Literatur vom gleichen Autor/der gleichen Autor*in
plus bei Google Scholar

Bibliografische Daten exportieren
 

Computing reachable sets via barrier methods on SIMD architectures

Titelangaben

Grüne, Lars ; Jahn, Thomas U.:
Computing reachable sets via barrier methods on SIMD architectures.
In: Eberhardsteiner, Josef ; Böhm, Helmut J. ; Rammerstorfer, Franz G. (eds.): Proceedings of the 6th European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS 2012) Held at the University of Vienna, Austria, September 10-14, 2012. - Vienna : Vienna University of Technology , 2012 . - S. 2076-2095
ISBN 9783950248197

Dies ist die aktuelle Version des Eintrags.

Weitere URLs

Abstract

We consider the problem of computing reachable sets of ODE-based control systems parallely on CUDA hardware. To this end, we modify an existing algorithm based on solving optimal control problems.
The idea is to simplify the optimal control problems to pure feasibility problems instead of minimizing an objective function. We show that an interior point algorithm is well suited for solving the resulting feasibility problems and leads to a sequence of linear systems of equations with identical matrix layout. If the problem is defined properly, these matrices are sparse and can be transformed into a hierarchical lower arrow form which can be solved on CUDA with sparse linear algebra and Cholesky’s method.
We demonstrate the performance of our new algorithm by computing the reachable sets of two test problems on a CPU implementation using several explicit and implicit Runge-Kutta methods of different order. The experiments reveal a significant speedup compared to the original optimal control algorithm.

Weitere Angaben

Publikationsform: Aufsatz in einem Buch
Begutachteter Beitrag: Ja
Zusätzliche Informationen: Contents:
1. Introduction
2. Principles of SIMD architectures
2.1 SIMD and thread enumeration
2.2 Memory considerations
2.3 Suitable algorithms
3. Algorithm Specification
3.1 The approach of Baier and Gerdts
3.2 An algorithm for computing reachable sets
3.3 Distributing the algorithm to the CUDA hardware
4. Solving the feasibility problem
4.1 The interior–point algorithm
4.2 Defining the restrictions
4.3 Exploiting sparsity
5. Numerical examples
5.1 Rayleigh
5.2 Kenderov
6. Conclusions
Keywords: reachable set; feasibility problem; sparse linear equation system; Runge-Kutta method; CUDA; parallelization; lower arrow form
Institutionen der Universität: Fakultäten
Fakultäten > Fakultät für Mathematik, Physik und Informatik
Fakultäten > Fakultät für Mathematik, Physik und Informatik > Mathematisches Institut
Fakultäten > Fakultät für Mathematik, Physik und Informatik > Mathematisches Institut > Lehrstuhl Mathematik V (Angewandte Mathematik) > Lehrstuhl Mathematik V (Angewandte Mathematik) - Univ.-Prof. Dr. Lars Grüne
Profilfelder
Profilfelder > Advanced Fields
Profilfelder > Advanced Fields > Nichtlineare Dynamik
Fakultäten > Fakultät für Mathematik, Physik und Informatik > Mathematisches Institut > Lehrstuhl Mathematik V (Angewandte Mathematik)
Titel an der UBT entstanden: Ja
Themengebiete aus DDC: 500 Naturwissenschaften und Mathematik > 510 Mathematik
Eingestellt am: 01 Apr 2015 06:32
Letzte Änderung: 09 Jan 2024 13:19
URI: https://eref.uni-bayreuth.de/id/eprint/9544

Zu diesem Eintrag verfügbare Versionen