Modeling Interactions of Autonomous Vehicles and Pedestrians with Deep Multi-Agent Reinforcement Learning for Collision Avoidance

Reliable pedestrian crash avoidance mitigation (PCAM) systems are crucial components of safe autonomous vehicles (AVs). The nature of the vehicle-pedestrian interaction where decisions of one agent directly affect the other agent's optimal behavior, and vice versa, is a challenging yet often neglected aspect of such systems. We address this issue by defining a Markov decision process (MDP) for a simulated driving scenario in which an AV driving along an urban street faces a pedestrian trying to cross an unmarked crosswalk. The AV's PCAM decision policy is learned through deep reinforcement learning (DRL). Since modeling pedestrians realistically is challenging, we compare two levels of intelligent pedestrian behavior. While the baseline model follows a predefined strategy, our advanced pedestrian model is defined as a second DRL agent. This model captures continuous learning and the uncertainty inherent in human behavior, making the vehicle-pedestrian interaction a deep multi-agent reinforcement learning (DMARL) problem. We benchmark the developed PCAM systems according to the agents' collision rate and the resulting traffic flow efficiency with a focus on the influence of observation uncertainty or noise on the decision-making of the agents. The results show that the AV is able to completely mitigate collisions under the majority of the investigated conditions, and that the DRL pedestrian model indeed learns a more intelligent crossing behavior.