系统仿真学报 ›› 2017, Vol. 29 ›› Issue (11): 2601-2608.doi: 10.16182/j.issn1004731x.joss.201711001

所属专题: 特约稿件

• 综述 •    下一篇

并行化碰撞检测算法综述

刘复昌, 王双建, 潘志庚, 王金荣   

  1. 杭州师范大学数字媒体与人机交互研究中心,浙江 杭州 310012
  • 收稿日期:2016-05-12 发布日期:2020-06-05
  • 作者简介:刘复昌(1982-),男,南京,博士,讲师,研究方向为图形图像。
  • 基金资助:
    国家自然基金青年科学基金(61502133),浙江省自然科学基金一般项目(LY16F020029),浙江省教育厅科研项目(Y201017442)

Survey on Parallel Collision Detection Algorithms

Liu Fuchang, Wang Shuangjian, Pan Zhigeng, Wang Jinrong   

  1. Digital Media and HCI Research Center, Hangzhou Normal University, Hangzhou 310012, China
  • Received:2016-05-12 Published:2020-06-05

摘要: 随着不同应用领域对实时碰撞检测算法需求的增长,利用多核CPU和GPU的并行处理能力来提高碰撞检测算法的处理速度已经得到了广泛的关注。文中回顾了碰撞检测算法的发展历史并从多个角度对目前现有的算法进行了分类归纳;介绍了十余种代表性的基于CPU和GPU并行化碰撞检测算法,并从算法的可扩展性和存储空间消耗以及任务量均衡化等方面分析了这些算法的优缺点。最后总结了并行化碰撞检测算法研究中存在的问题和新的发展方向以及常用的实验测试数据。

关键词: 碰撞检测, 综述, GPU, 并行化算法

Abstract: The demand for real-time collision detection is increasing in different applications. Exploiting the parallel computing capability of multi-core CPUs and GPUs to accelerate the speed of collision detection algorithms has attracted abroad attention. This paper reviews the development history of collision detection algorithms and classified the existing algorithms from multiple perspectives. Moreover, we analyze the strengths and weaknesses of more than ten representative parallel collision detection algorithms based on multi-core CPUs and GPUs from the aspects of the scalability, memory consumption and workload balancing. Finally, the problem of present parallel collision detection research and potential direction of following research and some representative benchmark data sets are presented.

Key words: collision detection, survey, GPU, parallel algorithms

中图分类号: