Statistical analysis of least-squares identification for robust control design: output error case with affine parameterization

Abstract

Precise, finite-data statistical properties are determined using a least-squares estimator based on an output error model with an affine parameter representation where the true system is of output error form, but is not in the model set. The purpose of the analysis is to show the effect of unmodeled dynamics on the resulting closed-loop system designed on the basis of the estimated transfer function. This simple problem set-up is prototypical of the interplay between system identification and robust control design. Show more