System identification in the presence of nonlinear sensors

Shuichi Adachi, Yasushi Okada, Jan M. Maciejowski

Research output: Contribution to journalConference article

Abstract

We consider a system identification problem in which input and output signals are measured using sensors with static nonlinear characteristics, such as saturation, dead zone, and so on. A new system identification method which estimates the sensor errors due to nonlinearity, as well as system parameters, is proposed. the key idea is to formulate the identification problem as a constrained optimization problem. the system parameters are then estimated using a quasi-Newton method. Effectiveness of the proposed method is examined through numerical experiments.

Original languageEnglish
Pages (from-to)185-190
Number of pages6
JournalIFAC Proceedings Volumes (IFAC-PapersOnline)
Volume37
Issue number12
DOIs
Publication statusPublished - 2004 Jan 1
Externally publishedYes
Event2004 IFAC Workshop on Adaptation and Learning in Control and Signal Processing, ALCOSP 2004 and IFAC Workshop on Periodic Control Systems, PSYCO 2004 - Yokohama, Japan
Duration: 2004 Aug 302004 Sep 1

Fingerprint

Identification (control systems)
Sensors
Constrained optimization
Newton-Raphson method
Experiments

Keywords

  • Constrained optimization problem
  • Modeling
  • System identification

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

System identification in the presence of nonlinear sensors. / Adachi, Shuichi; Okada, Yasushi; Maciejowski, Jan M.

In: IFAC Proceedings Volumes (IFAC-PapersOnline), Vol. 37, No. 12, 01.01.2004, p. 185-190.

Research output: Contribution to journalConference article

Adachi, Shuichi ; Okada, Yasushi ; Maciejowski, Jan M. / System identification in the presence of nonlinear sensors. In: IFAC Proceedings Volumes (IFAC-PapersOnline). 2004 ; Vol. 37, No. 12. pp. 185-190.
@article{5afec912f9ce4795a4dc1298f163154a,
title = "System identification in the presence of nonlinear sensors",
abstract = "We consider a system identification problem in which input and output signals are measured using sensors with static nonlinear characteristics, such as saturation, dead zone, and so on. A new system identification method which estimates the sensor errors due to nonlinearity, as well as system parameters, is proposed. the key idea is to formulate the identification problem as a constrained optimization problem. the system parameters are then estimated using a quasi-Newton method. Effectiveness of the proposed method is examined through numerical experiments.",
keywords = "Constrained optimization problem, Modeling, System identification",
author = "Shuichi Adachi and Yasushi Okada and Maciejowski, {Jan M.}",
year = "2004",
month = "1",
day = "1",
doi = "10.1016/S1474-6670(17)31465-9",
language = "English",
volume = "37",
pages = "185--190",
journal = "IFAC-PapersOnLine",
issn = "2405-8963",
publisher = "IFAC Secretariat",
number = "12",

}

TY - JOUR

T1 - System identification in the presence of nonlinear sensors

AU - Adachi, Shuichi

AU - Okada, Yasushi

AU - Maciejowski, Jan M.

PY - 2004/1/1

Y1 - 2004/1/1

N2 - We consider a system identification problem in which input and output signals are measured using sensors with static nonlinear characteristics, such as saturation, dead zone, and so on. A new system identification method which estimates the sensor errors due to nonlinearity, as well as system parameters, is proposed. the key idea is to formulate the identification problem as a constrained optimization problem. the system parameters are then estimated using a quasi-Newton method. Effectiveness of the proposed method is examined through numerical experiments.

AB - We consider a system identification problem in which input and output signals are measured using sensors with static nonlinear characteristics, such as saturation, dead zone, and so on. A new system identification method which estimates the sensor errors due to nonlinearity, as well as system parameters, is proposed. the key idea is to formulate the identification problem as a constrained optimization problem. the system parameters are then estimated using a quasi-Newton method. Effectiveness of the proposed method is examined through numerical experiments.

KW - Constrained optimization problem

KW - Modeling

KW - System identification

UR - http://www.scopus.com/inward/record.url?scp=85064413416&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064413416&partnerID=8YFLogxK

U2 - 10.1016/S1474-6670(17)31465-9

DO - 10.1016/S1474-6670(17)31465-9

M3 - Conference article

AN - SCOPUS:85064413416

VL - 37

SP - 185

EP - 190

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8963

IS - 12

ER -