Identification of dynamic characteristics of continuous linear models under conditions of complete parametric uncertainty

Citation
Av. Dmitriev et Ei. Druzhinin, Identification of dynamic characteristics of continuous linear models under conditions of complete parametric uncertainty, J COMP SYST, 38(3), 1999, pp. 376-385
Citations number
20
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL
ISSN journal
10642307 → ACNP
Volume
38
Issue
3
Year of publication
1999
Pages
376 - 385
Database
ISI
SICI code
1064-2307(199905/06)38:3<376:IODCOC>2.0.ZU;2-O
Abstract
A new algorithm for a direct computation of the order and dynamic character istics of a continuous linear model is designed by using data of a measurem ent unit of the state during the proper motion of an object. The algorithm uses the Hankel matrix, which is constructed by using averaged measurements of the state, A numerically stable realization of the proposed algorithm, including procedures for computation of the rank and pseudoinversion of the "averaged" Hankel matrix, is possible with the use of the technology of th e SVD decomposition of this matrix as well as without it.