BMW Group is expanding its physical AI work by making Plant Landshut a key site for developing software for AI-supported robotics in component production.

The German plant will focus on open software platforms, training data, simulation, motion planning, and robotic system training, complementing humanoid robotics pilots at BMW plants in Leipzig and Spartanburg.

Plant Landshut is leveraging its role as a centre of expertise for in-house component production

Its broad component production operations give BMW a real-world setting to test future robotics use cases, especially tasks requiring flexibility, fine motor skills and awareness of the working environment.

BMW said the software will help robots perceive their surroundings, assess situations, and determine suitable actions independently, with potential applications including component recognition, movement planning, gripping, and parts handling.

"At BMW Group Plant Landshut, we are combining software expertise with industrial practice in our component production", said Dr Wolfgang Bluemlhuber, head of Driving Dynamics and In-house Component Manufacturing.

At Plant Landshut, the project team captures data from real production environments and test setups, and integrates it into robot simulations

BMW is combining conventional programming with Vision-Language-Action AI models and demonstration-based learning, using production data, simulations, camera systems, motion capture suits, and data gloves to train robotic behaviour before live deployment.

At Landshut, BMW is working with Athenyx Robotics on an "intelligence stack" integrating perception, learning, simulation and decision-making, while also collaborating with the University of Applied Sciences Landshut on data generation, simulation, and use-case testing.