EnergyBrief
NEURA, E.ON and enviaM want to teach robots to inspect the power grid
NEURA Robotics, a robot maker based in Metzingen, and the energy companies E.ON and enviaM want to teach robots to inspect and monitor electrical installations, taking arduous or dangerous tasks off technicians. First results from trials on the live grid are expected in 2027.
The three companies’ joint press release, dated 5 November 2026 in Metzingen, announces a partnership to develop “physical AIAI systems that perceive, understand and act in the physical world: robots, autonomous vehicles, smart cameras.Read the glossary entry”, meaning artificial intelligence that controls robots, for critical energy infrastructure, and to put it into practice in stages.
The first tasks targeted are inspections, monitoring of critical installations, including substations, and help with protecting sites.
The robots’ skills will first be trained at the TUM RoboGym, a centre opened in October 2026 by NEURA Robotics and the Technical University of Munich near Munich airport, which E.ON is joining.
They will then be tested on the live grid by MITNETZ STROM, the grid operator of the enviaM group and a pilot partner. According to Stephan Lowis, chairman of the enviaM group’s management board, the first feedback is expected during 2027.
The press release notes that these installations are built for people, with stairs, doors and electrical cabinets, and argues that humanoid robots such as NEURA’s 4NE1 will in future be able to move around them without any modifications.
Update of 9 November: according to Humanoids Daily, the video accompanying the announcement shows robot demonstrations at NEURA’s premises, not work in a substation or at another site on the power grid. The partners have not disclosed the number of robots, the deployment sites or the performance thresholds expected for the pilot trials, the site says. The next step is to choose concrete use cases. According to the announcement, E.ON and enviaM keep control of their data and of the skills the robots learn, including the decision whether or not to share them.
Sources
- NEURA Robotics (blog, German) · neura-robotics.com
- Humanoids Daily · humanoidsdaily.com
Also worth reading
Illustration, not a photo of the robot