Controlled perturbation for certified geometric computing with fixed-precision arithmetic

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

12 Scopus citations

Abstract

Transforming geometric algorithms into effective computer programs is a difficult task. This transformation is particularly made hard by the basic assumptions of most theoretical geometric algorithms concerning the handling of robustness issues, namely issues related to arithmetic precision and degenerate input. Controlled perturbation, an approach to robust implementation of geometric algorithms we introduced in the late 1990's, aims at removing degeneracies and certifying correct predicate-evaluation, while using fixed-precision arithmetic. After exposing the key ideas underlying the scheme, we review the development of the approach over the past decade including variations and extensions, software implementation and applications. We conclude by pointing out directions for further development and major challenges.

Original languageEnglish
Title of host publicationMathematical Software, ICMS 2010 - Third International Congress on Mathematical Software, Proceedings
Pages92-95
Number of pages4
DOIs
StatePublished - 2010
Event3rd International Congress on Mathematical Software, ICMS 2010 - Kobe, Japan
Duration: 13 Sep 201017 Sep 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6327 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference3rd International Congress on Mathematical Software, ICMS 2010
Country/TerritoryJapan
CityKobe
Period13/09/1017/09/10

Funding

FundersFunder number
German-Israeli Foundation for Scientific Research and Development969/07
Israel Science Foundation236/06
Tel Aviv University

    Fingerprint

    Dive into the research topics of 'Controlled perturbation for certified geometric computing with fixed-precision arithmetic'. Together they form a unique fingerprint.

    Cite this