行人路径选择行为微观仿真建模研究

发布时间：2018-06-25 22:40

本文选题：行人微观仿真 + 社会力模型　；参考：《昆明理工大学》2017年硕士论文

【摘要】：随着全球大型活动的频繁组织和交通枢纽、体育馆和商城等各类大型步行设施的建设规模不断增大,设施内人群拥挤现象也日益加剧,行人交通流的安全性、高效性和舒适性等问题受到越来越多的关注。行人微观仿真是行人流再现和预测等研究的重要手段,为步行设施规划和行人流组织管理提供指导,本文在社会力模型的基础上研究行人路径选择的行为特性,探究行人自组织的规律。本文从行人微观仿真模型的模拟效果分项综述着手,提出了集自组织现象、异质人群(个体和同伴群)的模拟和全局路径选择功能于一体的微观仿真模型框架。分别从行人视觉信息的作用和同伴群的角度改进社会力模型,构建行人局部路径选择模型,并引入基于最短路径算法的全局路径选择模型。改进模型仿真显示,有视觉信息作用的模型弱化了仿真过程中震荡效应,考虑同伴群的交流分组的模型较好重现了同伴群随密度和人数增加的动态分组现象。行人流特性分析表明,同伴群的存在会降低了行人流速度,但其影响强度随人流密度的增加减弱;而同伴群动态交流分组模型的引入会提升仿真行人流的通行效率,使之更加接近真实行人流的运动特性。最后建立路网对行人全局路径选择行为进行仿真,结果发现行人交通量的水平是影响行人动态选择的主要影响因素,在低水平下行人首先考虑的是体力问题,但是当行人交通量增大,行人在权衡体力、行走的舒适性和行走时间时,更倾向于选择行走舒适性高和时间短的路线。本研究完善了社会力模型的描述能力,丰富了行人流的特性,对未来的步行设施设计提供理论依据。
[Abstract]:With the frequent organization and transportation hub of global large-scale activities, the construction scale of various large-scale walking facilities, such as gymnasiums and shopping malls, has been increasing, the crowd congestion in the facilities is also increasing day by day, and the safety of pedestrian traffic flow is also increasing. More and more attention has been paid to the problems of high efficiency and comfort. Pedestrian microscopic simulation is an important means to study pedestrian flow reappearance and prediction. It provides guidance for pedestrian facilities planning and pedestrian flow organization and management. Based on the social force model, this paper studies the behavior characteristics of pedestrian path selection. Explore the rules of pedestrian self-organization. Based on the summary of the simulation effect of pedestrian micro simulation model, this paper presents a micro simulation model framework which integrates the self-organizing phenomenon, the simulation of heterogeneous population (individual and peer group) and the function of global path selection. From the perspective of visual information and peer group, the social force model is improved, and the local pedestrian path selection model is constructed, and the global path selection model based on the shortest path algorithm is introduced. The improved model simulation shows that the visual information model weakens the oscillation effect in the simulation process, and the dynamic grouping phenomenon of peer group with increasing density and number of peers is well reproduced by the model considering the group of peers. The analysis of pedestrian flow characteristics shows that the presence of peer group will reduce the speed of pedestrian flow, but its influence intensity will weaken with the increase of passenger flow density, and the introduction of dynamic AC grouping model of peer group will improve the efficiency of simulating pedestrian flow. So that it is closer to the movement characteristics of the real pedestrian flow. Finally, the road network is established to simulate the overall path selection behavior of pedestrians. The results show that the level of pedestrian traffic volume is the main influencing factor of pedestrian dynamic selection, and the physical strength is the first consideration for pedestrians at low level. However, when the pedestrian traffic volume increases, pedestrians are more inclined to choose the route with high comfort and short time when weighing their physical strength, walking comfort and walking time. The research improves the description ability of social force model, enriches the characteristics of pedestrian flow, and provides theoretical basis for the design of walking facilities in the future.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491

【参考文献】