NEURA, E.ON and enviaM will train robots to inspect power grids
NEURA Robotics will train grid-inspection skills with E.ON and enviaM at its Munich RoboGym, with first field tests at MITNETZ STROM expected in 2027.
NEURA Robotics is teaming up with energy companies E.ON and enviaM to teach robots to work on power grids. The three partners announced the project on Wednesday, saying they will develop physical AI for inspecting and monitoring electricity infrastructure and for supporting workers on hazardous, physically demanding or repetitive jobs.
The skills will be trained first at TUM RoboGym powered by NEURA, the robotics training centre NEURA opened with the Technical University of Munich at Munich Airport in October. The centre covers more than 2,300 square metres, and companies from more than ten industries are already training robots there. E.ON now brings the energy sector into the mix. The pilot partner is MITNETZ STROM, enviaM's grid subsidiary and one of the largest regional distribution system operators in eastern Germany, with a network of about 71,000 kilometres. It expects to gain its first practical experience with the technology during 2027.
The first jobs on the list are inspections, monitoring of critical assets such as substations, and site security. Work in potentially hazardous environments will also be assessed. The training combines three ingredients: grid specialists demonstrating their workflows in sensor suits, with motion capture and teleoperation; data from robots performing the tasks themselves; and simulation to multiply the real-world experience. The robots learn with Neuron AI, NEURA's multimodal foundation model, which combines 3D vision, touch, audio, robot state, language and memory.
The humanoid angle is practical rather than decorative. Grid facilities were built for people — stairs, doors, control cabinets, tools — so a humanoid such as NEURA's 4NE1 could eventually move through them without the facilities being rebuilt. NEURA's Neuraverse platform is meant to turn each trained task into a reusable skill for compatible robots at other sites. E.ON and enviaM keep control of their own operational data and trained skills, and decide whether to share them.
For now, this is a plan, not a deployment. The partners disclosed no financial terms, no robot numbers and no deployment dates, and nothing is yet running in the grid. The scale of the prize is considerable: E.ON operates a distribution network of 1.6 million kilometres and serves around 47 million customers with about 78,000 employees, while the enviaM Group supplies around 1.1 million customers in eastern Germany. If the training pipeline works, the same approach could extend to other infrastructure where skilled workers are in short supply.