A Nonlinear Model Predictive Control Scheme with Multirate Integral Sliding Mode

In this paper, the Multirate Integral Sliding Mode (MRISM) control strategy for nonlinear discrete-time systems is proposed. The MRISM controller acts at a faster sampling time than a high level controller, and reduces the effect of model uncertainties and external disturbances, in order to obtain, at the next sampling instant of the high level controller, a value of the system state as close as possible to the nominal one. To obtain this result, the control variable is composed by two parts, one generated by the high level controller, and the other by the MRISM controller like in the classical integral sliding mode approach. The proposed strategy is used in connection with robust nonlinear model predictive control (MPC); the a-priori reduction of the disturbance terms proves to be very useful in order to guarantee the convergence properties of the MPC controller. As a result, the merging of the two controllers unites the advantages of sliding mode control (strong reduction of disturbances, low computational burden) to those of MPC (optimality, constraints handling).