Resolve Highway Conflict in Multi-Autonomous Vehicle Controls with Local State Attention
This work addresses conflict resolution for autonomous vehicles in highway merging scenarios, but it is incremental as it builds on existing multi-agent reinforcement learning methods with a novel attention module.
The paper tackled the problem of resolving local conflicts between autonomous vehicles in mixed-traffic environments by proposing a Local State Attention module to enhance state representation in multi-agent reinforcement learning. The results showed significant improvements in merging efficiency, particularly in high-density traffic settings, compared to existing baselines.
In mixed-traffic environments, autonomous vehicles must adapt to human-controlled vehicles and other unusual driving situations. This setting can be framed as a multi-agent reinforcement learning (MARL) environment with full cooperative reward among the autonomous vehicles. While methods such as Multi-agent Proximal Policy Optimization can be effective in training MARL tasks, they often fail to resolve local conflict between agents and are unable to generalize to stochastic events. In this paper, we propose a Local State Attention module to assist the input state representation. By relying on the self-attention operator, the module is expected to compress the essential information of nearby agents to resolve the conflict in traffic situations. Utilizing a simulated highway merging scenario with the priority vehicle as the unexpected event, our approach is able to prioritize other vehicles' information to manage the merging process. The results demonstrate significant improvements in merging efficiency compared to popular baselines, especially in high-density traffic settings.