Workspace Allocation for Team of Robots with Different Actuation Capabilities

Abstract

In this paper, the adaptation of multiple agents to actuation performance variations under localization uncertainty is investigated. The agents are modeled as nonholonomic wheeled mobile vehicles and Guaranteed Power Diagrams (GPD or GPVD) are used in order the robots to partition the workspace under the assumption that the locations of the agents are known within uncertainty circles. An online adaptive estimation algorithm is used for each agent to calculate the weight of its GPV-cell. So, the robots accomplish the coverage control to perform a collaborative task by giving larger portions of the workspace to the agents that have better actuation performances and smaller regions to the ones whose performances are worse than the other agents. Simulation results and ROS implementation show the effectiveness of the algorithm.