Randomised subspace system identification: complexities and error bounds

Recently, a novel randomised subspace system identification method (RandSID) for estimating LTI state-space models was proposed in Kedia, V., & Chakraborty, D. (2023), which aimed to address the computational issues faced by traditional subspace identification methods when the recorded dataset i...

Full description

Saved in:
Bibliographic Details
Published in:International journal of control Vol. 98; no. 12; pp. 2835 - 2846
Main Authors: Kedia, Vatsal, Chakraborty, Debraj
Format: Journal Article
Language:English
Published: Abingdon Taylor & Francis 02.12.2025
Taylor & Francis Ltd
Subjects:
Algorithms
Identification for Control
Identification methods
Randomised algorithms
Randomization
State-space modeling
Subspace Identification
Subspaces
System Identification
ISSN:0020-7179, 1366-5820
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recently, a novel randomised subspace system identification method (RandSID) for estimating LTI state-space models was proposed in Kedia, V., & Chakraborty, D. (2023), which aimed to address the computational issues faced by traditional subspace identification methods when the recorded dataset is large. In this article, we propose a modified version of RandSID and analyse the complexity of this proposed algorithm in terms of memory cost, data movement, flop count and computation time, and quantitatively establish the computational advantages over conventional algorithms. Further, we derive a bound on the ratio of mean squared errors for the proposed and conventional algorithms, thereby characterising the loss of accuracy specifically due to randomisation. The effectiveness of the proposed algorithm is established by various real and simulated case studies.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0020-7179
1366-5820
DOI:10.1080/00207179.2025.2485169