Abstract
In this paper, we present a robust controller design method which achieves not only robust stability but also performance robustness for a linear multivariable system with norm bounded or structured additive uncertainties. The performance robustness means that the deterioration of control performance is suppressed when we compare the time response for the real system with the desirable one generated by using the nominal system directly. In this approach, we assume that the control law consists of a state feedback with the feedback gain which is designed in order to generate the desirable transient behavior for the nominal system and a compensation input for the uncertainties. The compensation input is determined so that the upper bound of a quadratic cost function for the error system between the real and the nominal system is minimized. We show that stability of the error system is reduced to the solvability of a Riccati Equation and give a guide to set up a design parameter. Finally, numerical examples are presented.
