基于GPU的并行化人群实时仿真  

GPU-BASED PARALLELISED REAL-TIME CROWD SIMULATION

作　　者：邵明[1] 孙守迁[2] 

机构地区：[1]浙江大学计算机科学与技术学院,杭州310027 [2]浙江大学玉泉校区现代工业设计研究所,杭州310027

出　　处：《计算机应用与软件》2011年第1期8-10,25,共4页Computer Applications and Software

基　　金：国家自然科学基金重点项目(60533080);浙江省科技计划项目(2007C2396)

摘　　要：针对大规模实时人群仿真往往需要在极短时间内完成大量的计算,而一般的串行算法无法完成该类计算任务的现状,提出了一种基于GPU(Graphic processing unit)的并行化人群实时仿真算法。该算法利用人群仿真中人群个体的独立性,将个体近邻查询,个体位置更新两个计算过程并行化,从而实现了大规模人群的实时仿真。For large-scale real-time crowd simulation,it requires a large amount of calculations to be done in a very short period of time,and this can＇t be completed by general serial algorithms.In this paper we present a GPU-based parallelised algorithm for real-time crowd simulation.The algorithm takes advantage of the independence of individuals in crowd simulation,makes two calculations of neighbour query and location update of individual in parallel,so as to have realized the large-scale real-time crowd simulation.

关 键 词：并行计算 并行近邻查询 人群仿真 图形处理单元 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]