Operator-theoretic characterization of eventually monotone systems

Aivar Sootla, Alexandre Mauroy

Research output: Contribution to journalArticle

Abstract

Eventually monotone systems are dynamical systems whose solutions preserve a partial order in the initial condition after some initial transient. While monotone systems have a characterization in terms of their vector fields, eventually monotone systems have not been characterized in such an explicit manner. In order to provide a characterization, we drew inspiration from the results for linear systems, where eventually monotone (positive) systems are studied using the spectral properties of the system. We extend this spectral characterization to nonlinear systems by employing the Koopman operator framework. We also present a method to certify strong eventual monotonicity with respect to an unknown cone, a tool which to our best knowledge does not exist for monotone systems. These results are illustrated on biologically inspired numerical examples, which highlight the potential applicability of eventual monotonicity.

LanguageEnglish
Pages429-434
Number of pages6
JournalControl Systems Letters (L-CSS)
Volume2
Issue number3
DOIs
Publication statusPublished - 1 Jul 2018

Fingerprint

Monotone Systems
Operator
Monotonicity
Linear systems
Cones
Nonlinear systems
Dynamical systems
Positive Systems
Partial Order
Spectral Properties
Vector Field
Initial conditions
Cone
Nonlinear Systems
Dynamical system
Linear Systems
Unknown
Numerical Examples

Keywords

  • biological systems
  • Koopman operator
  • Monotone systems

Cite this

@article{80c4911a950e4af3806d9c403d038b7f,
title = "Operator-theoretic characterization of eventually monotone systems",
abstract = "Eventually monotone systems are dynamical systems whose solutions preserve a partial order in the initial condition after some initial transient. While monotone systems have a characterization in terms of their vector fields, eventually monotone systems have not been characterized in such an explicit manner. In order to provide a characterization, we drew inspiration from the results for linear systems, where eventually monotone (positive) systems are studied using the spectral properties of the system. We extend this spectral characterization to nonlinear systems by employing the Koopman operator framework. We also present a method to certify strong eventual monotonicity with respect to an unknown cone, a tool which to our best knowledge does not exist for monotone systems. These results are illustrated on biologically inspired numerical examples, which highlight the potential applicability of eventual monotonicity.",
keywords = "biological systems, Koopman operator, Monotone systems",
author = "Aivar Sootla and Alexandre Mauroy",
year = "2018",
month = "7",
day = "1",
doi = "10.1109/LCSYS.2018.2841654",
language = "English",
volume = "2",
pages = "429--434",
journal = "Control Systems Letters (L-CSS)",
issn = "2475-1456",
publisher = "IEEE",
number = "3",

}

Operator-theoretic characterization of eventually monotone systems. / Sootla, Aivar; Mauroy, Alexandre.

In: Control Systems Letters (L-CSS), Vol. 2, No. 3, 01.07.2018, p. 429-434.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Operator-theoretic characterization of eventually monotone systems

AU - Sootla, Aivar

AU - Mauroy, Alexandre

PY - 2018/7/1

Y1 - 2018/7/1

N2 - Eventually monotone systems are dynamical systems whose solutions preserve a partial order in the initial condition after some initial transient. While monotone systems have a characterization in terms of their vector fields, eventually monotone systems have not been characterized in such an explicit manner. In order to provide a characterization, we drew inspiration from the results for linear systems, where eventually monotone (positive) systems are studied using the spectral properties of the system. We extend this spectral characterization to nonlinear systems by employing the Koopman operator framework. We also present a method to certify strong eventual monotonicity with respect to an unknown cone, a tool which to our best knowledge does not exist for monotone systems. These results are illustrated on biologically inspired numerical examples, which highlight the potential applicability of eventual monotonicity.

AB - Eventually monotone systems are dynamical systems whose solutions preserve a partial order in the initial condition after some initial transient. While monotone systems have a characterization in terms of their vector fields, eventually monotone systems have not been characterized in such an explicit manner. In order to provide a characterization, we drew inspiration from the results for linear systems, where eventually monotone (positive) systems are studied using the spectral properties of the system. We extend this spectral characterization to nonlinear systems by employing the Koopman operator framework. We also present a method to certify strong eventual monotonicity with respect to an unknown cone, a tool which to our best knowledge does not exist for monotone systems. These results are illustrated on biologically inspired numerical examples, which highlight the potential applicability of eventual monotonicity.

KW - biological systems

KW - Koopman operator

KW - Monotone systems

UR - https://ieeexplore.ieee.org/document/8368266/

UR - http://www.mendeley.com/research/operatortheoretic-characterization-eventually-monotone-systems

UR - http://www.scopus.com/inward/record.url?scp=85057646620&partnerID=8YFLogxK

U2 - 10.1109/LCSYS.2018.2841654

DO - 10.1109/LCSYS.2018.2841654

M3 - Article

VL - 2

SP - 429

EP - 434

JO - Control Systems Letters (L-CSS)

T2 - Control Systems Letters (L-CSS)

JF - Control Systems Letters (L-CSS)

SN - 2475-1456

IS - 3

ER -