Automatic Intersection Management in Mixed Traffic Using Reinforcement Learning and Graph Neural Networks
This addresses urban traffic efficiency for mixed traffic scenarios, representing an incremental extension of prior work on fully automated traffic.
The paper tackles the problem of automatic intersection management in mixed traffic with both automated and human-driven vehicles, proposing a reinforcement learning and graph neural network approach that significantly increases vehicle throughput and reduces delays as the share of automated vehicles rises, with non-automated vehicles benefiting similarly.
Connected automated driving has the potential to significantly improve urban traffic efficiency, e.g., by alleviating issues due to occlusion. Cooperative behavior planning can be employed to jointly optimize the motion of multiple vehicles. Most existing approaches to automatic intersection management, however, only consider fully automated traffic. In practice, mixed traffic, i.e., the simultaneous road usage by automated and human-driven vehicles, will be prevalent. The present work proposes to leverage reinforcement learning and a graph-based scene representation for cooperative multi-agent planning. We build upon our previous works that showed the applicability of such machine learning methods to fully automated traffic. The scene representation is extended for mixed traffic and considers uncertainty in the human drivers' intentions. In the simulation-based evaluation, we model measurement uncertainties through noise processes that are tuned using real-world data. The paper evaluates the proposed method against an enhanced first in - first out scheme, our baseline for mixed traffic management. With increasing share of automated vehicles, the learned planner significantly increases the vehicle throughput and reduces the delay due to interaction. Non-automated vehicles benefit virtually alike.