Litcius/Paper detail

Hierarchical Recursive Gradient Parameter Identification for Multi‐Input ARX Systems With Partially‐Coupled Information Vectors

Feng Ding, Xiaoli Luan, Ling Xu, Zhang Xiao

2025International Journal of Adaptive Control and Signal Processing56 citationsDOI

Abstract

ABSTRACT The research object of coupling identification is for multivariate systems. It is required to study and explore recursive and iterative identification methods for multivariable systems when there exists information coupling and/or parameter coupling between their subsystems. For a multivariable system, namely a multiple‐input multiple‐output system, after parameterization, its identification model contains the same information vector in all its subsystems. For a multi‐input ARX system where there exists the information vector coupling, this paper derives the coupled identification model and investigates recursive parameter identification methods for such partially‐coupled information vector systems, and presents a hierarchical recursive gradient identification algorithm and a hierarchical multi‐innovation recursive gradient identification algorithm. Finally, the simulation example is provided to show the effectiveness of the proposed algorithms.

Topics & Concepts

Multivariable calculusIdentification (biology)Computer scienceCoupling (piping)Control theory (sociology)Multivariate statisticsSystem identificationAlgorithmMathematicsArtificial intelligenceData miningControl engineeringMachine learningEngineeringBotanyBiologyMeasure (data warehouse)Control (management)Mechanical engineeringControl Systems and IdentificationFault Detection and Control SystemsAdvanced Control Systems Optimization
Hierarchical Recursive Gradient Parameter Identification for Multi‐Input ARX Systems With Partially‐Coupled Information Vectors | Litcius