Comparison of different radial basis functions in dynamical systems

Argáez, Peter Giesl, Sigurdur Hafstein

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper we study the impact of using different radial basis functions for the computation of complete Lyapunov function candidates using generalised interpolation. We compare the numerical well-posedness of the discretised problem, condition numbers of the collocation matrices, and the quality of the solutions for Wendland functions ψ3,1 and ψ5,3, Gaussians, Inverse quadratics and Inverse multiquadrics, and Matérn kernels ψ(n+3)/2 and ψ(n+5)/2.

Original languageEnglish
Title of host publicationProceedings of the 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications, SIMULTECH 2021
EditorsGerd Wagner, Frank Werner, Tuncer Oren, Floriano De Rango
PublisherSciTePress
Pages394-405
Number of pages12
ISBN (Electronic)9789897585289
DOIs
Publication statusPublished - 1 Jan 2021
Event11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications, SIMULTECH 2021 - Virtual, Online
Duration: 7 Jul 20219 Jul 2021

Publication series

NameProceedings of the 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications, SIMULTECH 2021

Conference

Conference11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications, SIMULTECH 2021
CityVirtual, Online
Period7/07/219/07/21

Bibliographical note

Publisher Copyright:
Copyright © 2021 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved.

Other keywords

  • Complete Lyapunov Functions
  • Generalised Interpolation
  • Radial Basis Functions

Fingerprint

Dive into the research topics of 'Comparison of different radial basis functions in dynamical systems'. Together they form a unique fingerprint.

Cite this