Enclosing temporal evolution of dynamical systems using numerical methods

Olivier Bouissou, Alexandre Chapoutot, Adel Djoudi

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

Abstract

Numerical methods are necessary to understand the behaviors of complex hybrid systems used to design control-command systems. Especially, numerical integration methods are heavily used in simulation to compute approximations of the solution of differential equations, including non-linear and stiff solutions. Nevertheless, these methods only produce approximate results and they should not be used in formal verification methods as is. We propose a systematic way to make explicit Runge-Kutta integration method safe with respect to the mathematical solution. As side effect, we can hence compare different integration schemes in order to pick the right one in different situations.

Original languageEnglish
Title of host publicationNASA Formal Methods - 5th International Symposium, NFM 2013, Proceedings
Pages108-123
Number of pages16
DOIs
Publication statusPublished - 9 Sept 2013
Event5th International Symposium on NASA Formal Methods, NFM 2013 - Moffett Field, CA, United States
Duration: 14 May 201316 May 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7871 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference5th International Symposium on NASA Formal Methods, NFM 2013
Country/TerritoryUnited States
CityMoffett Field, CA
Period14/05/1316/05/13

Fingerprint

Dive into the research topics of 'Enclosing temporal evolution of dynamical systems using numerical methods'. Together they form a unique fingerprint.

Cite this