A Continuous Representation Of Switching Linear Dynamic Systems For Accurate Tracking

Parisa Karimi, Helmuth Naumer, Farzad Kamalabadi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose a method for tracking linear representations of a nonlinear dynamic system with time-varying parameters based on a continuous representation of its switching linear dynamic system (SLDS) model. Given approximate linear representations for a finite set of unknown intrinsic parameters of the dynamics, a combination of autoencoder-based dimensionality reduction and cubic curve-fitting are applied to learn the continuous manifold of dynamics embedded in the evolution operator. This representation enables a significant reduction of the squared Frobenius norm of error in maximum likelihood (ML) system identification relative to that of the original SLDS model. Numerical experiments also verify this result.

Original languageEnglish (US)
Title of host publicationProceedings of the 22nd IEEE Statistical Signal Processing Workshop, SSP 2023
PublisherIEEE Computer Society
Pages339-343
Number of pages5
ISBN (Electronic)9781665452458
DOIs
StatePublished - 2023
Event22nd IEEE Statistical Signal Processing Workshop, SSP 2023 - Hanoi, Viet Nam
Duration: Jul 2 2023Jul 5 2023

Publication series

NameIEEE Workshop on Statistical Signal Processing Proceedings
Volume2023-July

Conference

Conference22nd IEEE Statistical Signal Processing Workshop, SSP 2023
Country/TerritoryViet Nam
CityHanoi
Period7/2/237/5/23

Keywords

  • Dynamic systems
  • Koopman operator
  • Manifold learning
  • Online tracking
  • Switching linear dynamic system
  • Variational autoencoder

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Applied Mathematics
  • Signal Processing
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'A Continuous Representation Of Switching Linear Dynamic Systems For Accurate Tracking'. Together they form a unique fingerprint.

Cite this