Communication-aware coverage control for robotic sensor networks

Abstract

The purpose of this paper is to propose a control scheme to maximize area coverage and at the same time ensure reliable communication in networks of mobile robot sensors. The information that is generated at the sensors depends on the sensing capabilities of the sensors as well as on the frequency at which events occur in their vicinity, captured by appropriate probability density functions. This information is then routed to a fixed set of access points via a multi-hop network whose links model the probability that information packets are correctly decoded at their intended destinations. The proposed hybrid controller simultaneously optimizes coverage and routing of information by decoupling control in the continuous and discrete domains. The robots' motion is performed in continuous time, along the negative gradient of a cost function that combines the coverage objective and a barrier potential used to ensure satisfaction of desired communication rates. On the other hand, the communication variables are updated periodically, in discrete time, by the solution of an optimization problem, and constitute the switching signal in the continuous motion control. Simulation studies are conducted verifying the efficiency of the proposed algorithm.