On exponential convergence and robustness of least-squares identification

DOI10.1016/0167-6911(87)90099-5MaRDI QIDQ1088967zbMATH OpenOpenAlexFDO

Authors Markku T. Nihtilä

Publication date 1987

Published in Systems \& Control Letters (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/0167-6911(87)90099-5

zbMATH Keywords

exponential convergenceexponentially-weighted least-squares (EWLS) estimationextended persistency-of-excitationlinear regression models

Mathematics Subject Classification ID

Robustness and adaptive procedures (parametric inference) (62F35) Linear regression; mixed models (62J05) Estimation and detection in stochastic control theory (93E10) Identification in stochastic control theory (93E12)

Review text

This paper considers the exponentially-weighted least-squares (EWLS) estimation of the parameters in linear regression models. Its main results are: a) In the noise-free case the exponential convergence of the EWLS estimator is shown under an extended persistency-of-excitation condition; b) The exponential convergence in the noise-free case does not necessarily imply convergence in (bounded) noise; the basic assumption under which this non-robustness result is shown is that the regression vector is not persistently exciting (in the standard sense).

Recommendations

• Exponential convergence of recursive last squares with exponential forgetting factor
• On exponentially convergent parameter estimation with lack of persistency of excitation
• A recursive parameter estimator yielding exponential convergence under sufficient excitation
• Exponential convergence of a modified directional forgetting identification algorithm
• On parameter convergence in least squares identification and adaptive control

Cites work

• A globally convergent adaptive pole placement algorithm without a persistency of excitation requirement
• Dynamic system identification. Experiment design and data analysis
• Exponential convergence of recursive last squares with exponential forgetting factor
• Exponential stability of linear equations arising in adaptive identification
• Optimal finite-dimensional recursive identification in a polynomial output mapping class
• Persistency of excitation for nonminimal models of systems having purely deterministic disturbances
• Persistency of excitation in the presence of possibly unbounded signals
• Recursive estimation and time-series analysis. An introduction
• Robust adaptive control: Conditions for global stability
• Robustness of continuous-time adaptive control algorithms in the presence of unmodeled dynamics
• Theory and applications of adaptive control - a survey

Cited in

(13)

• A recursive parameter estimator yielding exponential convergence under sufficient excitation
• Necessary and sufficient regressor conditions for the global asymptotic stability of recursive least squares
• Identification problem of the exponential tilting estimator under misspecification
• On exponentially convergent parameter estimation with lack of persistency of excitation
• scientific article; zbMATH DE number 991575 (Why is no real title available?)
• scientific article; zbMATH DE number 3922504 (Why is no real title available?)
• On the consistency of a least squares identification procedure in linear evolution systems
• Estimation of convergence rate for robust identification algorithms
• scientific article; zbMATH DE number 522876 (Why is no real title available?)
• scientific article; zbMATH DE number 647688 (Why is no real title available?)
• A parametric approach for l/sub 1/ robust identification
• On parameter convergence in least squares identification and adaptive control
• Identifiability implies robust, globally exponentially convergent on-line parameter estimation

This page was built for publication: On exponential convergence and robustness of least-squares identification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1088967)