火灾中基于个体行为的人群疏散仿真  

Simulation of crowd evacuation in fire based on agent behavior

作　　者：陆卓谟[1] 秦文虎[2] 

机构地区：[1]东南大学自动化学院,南京210096 [2]东南大学仪器科学与工程学院,南京210096

出　　处：《东南大学学报（自然科学版）》2011年第6期1295-1299,共5页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金面上资助项目(60972001);江苏省社会发展基金资助项目(BE2009662)

摘　　要：为了指导火灾中的人群疏散,研究了一种新的人群疏散模型用于提高人群疏散仿真的真实性.通过分析个体对火灾警报的不同响应时间,个体在疏散过程中采取的不同运动速度,个体间的组运动方式,个体的出口选择等因素对人群疏散的影响,建立了个体行为特征与元胞自动机相结合的人群疏散模型.该模型包括个体对火灾警报响应模型、个体逃生速度模型、组运动模型和出口选择模型.仿真分析表明:火灾中不同的个体行为对建筑物内人群疏散的总体时间有较大影响;人群疏散仿真中引入个体的反应时间和组运动方式能使人群疏散仿真更加真实.For guiding the crowd evacuation in fire,a new crowd evacuation model based on agent＇s behavior which can improve the realism of crowd evacuation is proposed.Through analyzing the influences of many factors on the evacuation process,including agent＇s different reaction time when encountering the fire alarm,different moving velocities in the process of evacuation,application of group movement and the selection of exits,a model combining the characteristics of agent behaviors and cellular automata is established.The model includes the agent＇s response of fire alarm model,the agent＇s evacuation speed model,the group movement model and exit selection model.The simulation results show that different agent behavior in fire remarkably influence total time of crowd evacuation in building.Crowd evacuation simulation integrating agent＇s reaction time and group movement can improve the realism greatly.

关 键 词：火灾 人员疏散 元胞自动机 个体行为 人群仿真 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术] X928.03[自动化与计算机技术—计算机科学与技术]