BMW and TUM develop AI sensor fusion for traffic safety

Researchers from the Technical University of Munich (TUM) and the BMW Group have developed an artificial intelligence system designed to improve traffic safety through advanced sensor fusion. The system integrates data from various sources, including cameras, radar, and lidar mounted on a test vehicle, with measurements from infrastructure sensors located on sign bridges at intersections.

By combining these different perspectives, the AI creates a digital real-time map of the traffic situation that updates every 100 milliseconds. This digital twin can track cars, bicycles, and pedestrians with a positional accuracy of up to ten centimeters. Crucially, the system can predict the movement of these road users, allowing the vehicle to see into obscured areas—such as when a pedestrian emerges from behind a parked car or a cyclist approaches from a blind corner.

Professor Alois Knoll, head of the Chair of Robotics, Artificial Intelligence and Real-Time Systems at TUM, noted that the technology allows drivers to react faster to potential hazards by providing visibility into previously hidden zones.