【摘要】 针对公共场所发生火灾等事故时人群密集的特点,研究了紧急状态下人群行为的仿真。分析了紧急状态下人群的行为特点,定义了逃跑、跟随和拥挤行为;采用虚拟视觉和九宫避碰策略,实现了虚拟人行为的自治控制方式。针对紧急状态下人群逃生时个体行为之间的差异性,依据个体与个体之间的作用力、摩擦力动态改变个体的行进方向和依据人群密度采用人群疏散速度模型实时确定每个个体的运动速度,解决了个体之间的动态避碰、绕行和超越等问题。经实验验证,文中的人群行为仿真技术能达到较好的效果。更多还原

【Abstract】 Focus on the crowd denseness in public place when fires happen,study on the simulation of virtual crowd behaviors in emergency.The characteristics of real-life crowd behaviors in emergent statue is analyzed.According to this,escape,follow and crowd behaviors of virtual crowd are defined.Realized self-control model of virtual characters’ behaviors by introducing virtual vision and nine-unit collision avoidance.Consider the otherness of crowd behaviors in emergency,to settle the problem of dynamic collision avoidance,bypass and beyond,interaction and friction are introduced to dynamic change the forward direction,and real-time to determine individual’s movement speed by using crowd evacuate speed model that according to the crowd density.The results show that the simulation method of this article can get a satisfying effect.更多还原