Analytical and numerical estimates of reachable sets in a subdivision scheme

DOI zum Zitieren der Version auf EPub Bayreuth: https://doi.org/10.15495/EPub_UBT_00005055
URN to cite this document: urn:nbn:de:bvb:703-epub-5055-3

Title data

Riedl, Wolfgang ; Baier, Robert ; Gerdts, Matthias:
Analytical and numerical estimates of reachable sets in a subdivision scheme.
Mathematisches Institut, Universität Bayreuth, Institut für Mathematik und Rechneranwendung, Universität der Bundeswehr in Neubiberg/München
Bayreuth, Neubiberg/München , 2017 . - 15 S.

Format:	PDF
Name:	Riedl_Baier_Gerdts_analytical_numerical_estim_reach_sets_in_subdiv_scheme_2017_preprint.pdf
Version:	Preprint
Available under License	German copyright law. This contribution is freely accessible with the consent of the copyright holder on the basis of an (DFG-funded) alliance or national license. The document may be used free of charge for personal use. In addition, the reproduction, editing, distribution and any kind of exploitation require the written consent of the respective rights holder.

Download (1MB)

Abstract

Reachable sets for (discrete) nonlinear control problems can be described by feasible sets of nonlinear optimization problems. The objective function for this problem is set to minimize the distance from an arbitrary grid point of a bounding box to the reachable set. To avoid the high computational costs of starting the optimizer for all points in an equidistant grid, an adaptive version based on the subdivision framework known in the computation of attractors and invariant measures is studied. The generated box collections provide over-approximations which shrink to the reachable set for a decreasing maximal diameter of the boxes in the collection and, if the bounding box is too pessimistic, do not lead to an exploding number of boxes as examples show. Analytical approaches for the bounding box of a 3d funnel are gained via the Gronwall-Filippov-Wazewski theorem for differential inclusions or by choosing good reference solutions. An alternative self-finding algorithm for the bounding box is applied to a higher-dimensional kinematic car model.

Further data

Item Type:	Preprint, postprint
Additional notes (visible to public):	Contents: 1. Introduction 1.1 Reachability analysis 1.2 Preliminaries 1.3 Control problems and differential inclusions 1.4 Direct discretization via set-valued Runge-Kutta methods 2. Subdivision Algorithm for Reachable Sets and Its Convergence 2.1 Non-adaptive and adaptive algorithm 2.2 Convergence study 3. Analytical and Numerical Calculation of Bounding Boxes 3.1 Analytical approach 3.2 Numerical approach 4. Examples 4.1 Kenderov’s example 4.2 Car model 5. Conclusions
Keywords:	reachable sets; subdivision; direct discretization of optimal control; Filippov's theorem; nonlinear optimization
Subject classification:	Mathematics Subject Classification Code: 93B03 34A60 (49M25 49J53 65L07 93D23 93C10)
DDC Subjects:	500 Science > 510 Mathematics
Institutions of the University:	Faculties Faculties > Faculty of Mathematics, Physics und Computer Science Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics > Chair Mathematics V (Applied Mathematics) Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics > Chair Mathematics V (Applied Mathematics) > Chair Mathematics V (Applied Mathematics) - Univ.-Prof. Dr. Lars Grüne Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics > Chair Scientific Computing Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics > Chair Scientific Computing > Chair Scientific Computing - Univ.-Prof. Dr. Mario Bebendorf Profile Fields Profile Fields > Advanced Fields Profile Fields > Advanced Fields > Nonlinear Dynamics
Language:	English
Originates at UBT:	Yes
URN:	urn:nbn:de:bvb:703-epub-5055-3
Date Deposited:	15 Sep 2020 07:14
Last Modified:	15 Sep 2020 07:14
URI:	https://epub.uni-bayreuth.de/id/eprint/5055

Download Statistics

Downloads

Downloads per month over past year