Modeling and Analysis of Automated Lane Keeping Systems

Automated Lane Keeping Systems (ALKS) are systems that autonomously drive cars in certain situations. For approval, such systems have to fulfill certain standardized requirements. The standardization document describes several scenarios, that an ALKS must be able to handle at least as well as a typical human driver. Those scenarios depend on parameters such as car speed, vehicle size, reaction time, etc. For the approval process, the scenarios are currently simulated for thousands of different combinations of parameter values, which is costly.

The goal of the thesis will be to, at least partially, replace such simulations by an analysis of the global behavior of the scenarios. More concretely, the student will

1. create a completely formal mathematical model of the scenarios, and
2. apply existing symbolic or semi-symbolic algorithms for deducing information about the global parameter-dependent behavior of this mathematical model.

This thesis topic forms a project based on collaboration with the company TÜV Süd.