A decentralized approach for cooperative sweeping by multiple mobile robots

Abstract

Presents a decentralized cooperative sweeping approach for area sweeping by multiple mobile robots. Approaches for implementing both autonomous function and cooperative function are proposed and described. An online goal selection (OGS) algorithm is proposed for motion planning to guide autonomous behavior. To achieve cooperation, the robot society is organized as a decentralized market-like structure and a negotiation mechanism is used to resolve the task-sharing problem. A hybrid system architecture is implemented to combine both cooperative and autonomous functions. A simulation software environment is built for testing and evaluation.