RAILGUN: A Unified Convolutional Policy for Multi-Agent Path Finding Across Different Environments and Tasks
This addresses the NP-hard MAPF problem for applications like aerial swarms and warehouse automation by providing a centralized solution that overcomes limitations of decentralized approaches, though it is incremental as it builds on learning-based methods.
The paper tackles the Multi-Agent Path Finding (MAPF) problem by introducing RAILGUN, the first centralized learning-based policy that uses a CNN-based architecture to generalize across maps and agent numbers, outperforming most baselines and showing strong zero-shot generalization in experiments.
Multi-Agent Path Finding (MAPF), which focuses on finding collision-free paths for multiple robots, is crucial for applications ranging from aerial swarms to warehouse automation. Solving MAPF is NP-hard so learning-based approaches for MAPF have gained attention, particularly those leveraging deep neural networks. Nonetheless, despite the community's continued efforts, all learning-based MAPF planners still rely on decentralized planning due to variability in the number of agents and map sizes. We have developed the first centralized learning-based policy for MAPF problem called RAILGUN. RAILGUN is not an agent-based policy but a map-based policy. By leveraging a CNN-based architecture, RAILGUN can generalize across different maps and handle any number of agents. We collect trajectories from rule-based methods to train our model in a supervised way. In experiments, RAILGUN outperforms most baseline methods and demonstrates great zero-shot generalization capabilities on various tasks, maps and agent numbers that were not seen in the training dataset.