Controller reduction via minimum rank matrix approximation

In this paper a controller reduction method for discrete-time linear time-invariant systems is described. Using the bounded-real lemma, the proposed method generates reduced controllers with closed loop stability and H∞H∞ norm performance guarantee. Information of the full controller is used as a basis for reduction using singular value decomposition. This is different from traditional model reduction schemes such as weighted balanced truncation. Numerical assessment of the proposed method is given in the end of the paper.