Finite control set model predictive control assisted by a linear controller for true parameter uncertainty compensation

This work presents a new Finite Control Set Model Predictive Control (FCS-MPC) strategy based on an input reference tracking. The control structure admits a closed-loop representation featuring a linear dynamic and a perturbation term. Thus, the stability and the design process can be carried out by means of standard control methods. Besides, it reduces the computational burden because the minimization process does not consider the output predictions. In this work, the linear dynamic is included by designing an integral state-feedback linear controller. In addition, an offline optimization is proposed to find the optimal input which is applied in the next sampling instant. Finally, the results show that the proposed approach can lead to unbiased reference tracking even under severe parameters mismatch.