We study the problem of identification of an input to a linear finite-dimensional system. We assume that the input has a feedback form, which is related to a problem often encountered in fault detection. The method we use is to embed the identification problem in a class of inverse problems of dynamics for controlled systems. Two algorithms for identification of a feedback matrix based on the method of feedback control with a model are constructed. These algorithms are stable with respect to noise-corrupted observations and computational errors.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
We study the construction of an outer factor to a positive definite Popov function Pi(iomega) of a distributed parameter system. We assume that Pi(iomega) is a non-negative definite matrix with non-zero determinant. Coercivity is not assumed. We present a penalization approach which gives an outer factor just in the case when there exists any outer factor.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.