Set-Membership Identification of Systems with Parametric and Nonparametric Uncertainty

R. Kosut, M. Lau, and S. Boyd

IEEE Transactions on Automatic Control, 37(7):929-941, July 1992

A method is presented for parameter set estimation where the system model is assumed to contain both parametric and nonparametric uncertainty. In the disturbance-free case, the parameter set estimate is guaranteed to contain the parameter set of the true plant. In the presence of stochastic disturbances, the parameter set estimate obtained from finite data records is shown to have the property that it contains the true-plant parameter set with probability one as the data length tends to infinity.