Real-time CUDA based collision detection and physics based collision response simulation

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

Abstract

To improve the physical realism and guarantee the simulation efficiency in virtual surgery, we propose a real-time method to detect collisions and conduct collision response simulation based on an improved physical model. To guarantee the real-time efficiency, we design CUDA accelerated collision detection algorithm based on space subdivision, and calculate the intrusion depth of collision units to further reduce computation cost in the narrow phase. We improve the physical realism by introducing geometry information into penalty force calculation to better satisfy the Newton's third law, which can simultaneously retain the advantages of the punishment force method, such as easy calculation, high response speed, etc. We demonstrate the superiority of our method by performing extensive comparison with punishment force method and state-of-the-art commercial software ABAQUS.

Original languageEnglish
Title of host publicationProceedings of the 2012 4th International Symposium on Information Science and Engineering, ISISE 2012
Pages250-254
Number of pages5
DOIs
StatePublished - 2012
Event2012 4th International Symposium on Information Science and Engineering, ISISE 2012 - Shanghai, China
Duration: 14 Dec 201216 Dec 2012

Publication series

NameProceedings of the 2012 4th International Symposium on Information Science and Engineering, ISISE 2012

Conference

Conference2012 4th International Symposium on Information Science and Engineering, ISISE 2012
Country/TerritoryChina
CityShanghai
Period14/12/1216/12/12

Keywords

  • CUDA
  • collision detection
  • physics based collision response model
  • space subdivision

Fingerprint

Dive into the research topics of 'Real-time CUDA based collision detection and physics based collision response simulation'. Together they form a unique fingerprint.

Cite this