Optimal Control and Game Theoretic Approaches to Cooperative Control of a Team of Multi-Vehicle Unmanned Systems

Abstract

The main goal of this work is to design a team of multi-agent unmanned vehicles that can accomplish consensus over a common value for vehicles' output in a leaderless topology. The interaction between agents due to information exchange is modelled in the characterization of the dynamical model of each agent. The main approach to the problem is based on cooperative game theory, however in order to clarify the cooperative nature of the method, two approaches to this problem are considered: First a semi-decentralized optimal control strategy is designed based on minimization of individual costs using local information. Next, cooperative game theory is used to ensure the team cooperation by combining individual cost into a team cost function. This cooperative solution needs the full information of the team but results in lower cost for each agent. The choice of Nash-bargaining solution among a set of Pareto-efflcient solutions for this cooperative game guarantees minimum individual cost by maximizing the difference between the cooperative (centralized) and non-cooperative (semi-decentralized) individual cost.