Driver models play an important role for simulation and evaluation of planning algorithms for automated vehicles. We leverage driver models to accelerate learning a policy with Deep Reinforcement Learning by using principles from physics-informed deep learning. This allows to regularize the policy with a driver model and results in a better generalization to unseen driving situations and a higher sample-efficiency.
Most existing driver models cover the case of car following or lane changing. To also use physics-informed reinforcement learning in more interactive situations like merging, we also develop novel driver models that are able to approach traffic gaps.
Contact: M.Sc. Johannes Fischer