摘要：AbstractThe problem of reconstructing the internal state of a linear time-varying system through delayed measurements is addressed, under the assumption that the delay is known and constant. To estimate the state, a delayed version of the Kalman-Bucy observer is proposed, and a delay size ensuring the convergence of the observer is provided. This bound depends on the size of the system matrices and the observer gain. If the delay is larger, a scheme of cascade observers is proposed. This allows to handle arbitrary large delays at the cost of implementing more observers. The strategy is then applied to the problem of parameter estimation in a numerical exercise.
