Understanding which traffic light controls which lane is crucial to navigate intersections safely. Autonomous vehicles commonly rely on High Definition (HD) maps that contain information about the assignment of lanes and traffic lights. The manual provisioning of this information is tedious, expensive and not scalable. To remedy these issues, the proposed novel approach derives the assignments automatically from motion data. Effectiveness and safety are addressed and the dataset is made publicly available.
The paper results from a collaboration between Thomas Monninger, Senior Machine Learning Engineer at Mercedes-Benz Research & Development, Sunnyvale, California, Andreas Silvius Weber, Development Engineer, Autonomous Driving Maps & Map Learning at Mercedes-Benz AG, Germany and Prof. Dr. Steffen Staab.
Thomas Monninger, Andreas Silvius Weber, Steffen Staab. Semantic Map Learning of Traffic Light to Lane Assignment based on Motion Data. In: 26th IEEE International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain, September 24-28, 2023.