Robust MPC controller design: Infinite prediction horizon

We present a new technique for the synthesis of a robust model predictive controller with infinite prediction horizon. The proposed controller design approach ensures a parameter-dependent quadratic stability and guaranteed cost for linear polytopic uncertain systems. The proposed design procedure based on BMI or LMI reduces to off-line output feedback controller gain calculation minimizing an upper bound on the “worst case” an objective function. For the case of our approach and input (output) constraints at each sample time step for each input (output) scalar parameter is calculated, which radically decreases the computational load. The controller design is illustrated by examples.