Route planning method of multiple mobile robots for transportation task based on collision prospect

Multiple robot system with decentralized control attracts attention and is expected to play an important role in transportation systems in factories and warehouses. In transportation systems, it is necessary to plan a route that minimizes the transportation time with less collision and deadlock. The goal of this study is to calculate appropriate routes with reducing communication load and computational load. In this paper, we proposed to combine prior route planning and local avoidance. In order to reduce computational load, each robot predict probability of collision with other robots by comparing present and destination position. A route with less collision is calculated by reflecting the overlap of route which has probability of collision into dijkstra method. The proposed method is evaluated through the computer simulation and the experiment using real robots.