SM identification of approximating models for H/sub /spl infin// robust control

Set membership (SM) H/sub /spl infin// identification of mixed parametric and nonparametric models is investigated, which aims to estimate a low order approximating model and an identification error, giving a measure of the unmodeled dynamics in a form well suited for H/sub /spl infin// control methodologies. In particular, the problem of estimating the parameters of the parametric part and the H/sub /spl infin// bound on the modeling error is solved using frequency domain data, supposing l/sub /spl infin// bounded measurement errors and exponentially stable unmodeled dynamics. The effectiveness of the proposed procedure is tested on some numerical examples, showing the advantages of the proposed methods over the existing nonparametric H/sub /spl infin// identification approaches are shown, in terms of lower model order and of tightness in the modeling error bounds.