I'm interested in the general understanding of information processing principles and models of the human brain and their use for artificial intelligence as well as representations of knowledge in machine learning models.
My current research project is about self-driving vehicles with a focus on the prediction of surrounding traffic participants' behavior. For that, a model learns an appropriate semantic representation of the surrounding environment as well as the participants' trajectories in order to identify causal factors influencing their behavior. The prediction thus exceeds physics-based trajectory prediction approaches and captures typical driving behavior patterns as well as mutual interactions between participants. The ultimate goal is to ensure safe and robust automated driving in arbitrary mixed traffic scenarios.