The block regularised parameter estimator and its parallelisation

DOI10.1016/0005-1098(95)00015-OMaRDI QIDQ1899597zbMATH OpenOpenAlexFDO

Authors Jiří Kadlec, F. M. F. Gaston, George W. Irwin

Publication date 14 May 1996

Published in Automatica (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/0005-1098(95)00015-o

zbMATH Keywords

adaptive controlestimationparameter estimatorsystolic architecture

Mathematics Subject Classification ID

Parallel numerical computation (65Y05) System identification (93B30) Control/observation systems involving computers (process control, etc.) (93C83)

Review text

A new block regularised parameter estimator that is compatible with the requirements for implementation on a systolic architecture is presented. The array provides either the prediction error or the filtration error. It is shown that the processor utilisation of the proposed architecture is close to 100\%, that there are no global connections, and that data take only one direction in any given link. The proposed parameter estimator offers significant potential for adaptive control applications, particularly in the real-time domain.

Recommendations

• scientific article; zbMATH DE number 1079095
• A revisit to block and recursive least squares for parameter estimation
• scientific article; zbMATH DE number 1863311
• Parallel block coordinate minimization with application to group regularized regression
• A parallel implementation for parameter estimation with implicit models
• Parallel maximum likelihood estimator for multiple linear regression models
• Parallel restricted maximum likelihood estimation for linear models with a dense exogenous matrix
• Parallel random block-coordinate forward-backward algorithm: a unified convergence analysis
• Estimators for Gaussian models having a block-wise structure
• On shrinkage least squares estimation in a parallelism problem

Cites work

• Adaptation and tracking in system identification - a survey
• Bayesian system identification
• Factorization methods for discrete sequential estimation
• scientific article; zbMATH DE number 3875113 (Why is no real title available?)
• scientific article; zbMATH DE number 4215267 (Why is no real title available?)
• scientific article; zbMATH DE number 125161 (Why is no real title available?)
• On a general concept of forgetting
• Recursive forgetting algorithms
• Restricted exponential forgetting in real-time identification
• Systolic approach to square root information Kalman filtering
• Transputer implementation of Kalman filters

Cited in

(7)

• A fast lattice algorithm for PARCOR parameter estimation based on an approximate likelihood
• scientific article; zbMATH DE number 1014741 (Why is no real title available?)
• scientific article; zbMATH DE number 1079095 (Why is no real title available?)
• Contribution of Václav Peterka to adaptive control theory and practice
• Efficient block boundaries estimation in block-wise constant matrices: an application to HiC data
• Parallel restricted maximum likelihood estimation for linear models with a dense exogenous matrix
• Computation of Huber's \(M\)-estimates for a block-angular regression problem

This page was built for publication: The block regularised parameter estimator and its parallelisation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1899597)