Motion Planning for Multiple Mobile Robot Systems using Dynamic Networks

Abstract

Presented is a new approach to multi-robot motion planning that is based on the concept of planning within dynamic robot networks. The system enables multiple mobile robots that have limited ranges of sensing and communication to maneuver safely in dynamic, unstructured environments. As the robots move about their workspace, localized robot groups form networks within which world models can be shared and centralized motion plans can be calculated in parallel. The motion planning algorithm used within networks is based on kinodynamic randomized motion planning techniques that construct trajectories in real-time. Both simulations and real robot experiments are used to validate the system.