Roadmap-Optimal Multi-robot Motion Planning using Conflict-Based Search

Multi-Agent Pathfinding (MAPF) is the problem of finding a set of feasible paths for a set of agents with specific individual start and goal poses. It is considered computationally hard to solve. Conflict-based search (CBS) has shown optimality in developing solutions for multi-agent pathfinding problems in discrete spaces. However, neither CBS nor other discrete MAPF techniques can be directly applied to solve Multi-Agent Motion Planning (MAMP) problems, the continuous version on multi-agent pathfinding. In this work, we present the extension of the CBS discrete approach to solve Sampling-based Motion planning problems, and we show its capabilities to produce roadmap-optimal solutions for multi-robot motion planning problems.