WEIGHTED LEAST-SQUARES SET ESTIMATION FROM L(INFINITY)-NORM BOUNDED NOISE DATA

Authors
KEESMAN KJ
Citation
Kj. Keesman, WEIGHTED LEAST-SQUARES SET ESTIMATION FROM L(INFINITY)-NORM BOUNDED NOISE DATA, IEEE transactions on automatic control, 42(10), 1997, pp. 1456-1459
Citations number
15
Categorie Soggetti
Controlo Theory & Cybernetics","Robotics & Automatic Control","Engineering, Eletrical & Electronic
Journal title
IEEE transactions on automatic control → ACNP
ISSN journal
00189286
Volume
42
Issue
10
Year of publication
1997
Pages
1456 - 1459
Database
ISI
SICI code
0018-9286(1997)42:10<1456:WLSEFL>2.0.ZU;2-5
Abstract
The problem of parameter set estimation from pointwise bounded-error d ata is considered. The possibilities of employing l(2)-projection proc edures to solve the problem are explored, and exact as well as approxi mate outer-bounding solutions are proposed. In particular, the propert ies of weighted least squares set estimation in this l(infinity) norm bounded-error context and the implementation of a resulting minimum-vo lume parallelotope-bounding algorithm are discussed.