Immune network control for stigmergy based foraging behaviour of autonomous mobile robots

Abstract

Abstract The paper presents a series of experiments in a simulated environment where two autonomous mobile robots gather randomly distributed objects and cluster them on a pile. The co‐ordination of the robots' movements is achieved through stigmergy (an indirect form of communication through the environment). The random moves, necessary for stigmergy based foraging behaviour, make the task solution a time consuming process. In order to speed up the foraging behaviour, the immune network robot control is proposed. Stigmergic principles are coded in two artificial immune networks—for a collision free goal following behaviour and for an object picking up/dropping behaviour. Simulations confirm the improved performance of the foraging behaviour under the proposed immune network control. Copyright © 2006 John Wiley & Sons, Ltd.