Tracking of Nonlinear Systems via Model Based Predictive Control

Abstract This paper presents a new Model Predictive Control ( MPC ) algorithm for nonlinear systems which solves the output tracking problem for asymptotically constant references. Closed-loop stability of the equilibrium and asymptotic zero-error regulation are guaranteed. The method is based on the solution of a finite-horizon optimization problem where the future error and control deviation norms are minimized. In the optimization it is assumed that a stabilizing linear dynamic regulator is used at the end of the control horizon, although it is never applied in practice. State and control constraints can be included in the problem and preprogrammed reference signals can be considered. The performances of the method are discussed on the classical Continuous Stirred Tank Reactor ( CSTR ) control application.