A new stability robustness result for estimator-based controllers with application

An approach for designing robust compensators, consisting of a state estimator and a state-feedback controller, is presented. Fundamental to the method is the derivation of a general condition for the stability robustness of a linear multivariable multiplicative type. The derived condition directly relates the singular value of the multiplicative uncertainty matrix to the values of arbitrary controller and estimator gains. By incorporating a search algorithm with the stability robustness condition, an iterative scheme which allows the computation of a robust estimator gain (in the case of input uncertainty) is obtained. The design method is applied to a numerical example for illustration