Constrained receding horizon predictive control for nonlinear systems

This paper represents a receding horizon predictive control algorithm for constrained nonlinear systems which, unlike earlier works, can be solved by linear programming methods. Use is made of a terminal inequality constraint in conjunction with a cost penalizing an upper bound on the tracking error over a finite control horizon. The optimization procedure is based on predictions made by linearized incremental models at points of a given seed trajectory and the effects of linearization error are taken into account to give a bound on the predicted tracking error. The proposed algorithm is posed in the form of LP and its asymptotic stability can be guaranteed by proper selection of the terminal weights of the cost index.