. By harnessing NVIDIA’s platforms, they achieved a robot that combines advanced mechatronics, multimodal sensor fusion, and real-time AI. AEON will initially be piloted in factories and warehouses to scan and capture data from components, assembly lines, and storage areas. The captured data will be integrated into Hexagon’s Reality Capture Suite (RCS), a platform that enables collaboration, visualization, and sharing of reality-capture data. RCS utilizes Hexagon Data River (HxDR) and NVIDIA Omniverse running in the cloud, breaking down the limitations imposed by local infrastructure. Lucas Heinzle, vice president of R&D at Hexagon’s robotics division, noted that while digital twins offer significant benefits, their adoption has been slow in some industries. AEON’s advanced sensor suite helps bridge this gap by seamlessly integrating reality data capture with NVIDIA Omniverse, thus streamlining workflows and promoting the widespread use of digital twins. Hexagon’s use of the OpenUSD framework and development on Omniverse ensures that high-fidelity digital twins can be generated from scanned data, creating a continuous data flywheel for ongoing training and improvement of AEON. This collaborative effort with NVIDIA is pushing the boundaries of physical AI, providing scalable and efficient solutions for industries reliant on accurate real-world data. Industry experts have praised the partnership between Hexagon and NVIDIA, noting that it exemplifies the future of robotics in industry. The integration of advanced simulation and AI technologies not only accelerates robot development but also enhances their autonomy and capabilities. Hexagon's expertise in measurement technologies, combined with NVIDIA's robust AI and simulation platforms, positions AEON as a groundbreaking technology poised to revolutionize industrial operations. Companies in these sectors can expect significant improvements in efficiency, accuracy, and safety with the deployment of AEON.