Sweeping task of multiple mobile agents by utilizing behavior selection model with interaction-based efficacy dynamics

Abstract

This paper discusses what is given to control of multi-agent system with dynamic behavior selection model which we already proposed. Cricket shows adaptive behavior modification in both of individual and group level based on individual interactions. Principle of such ability is useful to control multi-agent system by bottom-up approach. In this paper, we compare our proposed model with other behavior selection models in sweeping task and discuss its effects from computer simulation results. © 2012 The Japan Society of Mechanical Engineers.