Multi-agent coordination for target tracking using fuzzy inference system in game theoretic framework

Abstract

This paper presents a multiple robot coordination method for target tracking problem. The coordination is to achieve a desired formation during the team operation. The individual decision of robots on the moving direction induces conflict situation within the team, a typical feature in the noncooperative game theory that makes the global coordination nontrivial. The contribution of the paper is a game theoretic approach that improves the convergence of target tracking independently of the initial weight of cost components. The method uses semi-cooperative Stackelberg equilibrium instead of Nash equilibrium, a new formation component in the individual cost functions and a fuzzy inference system for high level cost weight tuning. The results are simulated on a target tracking example where the target is followed by a team of three simple mobile robots.