Dynamic coordinated control laws in multiple agent models

Abstract

We present an active control scheme of a kinetic model of swarming. It has been shown previously that the global control scheme for the model, presented in \cite{JK04}, gives rise to spontaneous collective organization of agents into a unified coherent swarm, via a long-range attractive and short-range repulsive potential. We extend these results by presenting control laws whereby a single swarm is broken into independently functioning subswarm clusters. The transition between one coordinated swarm and multiple clustered subswarms is managed simply with a homotopy parameter. Additionally, we present as an alternate formulation, a local control law for the same model, which implements dynamic barrier avoidance behavior, and in which swarm coherence emerges spontaneously.