Agent-based guaranteed estimation and control of nonlinear systems

This paper presents a distributed guaranteed estimation and control technique for a class of nonlinear system. The system outputs are being measured by a set of agents whose objective is the estimation of the whole state. Furthermore, one agent is responsible for the control of the plant. Compared to the case of distributed estimation, the joint problem becomes harder to solve due to the fact that the agents ignore the actual control action being applied to the plant. The solution proposed makes use to zonotopes to find a set that contains the control action. The proposed algorithm has been tested with a simulation example.