Sampling-based Multi-robot Motion Planning

This paper describes a sampling-based approach to multi-robot motion planning. The proposed approach is centralized, which aims to reduce interference between mobile robots such as collision, congestion and deadlock, by increasing the number of waypoints. The implementation based on occupancy grid map is decomposed into three steps: the first step is to identify primary waypoints by using the Voronoi diagram, the second step is to generate additional waypoints by sampling the Voronoi diagram, and the last step is to assign the waypoints to robots by using the Hungarian method. The approach has been implemented and tested in simulation and the experimental results show a good system performance for multi-robot motion planning.