MIMOptMPC: A MATLAB™ toolbox for off-line robustification of multivariable MPC

Robustness of Model Predictive Control (MPC) has been extensively investigated, considering both on-line and off-line strategies. However the implementation of the theoretical background required within such methodologies may appear complicated to the users, in particular in terms of tuning parameters. To overcome this difficulty, this paper proposes a new powerful toolbox which permits off-line robustness improvement of multivariable unconstrained MPC via the convex optimization of the Youla parameter. This toolbox, so-called MIMOptMPC, provides robustified controllers, guaranteeing robust stability under model unstructured uncertainties and nominal performance specifications for the disturbances rejection, while being a user friendly and easily extensible tool. In addition, MIMOptMPC offers visualization facilities and allows relevant comparisons between results obtained with the different robustified MPC controllers.