社会行为驱动的疏散仿真方法研究  

Research on Evacuation Simulation Method Considering Social Behavior

作　　者：邓媛媛 郑利平[1] 蔡瑞文 Deng Yuanyuan;Zheng Liping;Cai Ruiwen(VCC Laboratory,School of Computer and Information,Hefei University of Technology,Hefei 230009,China)

机构地区：[1]合肥工业大学计算机与信息学院VCC实验室

出　　处：《系统仿真学报》2020年第1期130-141,共12页Journal of System Simulation

基　　金：国家重点研发计划(2016YFC0800100);安徽省科技强警项目(1704d0802177)

摘　　要：在紧急疏散场景中,个人具有的典型社会属性会对其疏散行为产生影响。将个体对环境的熟悉程度和个体所属团队这2种社会因素引入并应用到人群疏散仿真中,提出一种疏散仿真方法。利用RVO库的实时避碰技术模拟人群的动态运动过程,通过添加局部目标点以及对局部目标点的选择机制来动态模拟人群不同的社会行为。通过对不同熟悉度的人群以及不同团队数量的人群进行仿真实验,实验结果表明环境熟悉程度和团队因素对疏散时间存在一定影响,仿真结果具有较高真实性。In an emergency evacuation scenario, the typical social attributes of an individual impact their evacuation behavior. Two kinds of social factors, such as individual familiarity to the environment and the individual group, are introduced and applied in crowd evacuation simulation. An evacuation simulation method is proposed. The real-time collision avoidance technique of RVO library is used to simulate the dynamic motion of the population. The local target points and its selection mechanism are used to simulate the different social behaviors of the population. Experiments show that the familiarity to the environment and group factors have influence on the evacuation time. Experiments show that the method can achieve better simulation results.

关 键 词：疏散仿真 环境熟悉度 团队 RVO2(Reciprocal Velocity Obstacle) 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]