Robust parametric predictive control design for polytopically uncertain systems

Abstract

In this paper we present an algorithm for the design of robust model based predictive controller for polytopically uncertain systems via parametric programming. A min-max approach is adopted to design robust parametric predictive controllers, which guarantees feasible plant operation for maximum violation of polytopic uncertainty in system matrices and bounded input disturbances. The resulting piece wise affine optimal control law is a function of states and can be implemented online as a sequence of simple function evaluations. An example is presented to illustrate the details of the proposed robust parametric predictive controller design.