Brief Parameter identification with derivative shift operator parametrization

Authors:
Lin Guo;Masayoshi Tomizuka
Affiliations:
Department of Mechanical Engineering, University of California at Berkeley, Berkeley, CA 94710, USA;Department of Mechanical Engineering, University of California at Berkeley, Berkeley, CA 94710, USA
Venue:
Automatica (Journal of IFAC)
Year:
1999

Citing 8
Cited 1

Round-off error propagation in four generally-applicable, recursive, least-squares estimation schemes

Automatica (Journal of IFAC)
Computer-controlled systems: theory and design (2nd ed.)

Computer-controlled systems: theory and design (2nd ed.)
Adaptive Control

Adaptive Control
Parametrizations in Control, Estimation, and Filtering Problems: Accuracy Aspects

Parametrizations in Control, Estimation, and Filtering Problems: Accuracy Aspects
Digital Control and Estimation: A Unified Approach

Digital Control and Estimation: A Unified Approach
On properties of information matrices of delta-operator basedadaptive signal processing algorithms

IEEE Transactions on Signal Processing
Identification of continuous-time models

IEEE Transactions on Signal Processing
Paper: Error propagation properties of recursive least-squares adaptation algorithms

Automatica (Journal of IFAC)

Adaptive control: the issue of short sampling period

SMO'05 Proceedings of the 5th WSEAS international conference on Simulation, modelling and optimization

Quantified Score

Hi-index	22.14

Visualization

Abstract

A way of parametrizing linear systems using an operator which combines the two commonly used operators, the shift(q) and the delta(@d) operators, is proposed. It is shown that the proposed parametrization improves the numerical property of parameter identification process. By using the proposed operator the condition number of the information matrix in the identification algorithm is significantly lower than that using the q- or @d-operator. It is also shown that accurate and fast converging parameter identification can be achieved by using the new operator at fast sampling rates, especially for systems with relatively wide bandwidth. The numerical superiority of the new parametrization is demonstrated by simulation results.