Accelerated Hierarchical Collision Detection for Simulation using CUDA

Jimmy Alison Jørgensen, Andreas Rune Fugl, Henrik Gordon Petersen

Research output: Contribution to conference without publisher/journalPaperResearchpeer-review

Abstract

In this article we present a GPU accelerated, hybrid, narrow phase collision detection algorithm for simulation
purposes. The algorithm is based on hierarchical bounding volume tree structures of oriented bounding boxes
(OBB) that in the past has shown to be efficient for collision detection.
The hierarchical nature of the bounding volume structure complicates an efficient implementation on massively
parallel architectures such as modern graphics cards and we therefore propose a hybrid method where only box
and triangle overlap tests and transformations are offloaded to the graphics card.
When exploiting coarse-grained parallelism in grasping and stacking simulations, requiring all-contacts resolu-
tion, a performance gain of up to 7x compared to the collision detection package PQP is obtained.
Original languageEnglish
Publication date2010
Number of pages8
DOIs
Publication statusPublished - 2010

Fingerprint

Graphics processing unit

Keywords

  • Collisiondetection
  • CUDA
  • Simulation

Cite this

@conference{9578b7fc73334a939dc975bf41d26ddd,
title = "Accelerated Hierarchical Collision Detection for Simulation using CUDA",
abstract = "In this article we present a GPU accelerated, hybrid, narrow phase collision detection algorithm for simulationpurposes. The algorithm is based on hierarchical bounding volume tree structures of oriented bounding boxes(OBB) that in the past has shown to be efficient for collision detection.The hierarchical nature of the bounding volume structure complicates an efficient implementation on massivelyparallel architectures such as modern graphics cards and we therefore propose a hybrid method where only boxand triangle overlap tests and transformations are offloaded to the graphics card.When exploiting coarse-grained parallelism in grasping and stacking simulations, requiring all-contacts resolu-tion, a performance gain of up to 7x compared to the collision detection package PQP is obtained.",
keywords = "Collisiondetection, CUDA, Simulation",
author = "J{\o}rgensen, {Jimmy Alison} and Fugl, {Andreas Rune} and Petersen, {Henrik Gordon}",
year = "2010",
doi = "10.2312/PE/vriphys/vriphys10/097-104",
language = "English",

}

Accelerated Hierarchical Collision Detection for Simulation using CUDA. / Jørgensen, Jimmy Alison; Fugl, Andreas Rune; Petersen, Henrik Gordon.

2010.

Research output: Contribution to conference without publisher/journalPaperResearchpeer-review

TY - CONF

T1 - Accelerated Hierarchical Collision Detection for Simulation using CUDA

AU - Jørgensen, Jimmy Alison

AU - Fugl, Andreas Rune

AU - Petersen, Henrik Gordon

PY - 2010

Y1 - 2010

N2 - In this article we present a GPU accelerated, hybrid, narrow phase collision detection algorithm for simulationpurposes. The algorithm is based on hierarchical bounding volume tree structures of oriented bounding boxes(OBB) that in the past has shown to be efficient for collision detection.The hierarchical nature of the bounding volume structure complicates an efficient implementation on massivelyparallel architectures such as modern graphics cards and we therefore propose a hybrid method where only boxand triangle overlap tests and transformations are offloaded to the graphics card.When exploiting coarse-grained parallelism in grasping and stacking simulations, requiring all-contacts resolu-tion, a performance gain of up to 7x compared to the collision detection package PQP is obtained.

AB - In this article we present a GPU accelerated, hybrid, narrow phase collision detection algorithm for simulationpurposes. The algorithm is based on hierarchical bounding volume tree structures of oriented bounding boxes(OBB) that in the past has shown to be efficient for collision detection.The hierarchical nature of the bounding volume structure complicates an efficient implementation on massivelyparallel architectures such as modern graphics cards and we therefore propose a hybrid method where only boxand triangle overlap tests and transformations are offloaded to the graphics card.When exploiting coarse-grained parallelism in grasping and stacking simulations, requiring all-contacts resolu-tion, a performance gain of up to 7x compared to the collision detection package PQP is obtained.

KW - Collisiondetection

KW - CUDA

KW - Simulation

U2 - 10.2312/PE/vriphys/vriphys10/097-104

DO - 10.2312/PE/vriphys/vriphys10/097-104

M3 - Paper

ER -