![]() |
Charlotte Stonestreet
Managing Editor |
| Home> | IIot & Smart Technology | >AI | >Partnership launched to bring physical AI to critcal energy infratructure |
| Home> | AUTOMATION | >Robots | >Partnership launched to bring physical AI to critcal energy infratructure |
Partnership launched to bring physical AI to critcal energy infratructure
16 November 2026
NEURA Robotics, E.ON and enviaM are partnering to develop Physical AI for critical energy infrastructure and progressively deploy it in practice.

Using Neuraverse, NEURA’s open Physical AI platform, the partners aim to develop practical applications for grid operations, from inspecting and monitoring technical assets to supporting potentially hazardous, physically demanding or repetitive tasks. Relevant robotic skills will first be trained and tested at TUM RoboGym powered by NEURA, one of the world’s largest scientific robotics training centres. MITNETZ STROM, the pilot partner and the enviaM Group’s grid subsidiary, will then test initial applications in real-world grid operations.
To jointly develop robotic skills for grid operations, E.ON is joining TUM RoboGym powered by NEURA, which NEURA Robotics and the Technical University of Munich (TUM) opened at LabCampus at Munich Airport in October 2026. Covering more than 2,300 square metres, the centre brings together research, robot training and industrial applications. Companies from more than ten industries are already training robots to perform specific workflows there; E.ON now brings the energy sector into the mix. The initial focus will be on selected tasks such as inspections, monitoring critical assets including substations, and supporting site security. Tasks in potentially hazardous environments will also be assessed as possible applications. The skills will first be developed under controlled conditions before being tested with MITNETZ STROM in real-world grid operations.
At the outset of the partnership, interdisciplinary teams from E.ON, MITNETZ STROM and NEURA will identify specific use cases in grid operations. Skilled workers will contribute their experience and requirements, while NEURA will work with them to develop the necessary robotic skills and provide the training and software infrastructure through the NEURA Gym and Neuraverse. Insights from the tests will inform the further development of both the skills and the platform. The partners are thus creating an approach that goes beyond a one-off demonstration and is intended to enable repeatable, scalable Physical AI applications in energy infrastructure over the long term.
How does a robot learn to inspect an asset that, until now, only experienced specialists have been familiar with? The NEURA Gym combines three approaches to gaining physical experience:
- Human demonstrations: Grid operations specialists demonstrate their workflows. Sensor-equipped suits, motion capture systems and teleoperation make their practical expertise available for training.
- Real-world robot data: Robots perform the tasks, while their sensors record movements, physical contact and the outcomes of their actions.
- Simulation and Data Amplifier: Virtual training environments build on real-world experience with additional variations and scenarios. The skills are then validated on real robots.
Robots learn with Neuron AI, NEURA’s multimodal foundation model for Physical AI, which combines 3D vision, touch, audio, robot state and language, as well as memory. This can make a difference in inspection tasks: a robot captures not only what it sees, but also what it hears and feels. Grid facilities are built for people, with stairs, doors, control cabinets and tools. In future, humanoid robots such as NEURA’s 4NE1 will be able to navigate these environments without requiring modifications to the facilities. This will allow them to access areas previously out of reach for conventional machines.
Neuraverse, NEURA’s open Physical AI platform, provides the central link between training and subsequent deployment. It connects robots, AI models, developers and industry partners, enabling trained skills to be managed, deployed and progressively refined. The aim is for an individual application trained in the NEURA Gym to become a reusable skill for compatible robots and additional deployment sites. E.ON and enviaM always retain control of their own data and trained skills and decide for themselves whether, and with whom, to share them.
- CAD to CAM digital continuity
- Fast Payback In High Power Applications
- McVitie's owner pumps £68m into British operations
- Managed IoT connectivity available in AWS Marketplace
- Get the key to code quality at PPMA
- The digital revolution in vehicle manufacturing
- Raspberry Pi CM4-based industrial controller
- Aerospace - from virtual to reality
- Are you up to date with the Data Act?
- How machine learning is transforming predictive maintenance
- Balls to Gravity
- ROBOTS GIVE FASTER CELL TOOL CHANGE TIMES
- ROBOT FOR DELICATE ASSEMBLY
- NUMBER OF ROBOTS IN UK ON THE RISE
- The future is bright - The future is automation!
- High Demand For Grant
- More Capacity, Less Space
- FREEBIRD - THE THIRD DIMENSION
- PRECISION ROBOT CELL FOR THE PHARMACEUTICAL INDUSTRY
- ROBOT SALES REACH ALL TIME HIGH

















