Source seeking by distributed swarm robots with sample variance control

This paper deals with a multi-source seeking problem by swarm robots. The source seeking is a task to find sources in a field, e.g., fire sites and leak places of chemicals or radiation. By using swarm robots, it is expected to find multiple sources effectively. In this paper, we design a distributed controller to appropriately spread the robots out on the multiple sources not to gather on one source. For this purpose, our proposed method adjusts the sample variance of the robots' positions. Moreover, the proposed controller is distributed with which each robot only needs its local information. Therefore, the resultant swarm-robot system is leaderless, where the robots can continue their task even if some of the robots are broken down. Finally, the effectiveness of the proposed method is illustrated by simulations.