Learning to Resolve Conflicts for Multi-Agent Path Finding with Conflict-Based Search
This work provides an incremental improvement to the conflict resolution strategy in CBS, which is a state-of-the-art algorithm for multi-agent path finding.
This paper addresses the problem of conflict selection in Conflict-Based Search (CBS) for multi-agent path finding. By learning from an oracle, the proposed method significantly improves success rates, reduces search tree sizes, and decreases runtimes compared to the current state-of-the-art CBS solver.
Conflict-Based Search (CBS) is a state-of-the-art algorithm for multi-agent path finding. At the high level, CBS repeatedly detects conflicts and resolves one of them by splitting the current problem into two subproblems. Previous work chooses the conflict to resolve by categorizing the conflict into three classes and always picking a conflict from the highest-priority class. In this work, we propose an oracle for conflict selection that results in smaller search tree sizes than the one used in previous work. However, the computation of the oracle is slow. Thus, we propose a machine-learning framework for conflict selection that observes the decisions made by the oracle and learns a conflict-selection strategy represented by a linear ranking function that imitates the oracle's decisions accurately and quickly. Experiments on benchmark maps indicate that our method significantly improves the success rates, the search tree sizes and runtimes over the current state-of-the-art CBS solver.