Environment and Context

For automated vehicles, sensors such as camera systems, LiDAR or radar, as well as special algorithms, act as the eyes and brain of the machine. This sub-project aims to determine how these sensors can be targeted to reliably capture and interpret the most challenging urban scenarios – even under difficult conditions. Artificial intelligence and machine learning play a key role here. These methods can be used to, for example, reliably identify and track road users in confusing traffic situations, such as intersections with limited visibility or in double-parking situations.

Main Topics

 Perception and environment

  • Requirements for sensor technology, adaptive sensors; system robustness; Concealed areas
  • Object modelling; AI-based methods for perception; multimodal sensor setup


  • Novel, scalable fusion concepts; increased robustness; AI-based fusion;
  • Increase in fusion accuracy; object standardisation; comparison of different fusion concepts


  • Innovative concepts and technologies for localisation; online HD maps;
  • Basic models, 3D models and scenarios