A simulation based MPC technique for feedback linearizable systems with input constraints

This paper proposes a novel methodology for applying MPC for nonlinear, feedback linearizable systems with input constraints. Earlier approaches coupling MPC and Feedback linearization techniques were limited by a basic factor; although the system dynamics were transformed to a linear system via feedback linearization, the initial input constraints were mapped to a set of nonlinear, and in general non convex bounds, and hence there is no guarantee that the global optimum will be found. The main advantage of the approach proposed in this paper is that by using an iterative process, at every timestep in the resulting optimization problem both dynamics and constraints are linear. The efficiency and robustness of the proposed scheme is verified via simulations in two case studies. The stability of the zero dynamics in these examples is investigated numerically, by using a two-stage approach based on reachability analysis.