Optimization-Based Design of Bounded-Error Estimators Robust to Missing Data

Kwesi J. Rutledge, Sze Yong, Necmiye Ozay

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Non-asymptotic bounded-error state estimators that provide hard bounds on the estimation error are crucial for safety-critical applications. This paper proposes a class of optimal bounded-error affine estimators to achieve a novel property we are calling Equalized Recovery that can be computed by leveraging ideas from the dual problem of affine finite horizon optimal control design. In particular, by using Q-parametrization, the estimator design problem is reduced to a convex optimization problem. An extension of this estimator to handle missing data (e.g., due to package drops or sensor glitches) is also proposed. These ideas are illustrated with a numerical example motivated by vehicle safety systems.

Original languageEnglish (US)
Pages (from-to)157-162
Number of pages6
JournalIFAC-PapersOnLine
Volume51
Issue number16
DOIs
StatePublished - Jan 1 2018

Fingerprint

Convex optimization
Security systems
Error analysis
Recovery
Sensors

Keywords

  • bounded-error estimation
  • missing data
  • Robust estimators

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Optimization-Based Design of Bounded-Error Estimators Robust to Missing Data . / Rutledge, Kwesi J.; Yong, Sze; Ozay, Necmiye.

In: IFAC-PapersOnLine, Vol. 51, No. 16, 01.01.2018, p. 157-162.

Research output: Contribution to journalArticle

Rutledge, Kwesi J. ; Yong, Sze ; Ozay, Necmiye. / Optimization-Based Design of Bounded-Error Estimators Robust to Missing Data In: IFAC-PapersOnLine. 2018 ; Vol. 51, No. 16. pp. 157-162.
@article{131df98ca83f49d7a8ff8d3b08d12aad,
title = "Optimization-Based Design of Bounded-Error Estimators Robust to Missing Data⁎",
abstract = "Non-asymptotic bounded-error state estimators that provide hard bounds on the estimation error are crucial for safety-critical applications. This paper proposes a class of optimal bounded-error affine estimators to achieve a novel property we are calling Equalized Recovery that can be computed by leveraging ideas from the dual problem of affine finite horizon optimal control design. In particular, by using Q-parametrization, the estimator design problem is reduced to a convex optimization problem. An extension of this estimator to handle missing data (e.g., due to package drops or sensor glitches) is also proposed. These ideas are illustrated with a numerical example motivated by vehicle safety systems.",
keywords = "bounded-error estimation, missing data, Robust estimators",
author = "Rutledge, {Kwesi J.} and Sze Yong and Necmiye Ozay",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.ifacol.2018.08.027",
language = "English (US)",
volume = "51",
pages = "157--162",
journal = "IFAC-PapersOnLine",
issn = "2405-8963",
publisher = "IFAC Secretariat",
number = "16",

}

TY - JOUR

T1 - Optimization-Based Design of Bounded-Error Estimators Robust to Missing Data⁎

AU - Rutledge, Kwesi J.

AU - Yong, Sze

AU - Ozay, Necmiye

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Non-asymptotic bounded-error state estimators that provide hard bounds on the estimation error are crucial for safety-critical applications. This paper proposes a class of optimal bounded-error affine estimators to achieve a novel property we are calling Equalized Recovery that can be computed by leveraging ideas from the dual problem of affine finite horizon optimal control design. In particular, by using Q-parametrization, the estimator design problem is reduced to a convex optimization problem. An extension of this estimator to handle missing data (e.g., due to package drops or sensor glitches) is also proposed. These ideas are illustrated with a numerical example motivated by vehicle safety systems.

AB - Non-asymptotic bounded-error state estimators that provide hard bounds on the estimation error are crucial for safety-critical applications. This paper proposes a class of optimal bounded-error affine estimators to achieve a novel property we are calling Equalized Recovery that can be computed by leveraging ideas from the dual problem of affine finite horizon optimal control design. In particular, by using Q-parametrization, the estimator design problem is reduced to a convex optimization problem. An extension of this estimator to handle missing data (e.g., due to package drops or sensor glitches) is also proposed. These ideas are illustrated with a numerical example motivated by vehicle safety systems.

KW - bounded-error estimation

KW - missing data

KW - Robust estimators

UR - http://www.scopus.com/inward/record.url?scp=85052621674&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052621674&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2018.08.027

DO - 10.1016/j.ifacol.2018.08.027

M3 - Article

AN - SCOPUS:85052621674

VL - 51

SP - 157

EP - 162

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8963

IS - 16

ER -