Centralized Decoupled Path Planning Algorithm for Multiple Robots Using the Temporary Goal Configurations

Abstract

This paper presents a problem of finding collision-free paths for multiple robots with individual start configurations and goal configurations. With the decoupled method, we try to find the collision-free paths if the solutions exist. The proposed algorithm consists of two phases. In the first phase, the path is obtained from the path planner for a single robot ignoring the existence of the other robots. To avoid mutual collisions, we assign temporarily either changed goal configurations to the robots which can be regarded as the obstacles or changed start configuration of the robot trying to move. We show that the proposed algorithm solves path planning problem in various scenarios.