Informed Reinforcement Learning for Situation-Aware Traffic Rule Exceptions
This work addresses the need for interpretable and situation-aware decision-making in autonomous driving, representing an incremental improvement over existing methods.
The paper tackles the problem of autonomous driving in complex scenarios by integrating a structured rulebook into reinforcement learning, resulting in high completion rates for complex scenarios with model-based agents.
Reinforcement Learning is a highly active research field with promising advancements. In the field of autonomous driving, however, often very simple scenarios are being examined. Common approaches use non-interpretable control commands as the action space and unstructured reward designs which lack structure. In this work, we introduce Informed Reinforcement Learning, where a structured rulebook is integrated as a knowledge source. We learn trajectories and asses them with a situation-aware reward design, leading to a dynamic reward which allows the agent to learn situations which require controlled traffic rule exceptions. Our method is applicable to arbitrary RL models. We successfully demonstrate high completion rates of complex scenarios with recent model-based agents.