We propose algorithms to automatically deploy a group of mobile robots and provide coverage of a non-convex environment with communication limitations. In settings such as hilly terrain or for underwater ocean gliders, peer-to-peer communication can be impossible and frequent communication to a central base station may be impractical.
This paper instead explores how to perform coverage control when each robot has only asynchronous and sporadic communication with a base station. The proposed algorithms rely upon overlapping territories, monotonically minimize suitable cost functions, and provably converge to a centroidal Voronoi partition. We also describe how the use of overlapping territories allows our algorithms to smoothly handle dynamic changes to the robot team.