Multiple source seeking via distributed sample-variance control of swarm robots

Abstract

In this article, we deal with a multiple source seeking problem via distributed control of swarm robots. Source seeking is the task of finding sources of hazardous events, which can be effectively carried out by using swarm robots. However, the robots might gather on a few of the sources if they are equipped only with locally detectable sensor and short-range communication devices. To overcome this difficulty, in this study, we propose a method to appropriately spread the robots out by using only local information. The key of the method is to control the sample variance of the robot coordinates based on a distributed-optimization technique. Although the sample variance depends on all robot coordinates, the proposed method enables each robot to compute its value via short-range communication between robots using local information on neighbouring robots. Finally, simulation results are presented to illustrate the effectiveness of the proposed method.