An actively adaptive control policy for linear models

Considers the quadratic optimal control of a discrete-time linear system with unknown parameters. The control action affects both the future behavior of the system and the precision of the future state and parameter estimates (dual effect). Using the information matrix, the posterior uncertainty on...

Full description

Saved in:

Bibliographic Details
Published in:	IEEE transactions on automatic control Vol. 41; no. 6; pp. 855 - 858
Main Authors:	Pronzato, L., Kulcsar, C., Walter, E.
Format:	Journal Article
Language:	English
Published:	New York, NY IEEE 01-06-1996 Institute of Electrical and Electronics Engineers
Subjects:	Adaptive control Applied sciences Computer science; control theory; systems Control systems Control theory. Systems Dynamic programming Exact sciences and technology Feedback control Linear systems Optics Optimal control Parameter estimation Physics State estimation Stochastic processes Uncertainty Autoregressive model Discrete system Linear system Adaptive control Optimal control Linear model
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Considers the quadratic optimal control of a discrete-time linear system with unknown parameters. The control action affects both the future behavior of the system and the precision of the future state and parameter estimates (dual effect). Using the information matrix, the posterior uncertainty on the parameters can be predicted, in order to design an actively adaptive control policy. The case of an autoregressive model is considered. Illustrative examples are presented, together with possible extensions and remaining open questions.
Bibliography:	ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23
ISSN:	0018-9286 1558-2523
DOI:	10.1109/9.506238