We consider the problem of estimating the state in multi-agent decision and control systems. A novel approach to state estimation is developed that uses partial order theory in order to overcome some of the severe computational complexity issues arising in multi-agent systems. State estimation algorithms are developed that enjoy provable convergence properties and are scalable with the number of agents. Application examples are considered, which include state estimation in competitive multi-robot systems. A final application example shows how to extend the proposed state estimation approach to the context of monitoring distributed environments.
Back to Control Seminars Page