论文部分内容阅读
总结了社会力模型的研究现状、存在问题和基本原理.将分子动力学的Gear预测校正法应用到社会力仿真模型中,并根据柯朗条件确定了行人仿真显式方法的最小时间步长,引入了链接列表元胞(linked-listcell method)加速算法.最后设计了编程实现框架,采用VC++实现了上述方法,并进行了仿真实验,再现了自动渠化等行人群自组织现象,仿真结果显示减速避让机制的加入可以得到更符合实际的效果,而链接列表元胞法可大幅缩减仿真时间.
This paper summarizes the research status, existing problems and basic principles of the social force model.Applied Gear predictive-correction method to the social force simulation model, and determines the minimum time step of the pedestrian simulation explicit method according to the condition of Cochrane, The paper introduced the accelerated algorithm of linked-list cell method.Finally, the programming framework was designed and the above method was implemented by using VC ++, and the simulation experiments were carried out to reproduce the self-organizing phenomenon of crowd-crowd such as automatic channelization. The simulation results show Deceleration avoidance mechanism can be added to achieve more realistic results, and linked list cellular method can significantly reduce the simulation time.