Nonlinear system identification with regularized Tensor Network B-splines

Ridvan Karagoz, Kim Batselier

Research output: Contribution to journalArticleScientificpeer-review

6 Downloads (Pure)

Abstract

This article introduces the Tensor Network B-spline (TNBS) model for the regularized identification of nonlinear systems using a nonlinear autoregressive exogenous (NARX) approach. Tensor network theory is used to alleviate the curse of dimensionality of multivariate B-splines by representing the high-dimensional weight tensor as a low-rank approximation. An iterative algorithm based on the alternating linear scheme is developed to directly estimate the low-rank tensor network approximation, removing the need to ever explicitly construct the exponentially large weight tensor. This reduces the computational and storage complexity significantly, allowing the identification of NARX systems with a large number of inputs and lags. The proposed algorithm is numerically stable, robust to noise, guaranteed to monotonically converge, and allows the straightforward incorporation of regularization. The TNBS-NARX model is validated through the identification of the cascaded watertank benchmark nonlinear system, on which it achieves state-of-the-art performance while identifying a 16-dimensional B-spline surface in 4 s on a standard desktop computer. An open-source MATLAB implementation is available on GitHub.

Original languageEnglish
Article number109300
Number of pages9
JournalAutomatica
Volume122
DOIs
Publication statusPublished - 2020

Keywords

  • B-splines
  • Curse of dimensionality
  • NARX
  • Nonlinear system identification
  • Tensor network

Fingerprint Dive into the research topics of 'Nonlinear system identification with regularized Tensor Network B-splines'. Together they form a unique fingerprint.

Cite this