Highly Realistic Virtual Testing of Automated Driving Systems as an Opportunity to Accelerate the Mobility Revolution

Authors

R. Degen, Cologne University of Applied Sciences

Summary

Automated driving is seen as a great source of hope for solving modern mobility problems. This includes effects such as increasing traffic safety in areas with high traffic density, reducing parking space and traffic volume in the city, and increasing resource efficiency through intelligent route prediction.

This paper presents an approach for virtual testing and verification of automated driving functions that interactively connects urban traffic participants in a virtual reality (VR) environment. The basis for this is a georealistic VR model of downtown Cologne, which reactively links to the vehicle periphery of a highly automated vehicle as well as a real, humanoid road user. A specific measurement technique captures both the vehicle and the person in a driving-dynamically and biomechanically correct way, respectively, and mirrors their behavior into the virtual environment. The use of selected sensor models ensures the interaction between the protagonists.

This approach enables the safe and valid testing of critical traffic scenarios in everyday urban mobility. In parallel, this results in a sustainable collection of data that can be fed back into the development process of driving functions without having to perform count-less real testkilometers.