Team-Based Blanket Coverage Control for Heterogeneous Multi-agent Systems

This paper presents a team-based blanket coverage strategy that allows for deploying a group of heterogeneous agents in river-like regions. The proposed team-based partition method aims at minimizing an accumulated sensing cost function. To this end, the coverage region is first partitioned into a blanket coverage area at the team level using the Voronoi diagram, and then the blanket area assigned to the team is partitioned among its members by the so-called Power diagram. The blanket coverage area, a region with dynamic boundaries, is driven to move within the mission region. In the process, a gradient descent-based control law is designed by introducing the dynamics of the boundaries into the controller. At the same time, the coverage areas assigned to agents can be adjusted online by an adaptation control law, where the performance variations among the agents are taken into account to change their weight factors. The proposed control laws guarantee the agents could converge to a locally optimal coverage configuration.