Summary of the Dissertation

Automated driving features, which are designed to relieve the driver by taking over driving tasks, are becoming increasingly common in passenger cars.

However, significant regional variations in factors affecting the road network can be observed, which also impact the functional performance as experienced by the user.

Against this backdrop, this dissertation presents a novel validation approach based on a framework that combines the concepts of a virtual test drive with a model-in-the-loop simulation, using representative routes as input. In this way, validation statements representative of the entire user base can be made, and, for example, the Operation Design Domain (ODD) of predictive functions can be better quantified.