Multi-Agent Path Finding via Tree LSTM
This work addresses the performance gap in MAPF for RL researchers and practitioners, representing a strong incremental improvement over existing RL approaches.
The paper tackled the problem of Multi-Agent Path Finding (MAPF) in the Flatland3 Challenge, where previous RL methods underperformed compared to OR methods, and achieved a score of 125.3, which is several times higher than the best prior RL solution and comparable to top OR methods.
In recent years, Multi-Agent Path Finding (MAPF) has attracted attention from the fields of both Operations Research (OR) and Reinforcement Learning (RL). However, in the 2021 Flatland3 Challenge, a competition on MAPF, the best RL method scored only 27.9, far less than the best OR method. This paper proposes a new RL solution to Flatland3 Challenge, which scores 125.3, several times higher than the best RL solution before. We creatively apply a novel network architecture, TreeLSTM, to MAPF in our solution. Together with several other RL techniques, including reward shaping, multiple-phase training, and centralized control, our solution is comparable to the top 2-3 OR methods.