摘要
对适用于连续空间的基于力的行人仿真模型进行综述,并以社会力模型为例,指出了其本质局限性.在此基础上提出了全新的适用于连续空间的微观行人仿真模型:邻域决策模型.该模型使用"方向选择与速率调整"相结合的方法对行人的微观动作进行描述,并采用邻域作用原理驱动行人的决策过程.最后,通过一系列复杂场景内的单人行走仿真试验,分析了在不同参数配置下模型所模拟出的丰富的行人特性.说明该模型能够合理地模拟行人的非线性决策特征,与基于力的模型相比,更加符合行人的自然行走过程,证实了这种新的仿真模型的可行性.
The Force-based Pedestrians Simulation Models which are applicable to continuous space are reviewed in this paper. Taking Social Force Model （SFM）for example, the limitations of these models are pointed out. To improve the limitations, a new simulation model .Neighborhood-based Decision-Making Model in continuous space was proposed. This model adopts a new descriptive method of pedestrians＇ movement combining direction choice and speed adjusting, and drives pedestrians＇ decisionmaking process based on Neighborhood-based Impact Principle. Finally, a series of single pedestrian walking simulation test in the complex scenarios was presented and the various walking characters from different parameters were in-deep analyzed. The simulation results proved the feasibility of the Neighborhood-based Decision-Making Model.
出处
《武汉理工大学学报（交通科学与工程版）》
2009年第5期815-818,共4页
Journal of Wuhan University of Technology（Transportation Science & Engineering）
基金
国家科技支撑计划项目资助(批准号:2007BAK35B06)
关键词
方向选择
独立作用原理
邻域作用原理
社会力模型
邻域决策模型
direction choice
neighborhood-based impact principle
neighborhood-based impact principle
social force model
neighborhood based decision making model
