Distance Measures for Strong Observability and Strong Detectability of Systems with Direct Feedthrough

Roland Falkensteiner; Richard Seeber; Markus Reichhartinger; Martin Horn

doi:10.1109/LCSYS.2020.3004279

Distance Measures for Strong Observability and Strong Detectability of Systems with Direct Feedthrough

Roland Falkensteiner^*, Richard Seeber, Markus Reichhartinger, Martin Horn

^*Corresponding author for this work

Institute of Automation and Control (4430)

Research output: Contribution to journal › Article › peer-review

Abstract

This letter presents distance measures for strong observability and strong detectability of linear, time-invariant systems with direct feedthrough. In contrast to standard binary criteria, the proposed measures allow to quantify these properties numerically. An optimization based approach for calculating these measures is derived. Meaningful conditions for such measures are introduced and the properties of the proposed measures regarding these conditions are analyzed. In order to demonstrate their usefulness, the measures are applied to optimal sensor placement for reconstruction of unknown forces on a mechanical structure.

Original language	English
Article number	9122594
Pages (from-to)	565-570
Number of pages	6
Journal	IEEE Control Systems Letters
Volume	5
Issue number	2
DOIs	https://doi.org/10.1109/LCSYS.2020.3004279
Publication status	Published - 2021

Keywords

Linear systems
Observers for linear systems
observers for linear systems

ASJC Scopus subject areas

Control and Optimization
Control and Systems Engineering

Access to Document

10.1109/LCSYS.2020.3004279

https://ieeexplore.ieee.org/document/9122594/

Cite this

@article{d891a6a854be45b78b5b6669a09a0276,

title = "Distance Measures for Strong Observability and Strong Detectability of Systems with Direct Feedthrough",

abstract = "This letter presents distance measures for strong observability and strong detectability of linear, time-invariant systems with direct feedthrough. In contrast to standard binary criteria, the proposed measures allow to quantify these properties numerically. An optimization based approach for calculating these measures is derived. Meaningful conditions for such measures are introduced and the properties of the proposed measures regarding these conditions are analyzed. In order to demonstrate their usefulness, the measures are applied to optimal sensor placement for reconstruction of unknown forces on a mechanical structure.",

keywords = "Linear systems, Observers for linear systems, observers for linear systems",

author = "Roland Falkensteiner and Richard Seeber and Markus Reichhartinger and Martin Horn",

year = "2021",

doi = "10.1109/LCSYS.2020.3004279",

language = "English",

volume = "5",

pages = "565--570",

journal = "IEEE Control Systems Letters",

issn = "2475-1456",

publisher = "IEEE Control Systems Society",

number = "2",

}

TY - JOUR

T1 - Distance Measures for Strong Observability and Strong Detectability of Systems with Direct Feedthrough

AU - Falkensteiner, Roland

AU - Seeber, Richard

AU - Reichhartinger, Markus

AU - Horn, Martin

PY - 2021

Y1 - 2021

N2 - This letter presents distance measures for strong observability and strong detectability of linear, time-invariant systems with direct feedthrough. In contrast to standard binary criteria, the proposed measures allow to quantify these properties numerically. An optimization based approach for calculating these measures is derived. Meaningful conditions for such measures are introduced and the properties of the proposed measures regarding these conditions are analyzed. In order to demonstrate their usefulness, the measures are applied to optimal sensor placement for reconstruction of unknown forces on a mechanical structure.

AB - This letter presents distance measures for strong observability and strong detectability of linear, time-invariant systems with direct feedthrough. In contrast to standard binary criteria, the proposed measures allow to quantify these properties numerically. An optimization based approach for calculating these measures is derived. Meaningful conditions for such measures are introduced and the properties of the proposed measures regarding these conditions are analyzed. In order to demonstrate their usefulness, the measures are applied to optimal sensor placement for reconstruction of unknown forces on a mechanical structure.

KW - Linear systems

KW - Observers for linear systems

KW - observers for linear systems

UR - http://www.scopus.com/inward/record.url?scp=85089555320&partnerID=8YFLogxK

U2 - 10.1109/LCSYS.2020.3004279

DO - 10.1109/LCSYS.2020.3004279

M3 - Article

SN - 2475-1456

VL - 5

SP - 565

EP - 570

JO - IEEE Control Systems Letters

JF - IEEE Control Systems Letters

IS - 2

M1 - 9122594

ER -

Distance Measures for Strong Observability and Strong Detectability of Systems with Direct Feedthrough

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this