Using Collision Momentum in Deep Reinforcement Learning Based Adversarial Pedestrian Modeling
This work addresses the need for specialized pedestrian simulation algorithms to test and improve autonomous driving systems, though it appears incremental as it builds on existing reinforcement learning approaches.
The paper tackles the problem of generating pedestrian behaviors that reveal weaknesses in automated vehicle controllers, particularly in extreme scenarios, by proposing a reinforcement learning algorithm that specifically targets collisions and produces more severe collisions to identify failure modes.
Recent research in pedestrian simulation often aims to develop realistic behaviors in various situations, but it is challenging for existing algorithms to generate behaviors that identify weaknesses in automated vehicles' performance in extreme and unlikely scenarios and edge cases. To address this, specialized pedestrian behavior algorithms are needed. Current research focuses on realistic trajectories using social force models and reinforcement learning based models. However, we propose a reinforcement learning algorithm that specifically targets collisions and better uncovers unique failure modes of automated vehicle controllers. Our algorithm is efficient and generates more severe collisions, allowing for the identification and correction of weaknesses in autonomous driving algorithms in complex and varied scenarios.