Change detection in the dynamics with recursive subspace indentification

H. Oku; G. Nijsse; M.H.G. Verhaegen; V. Verdult

doi:10.1109/.2001.980602

Change detection in the dynamics with recursive subspace indentification

H. Oku, G. Nijsse, M.H.G. Verhaegen, V. Verdult

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

160 Downloads (Pure)

Abstract

We propose a tracking mechanism to follow time-variations in the dynamics of a linear system with a recursive identification of the PI-MOESP scheme. The proposed mechanism consists of change detection scheme followed by a re-initialization of the recursive calculations. The change detection is based on the least-squares interpretation of the calculation in the subspace scheme and detects whether the estimates of the recursive solution without exponential forgetting lies in the confidence interval of the estimates obtained with a second finite-window length solution to the least-squares problem. When a change has been detected, the estimate by the recursive implementation is re-initialized via the solution of a constrained least-squares problem. One numerical example is presented to illustrate that our change detection and reinitialization scheme can detect incipient changes in the system dynamics without detecting changes in input dynamics

Original language	English
Title of host publication	Proceedings of the 40th IEEE Conference on Decision and Control
Place of Publication	Orlando, Forida, USA
Publisher	IEEE
Number of pages	6
Volume	3
ISBN (Print)	0-7803-7063-5
DOIs	https://doi.org/10.1109/.2001.980602
Publication status	Published - 4 Dec 2001
Event	40th IEEE Conference on Decision and Control, CDC 2001 - Hyatt Regency Grand Cypress, Orlando, United States Duration: 4 Dec 2001 → 7 Dec 2001

Conference

Conference	40th IEEE Conference on Decision and Control, CDC 2001
Abbreviated title	CDC
Country/Territory	United States
City	Orlando
Period	4/12/01 → 7/12/01

Access to Document

10.1109/.2001.980602

Cite this

@inproceedings{a12790d035f2407b806401144613217e,

title = "Change detection in the dynamics with recursive subspace indentification",

abstract = "We propose a tracking mechanism to follow time-variations in the dynamics of a linear system with a recursive identification of the PI-MOESP scheme. The proposed mechanism consists of change detection scheme followed by a re-initialization of the recursive calculations. The change detection is based on the least-squares interpretation of the calculation in the subspace scheme and detects whether the estimates of the recursive solution without exponential forgetting lies in the confidence interval of the estimates obtained with a second finite-window length solution to the least-squares problem. When a change has been detected, the estimate by the recursive implementation is re-initialized via the solution of a constrained least-squares problem. One numerical example is presented to illustrate that our change detection and reinitialization scheme can detect incipient changes in the system dynamics without detecting changes in input dynamics",

author = "H. Oku and G. Nijsse and M.H.G. Verhaegen and V. Verdult",

year = "2001",

month = dec,

day = "4",

doi = "10.1109/.2001.980602",

language = "English",

isbn = "0-7803-7063-5",

volume = "3",

booktitle = "Proceedings of the 40th IEEE Conference on Decision and Control",

publisher = "IEEE",

address = "United States",

note = "40th IEEE Conference on Decision and Control, CDC 2001, CDC ; Conference date: 04-12-2001 Through 07-12-2001",

}

TY - GEN

T1 - Change detection in the dynamics with recursive subspace indentification

AU - Oku, H.

AU - Nijsse, G.

AU - Verhaegen, M.H.G.

AU - Verdult, V.

PY - 2001/12/4

Y1 - 2001/12/4

N2 - We propose a tracking mechanism to follow time-variations in the dynamics of a linear system with a recursive identification of the PI-MOESP scheme. The proposed mechanism consists of change detection scheme followed by a re-initialization of the recursive calculations. The change detection is based on the least-squares interpretation of the calculation in the subspace scheme and detects whether the estimates of the recursive solution without exponential forgetting lies in the confidence interval of the estimates obtained with a second finite-window length solution to the least-squares problem. When a change has been detected, the estimate by the recursive implementation is re-initialized via the solution of a constrained least-squares problem. One numerical example is presented to illustrate that our change detection and reinitialization scheme can detect incipient changes in the system dynamics without detecting changes in input dynamics

AB - We propose a tracking mechanism to follow time-variations in the dynamics of a linear system with a recursive identification of the PI-MOESP scheme. The proposed mechanism consists of change detection scheme followed by a re-initialization of the recursive calculations. The change detection is based on the least-squares interpretation of the calculation in the subspace scheme and detects whether the estimates of the recursive solution without exponential forgetting lies in the confidence interval of the estimates obtained with a second finite-window length solution to the least-squares problem. When a change has been detected, the estimate by the recursive implementation is re-initialized via the solution of a constrained least-squares problem. One numerical example is presented to illustrate that our change detection and reinitialization scheme can detect incipient changes in the system dynamics without detecting changes in input dynamics

U2 - 10.1109/.2001.980602

DO - 10.1109/.2001.980602

M3 - Conference contribution

SN - 0-7803-7063-5

VL - 3

BT - Proceedings of the 40th IEEE Conference on Decision and Control

PB - IEEE

CY - Orlando, Forida, USA

T2 - 40th IEEE Conference on Decision and Control, CDC 2001

Y2 - 4 December 2001 through 7 December 2001

ER -

Change detection in the dynamics with recursive subspace indentification

Abstract

Conference

Access to Document

Fingerprint

Cite this