Distributed coverage of nonconvex environments

Abstract

Sensor networks and multi-agent robotic systems have been receiving increasing attention in recent times. This is due in no small part to the remarkable advances made in recent years in the development of small, agile, relatively inexpensive sensor nodes with mobile and networking capabilities. These sensor nodes are envisioned to be the basic components of complex networks intended to perform a wide variety of tasks. These include search and rescue, exploration, environmental monitoring, location-aware computing, and the maintaining of structures. The potential advantages of employing arrays of robotic sensors are numerous. For instance, certain tasks are difficult, if not impossible, when performed by a single agent. Further, a group of agents inherently provides robustness to failures of single agents or communication links. The existence of such motion-enabled sensing devices and the anticipated development of still more advanced versions raise compelling questions. A particularly important issue is whether large numbers of such small autonomous devices will be successfully deployed as a search team to cooperatively carry out a prescribed task reliably, robustly and adaptively, without a centralized controller and with limited communications among its members. Motivated by these future scenarios, this chapter focuses on algorithms for visually-guided agents, i.e., mobile robotic agents with line-of-sight sensing and communication capabilities, to solve a distributed version of the Art Gallery Problem. In the remainder of the introduction, we describe the problem in its original context, broadly highlight the characteristics of visually-guided agents and reformulate the original problem with respect to visually-guided agents.