Kernel-Based Models for System Analysis

DOI10.1109/TAC.2022.3218944MaRDI QIDQ6047030zbMATH OpenOpenAlexFDO

Authors Henk J. van Waarde, Rodolphe Sepulchre

Publication date 9 October 2023

Published in IEEE Transactions on Automatic Control (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2110.11735

zbMATH Keywords

machine learningsystem identificationmodelingnonlinear systemsidentification for control

Mathematics Subject Classification ID

Systems theory; control (93-XX)

Abstract: This paper introduces a computational framework to identify nonlinear input-output operators that fit a set of system trajectories while satisfying incremental integral quadratic constraints. The data fitting algorithm is thus regularized by suitable input-output properties required for system analysis and control design. This biased identification problem is shown to admit the tractable solution of a regularized least squares problem when formulated in a suitable reproducing kernel Hilbert space. The kernel-based framework is a departure from the prevailing state-space framework. It is motivated by fundamental limitations of nonlinear state-space models at combining the fitting requirements of data-based modeling with the input-output requirements of system analysis and physical modeling.

Recommendations

• scientific article; zbMATH DE number 6698762
• Kernel-based methods for parameter estimation in multidimensional systems
• Kernel methods for the approximation of nonlinear systems
• scientific article; zbMATH DE number 1984188
• Nonparametric system identification by kernel methods
• Kernel model-based diagnosis
• A new kernel-based approach for linear system identification
• Kernels for linear time invariant system identification
• Kernel methods in system identification, machine learning and function estimation: a survey
• 9 Kernel methods for surrogate modeling

Cited in

(5)

• scientific article; zbMATH DE number 6698762 (Why is no real title available?)
• Kernel-like Search for Robust Emergency System Designing
• Data-efficient global extremum seeking framework for constrained optimization of nonlinear dynamical systems
• Learning Hamiltonian dynamics with reproducing kernel Hilbert spaces and random features
• A graph learning approach for kernel-based system identification with manifold regularization

This page was built for publication: Kernel-Based Models for System Analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6047030)