Observer-based robust constrained model predictive control

This paper addresses the problem of observer-based robust constrained model predictive control with polytopic uncertain systems and norm-bound uncertain systems. It constructs the cost function directly by using the augmented system composed by states and estimation errors, and the observer gain need not be given in advance. In addition, this paper considers the constraints of system states and input in the designing of controller and observer, which relaxes the existing results to a certain extent. The performance of the proposed design method is demonstrated by an example.