Distributed Heuristic Multi-Agent Path Finding with Communication
This work addresses the challenge of cooperation in congested scenarios for robotic systems, presenting an incremental improvement over existing learning-based methods.
The paper tackles the problem of achieving collision-free policies in multi-agent path finding by combining communication with deep Q-learning and using heuristic guidance from potential shortest paths, resulting in high success rates and low average steps in obstacle-rich environments.
Multi-Agent Path Finding (MAPF) is essential to large-scale robotic systems. Recent methods have applied reinforcement learning (RL) to learn decentralized polices in partially observable environments. A fundamental challenge of obtaining collision-free policy is that agents need to learn cooperation to handle congested situations. This paper combines communication with deep Q-learning to provide a novel learning based method for MAPF, where agents achieve cooperation via graph convolution. To guide RL algorithm on long-horizon goal-oriented tasks, we embed the potential choices of shortest paths from single source as heuristic guidance instead of using a specific path as in most existing works. Our method treats each agent independently and trains the model from a single agent's perspective. The final trained policy is applied to each agent for decentralized execution. The whole system is distributed during training and is trained under a curriculum learning strategy. Empirical evaluation in obstacle-rich environment indicates the high success rate with low average step of our method.