An extension of two degree-of-freedom of generalized predictive control for M-input M-output systems based on state space approach

This paper proposes an extension method of two degree-of-freedom generalized predictive control (GPC) for m-input m-output systems by using coprime factorization approach. The extended controller has a new design parameter which can change controller poles without changing closed-loop poles. And the proposed method reveals an effect of integral compensation only if there exists modeling error and/or disturbance. Therefore, performance degradation due to integral compensation, such as a slow response or an excessive control effort, can be avoided when the controlled system has no perturbation.