Alexander Katriniok

The automation of road vehicles bares significant potential to enhance efficiency and road safety. Moreover, it supports the development of novel automotive applications like autonomous ride services or autonomous good delivery – with societal, economic and ecological impact. With an increasing level of automation, automated vehicles (AVs) require advanced decision making and motion planning (short: planning) algorithms as well as control methods to operate autonomously in challenging traffic scenarios. Especially those traffic scenarios, which involve interactions between AVs and human driven vehicles and/or even vulnerable road users (like pedestrians or cyclists), challenge these autonomous systems by means of uncertainty related to other road users’ actions and reactions. Complementing AVs with communication capabilities offers considerable potential for further performance improvements. Through information exchange, uncertainty about the behavior of other connected road users can significantly be reduced and AVs may even operate in a cooperative fashion.

To holistically address these challenges, my research focuses on real-time optimization-based control, control with rigorous safety guarantees, distributed optimization-based control and learning-based control, with application to motion planning, decision making and control of autonomous vehicles in uncertain environments. Besides these planning and control problems on vehicle-level, my research also investigates the interplay between vehicle-level control and control on transportation system level to fertilize efficient and sustainable mobility.