首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于改进社会力的单向流电动自行车行为建模研究
引用本文:王维莉,卢晓磊,张旺,熊浩然,余冕.基于改进社会力的单向流电动自行车行为建模研究[J].交通运输系统工程与信息,2022,22(5):223-232.
作者姓名:王维莉  卢晓磊  张旺  熊浩然  余冕
作者单位:上海海事大学,物流研究中心,上海 201306
基金项目:国家自然科学基金;上海市科技创新行动计划。
摘    要:电动自行车作为一种便捷高效的出行工具,已在许多国家被广泛使用,但对电动自行车微观行为的研究仍十分有限。在上海海事大学对电动自行车驾驶行为进行观测实验,提取电动自行车轨迹数据,以纵向间距、速度差、侧向净距、水平间距作为特征变量,利用CART(Classification and Regression Tree)决策树分类结果建立电动自行车骑行决策行为规则。然后,基于社会力模型,分别引入超越力、跟随力等改进行为力模拟电动自行车的超越、跟随等骑行决策行为。考虑公式超越力、固定值超越力和临时目标点超越力这3种超越力形式,通过实际观测数据和仿真结果比较进行参数标定和模型验证,选取单向流电动自行车行驶场景开展数值仿真分析。结果表明:临时目标点超越力的改进社会力模型在超越过程中的轨迹误差最小,电动自行车超越时所需要的横向间距与超车时横向速度、超车完成时间成正比,最佳超车横向间距为2 m。

关 键 词:城市交通  微观行为建模  改进社会力模型  电动自行车行为  CART决策树  
收稿时间:2022-05-17

Modeling of Electric Bicycle Behavior in Unidirectional Flow Based on Improved Social Forces
WANG Wei-li,LU Xiao-lei,ZHANG Wang,XIONG Hao-ran,YU Mian.Modeling of Electric Bicycle Behavior in Unidirectional Flow Based on Improved Social Forces[J].Transportation Systems Engineering and Information,2022,22(5):223-232.
Authors:WANG Wei-li  LU Xiao-lei  ZHANG Wang  XIONG Hao-ran  YU Mian
Institution:Logistics Research Center, Shanghai Maritime University, Shanghai 201306, China
Abstract:As a convenient and efficient transportation tool, the electric bicycle is widely used in many countries, but the research on its microscopic behavior is still limited. In this study, the observational experiment on the driving behavior of electric bicycles was conducted at Shanghai Maritime University, and electric bicycle trajectories were extracted. Longitudinal spacing, speed difference, lateral spacing, and horizontal spacing are regarded as characteristic variables, and the decision-making behavior rules of electric bicycles are established by using the classification results of the CART decision tree. Then, based on the social force model, overtaking force and following force are introduced to simulate the decision-making behavior of electric bicycles. Three forms of overtaking force are considered. Specifically, the overtaking force is respectively represented by equations, fixed values, and calculated by adding a temporary target point. The model is calibrated and validated through the comparison of the observation data and a series of simulation tests, and the scenario that electric bicycles drive in unidirectional flow is selected for numerical simulation analysis. The results show that the trajectory error of the improved social force model is the smallest, in which the overtaking force is calculated by adding a temporary target point. Moreover, the horizontal spacing required for the electric bicycle overtaking is directly proportional to the horizontal speed and overtaking completion time. Lastly, the best horizontal spacing in overtaking is 2 m.
Keywords:urban traffic  microscopic behavior modeling  improved social force model  electric bicycle behavior    CART decision tree  
点击此处可从《交通运输系统工程与信息》浏览原始摘要信息
点击此处可从《交通运输系统工程与信息》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号