Interval identification of linear systems with bounded errors in both input and output variables

An approach for evaluating upper and lower bounds of the parameters of linear systems is proposed to handle the case when the input and output disturbances are unknown but bounded. The authors consider the concept of interval identification as a process of constructing interval models and estimating unknown plant parameters in the form of closed intervals from experimental data. They not only include the usual additive disturbance, but also include disturbance in the input matrix.<<ETX>>