Abstract
The inverted interactorizing is the basic technique for the model matching control, the adaptive control, the decoupling control, the disturbance decoupling control, etc. However, this technique can be applied only to the minimum phase plant because all zeros of the plant are cancelled by the closed loop poles. This is due to the fact that the interactor matrix corresponds to zeros at infinity of the plant. In this paper, the generalized inverted interactorizing of linear multivariable non-minimum phase plants by the state feedback is concerned. This is the problem to find the state feedback which makes the closed loop transfer matrix coincide with the inverse of the generalized interactor which corresponds to not only zeros at infinity but also finite unstable zeros of the plant. Since it can be shown that this state feedback cancels only stable zeros of the plant, the closed loop system is internally stable. If the plant is a minimum phase system, the generalized interactor reduces to the ordinary interactor, so the proposed method can be regarded as the generalized technique of the inverted interactorizing to the non-minimum phase systems. As a special case of the inverted interactorizing, it is also shown that the plant is decouplable and internally stable if and only if its generalized interactor is diagonal.
