Collective Behavior of Robots Swarm

Abstract

a great interest has evolved recently in the study and synthesis of cooperative robots with decentralized control. The basic problem is the development of efficient algorithms that could be inspired from the behavior of social insects. This paper develops from different perspective; the theory of decentralized discrete control systems. This leads to a little increase in the computational complexity in each individual, still non-intelligent, robot. Swarm intelligence appears as a result of the mutual control signals that replace coordination signals in usual centralized control systems. The study of asymptotic stability and the convergence to the global goal follows directly from the stability of algorithms derived within system theory and are reached stepwise in an efficient way. Through the paper we derive the necessary algorithms and procedures and finally we introduce the experimental results that validate the developed theory.