Set--Membership Identification of Parametric Systems

This paper considers the problem of robust, set membership identification of parametric LTI plants, using frequency domain data. We consider the case of noisy data, and provide tractable, LMI-based conditions for computing inner and outer approximations to the set of parameters so that the resulting plant is consistent with a given a priori information and interpolate the experimental data. The results presented here are useful for model (in)validation, robust control synthesis and fault diagnosis