Skip to main navigation Skip to search Skip to main content

High-order least squares identification: A new approach

  • Rajamani Doraiswami
  • , Lahouari Cheded

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

A high-order least squares (HOLS) identification method is proposed to identify a system which is subject to disturbances affecting both the dynamical equations and the measurement noise. A linear regression model of the system is obtained based on the Kalman filter structure whose equation error is a colored noise process generated by the residual of the Kalman filter. A high-order regression model is derived by whitening the colored noise process so that the equation error is a zero-mean white noise process. The high-order model is estimated using the least-square method. A reduced-order model is derived from the high-order model using the frequency-weighted least-squares method. The proposed scheme has been successfully evaluated on a number of simulated and physical systems and favorably compared with the prediction error method (PEM).

Original languageEnglish
Title of host publicationICINCO 2013 - Proceedings of the 10th International Conference on Informatics in Control, Automation and Robotics
Pages147-152
Number of pages6
StatePublished - 2013

Publication series

NameICINCO 2013 - Proceedings of the 10th International Conference on Informatics in Control, Automation and Robotics
Volume1

Keywords

  • High Order Model
  • Kalman Filter
  • Least Squares Method
  • Prediction Error Method

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems
  • Control and Systems Engineering

Fingerprint

Dive into the research topics of 'High-order least squares identification: A new approach'. Together they form a unique fingerprint.

Cite this