Performance limitations of adaptive parameter estimation and system identification algorithms in the absence of excitation

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

We address the problem of fundamental performance limitations in adaptive parameter estimation and system identification, occurring in environments where perturbations are present but there is lack of sufficient excitation. We construct a simple yet general bursting scenario to derive an analytical lower bound on the worst-case peak steady-state error for a wide class of parameter estimation and system identification algorithms. Our results show that in the absence of any input constraints, arbitrarily small perturbations impose a serious performance limitation, in the sense that the worst-case performance deteriorates proportionally with the size of the parametric uncertainty set.

Original languageEnglish (US)
Pages (from-to)549-560
Number of pages12
JournalAutomatica
Volume32
Issue number4
DOIs
StatePublished - Apr 1996

Fingerprint

Parameter estimation
Identification (control systems)
Uncertainty

Keywords

  • Burst phenomena
  • Parameter estimation algorithms
  • Performance limits
  • Persistent excitation
  • System identification algorithms
  • Time-varying systems

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

@article{95ead51a69aa42bd8ffabbeb73412645,
title = "Performance limitations of adaptive parameter estimation and system identification algorithms in the absence of excitation",
abstract = "We address the problem of fundamental performance limitations in adaptive parameter estimation and system identification, occurring in environments where perturbations are present but there is lack of sufficient excitation. We construct a simple yet general bursting scenario to derive an analytical lower bound on the worst-case peak steady-state error for a wide class of parameter estimation and system identification algorithms. Our results show that in the absence of any input constraints, arbitrarily small perturbations impose a serious performance limitation, in the sense that the worst-case performance deteriorates proportionally with the size of the parametric uncertainty set.",
keywords = "Burst phenomena, Parameter estimation algorithms, Performance limits, Persistent excitation, System identification algorithms, Time-varying systems",
author = "Konstantinos Tsakalis",
year = "1996",
month = "4",
doi = "10.1016/0005-1098(95)00163-8",
language = "English (US)",
volume = "32",
pages = "549--560",
journal = "Automatica",
issn = "0005-1098",
publisher = "Elsevier Limited",
number = "4",

}

TY - JOUR

T1 - Performance limitations of adaptive parameter estimation and system identification algorithms in the absence of excitation

AU - Tsakalis, Konstantinos

PY - 1996/4

Y1 - 1996/4

N2 - We address the problem of fundamental performance limitations in adaptive parameter estimation and system identification, occurring in environments where perturbations are present but there is lack of sufficient excitation. We construct a simple yet general bursting scenario to derive an analytical lower bound on the worst-case peak steady-state error for a wide class of parameter estimation and system identification algorithms. Our results show that in the absence of any input constraints, arbitrarily small perturbations impose a serious performance limitation, in the sense that the worst-case performance deteriorates proportionally with the size of the parametric uncertainty set.

AB - We address the problem of fundamental performance limitations in adaptive parameter estimation and system identification, occurring in environments where perturbations are present but there is lack of sufficient excitation. We construct a simple yet general bursting scenario to derive an analytical lower bound on the worst-case peak steady-state error for a wide class of parameter estimation and system identification algorithms. Our results show that in the absence of any input constraints, arbitrarily small perturbations impose a serious performance limitation, in the sense that the worst-case performance deteriorates proportionally with the size of the parametric uncertainty set.

KW - Burst phenomena

KW - Parameter estimation algorithms

KW - Performance limits

KW - Persistent excitation

KW - System identification algorithms

KW - Time-varying systems

UR - http://www.scopus.com/inward/record.url?scp=0030126072&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030126072&partnerID=8YFLogxK

U2 - 10.1016/0005-1098(95)00163-8

DO - 10.1016/0005-1098(95)00163-8

M3 - Article

AN - SCOPUS:0030126072

VL - 32

SP - 549

EP - 560

JO - Automatica

JF - Automatica

SN - 0005-1098

IS - 4

ER -