A Cooperative Game Theoretical Approach to Distributed Iterative Command Governor Schemes

Abstract

A novel distributed Command Governor (CG) supervision strategy relying on cooperative game theoretical concepts is presented for multi-agent networked systems subject to pointwise-in-time coordination constraints. Unlike non-cooperative distributed CG schemes, here all agents contribute individually to the minimization of a global performance index. As a result, these methods are able to achieve Pareto-optimal solutions, not only in steady-state conditions as the non-cooperative ones, but also during transients and are not hampered by the presence of undesirable non-Pareto Nash-equilibria or deadlock situations. Other noticeable difference with respect to non-cooperative methods is that all agents need to exchange data among them several times within a decision step in order to coordinate their local optimization procedures and arrive to the optimal solution. The properties of the algorithm are fully investigated. A final example is presented where the proposed distributed solution is contrasted with both centralized CG solutions and distributed CG methods based on non-cooperative game theoretical concepts.