城市快速路路段行程时间估计方法  

Travel Time Estimation Method of Urban Expressway Section

作　　者：张伟健 邴其春 沈富鑫 胡嫣然 高鹏 ZHANG Weijian;BING Qichun;SHEN Fuxin;HU Yanran;GAO Peng(School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China;Qingdao Transportation Public Service Center,Qingdao Shandong 266100,China)

机构地区：[1]青岛理工大学机械与汽车工程学院,山东青岛266520 [2]青岛市交通运输公共服务中心,山东青岛266100

出　　处：《广西师范大学学报（自然科学版）》2023年第2期49-57,共9页Journal of Guangxi Normal University:Natural Science Edition

基　　金：山东省重点研发计划项目(2019GGX101038);山东省自然科学基金(ZR2019MG012)。

摘　　要：为有效提高城市快速路路段行程时间估计精度,本文提出一种基于自适应滤波和网格虚拟车的行程时间估计方法。首先,利用自适应平滑滤波插值获得固定检测器之间的速度数据,实现固定检测器之间的时空速度场重构。然后将重构的速度场细分为若干网格单元,通过计算虚拟车在重构时空速度场的虚拟轨迹估计路段行程时间。最后,选取城市快速路感应线圈实测数据进行实例验证。实验结果表明,本文方法在自由流、平稳流和拥堵流3个运行时段下,均方根误差分别为0.15、0.35和1.06,平均相对误差均低于10%,具有较高的估计精度。In order to effectively improve the travel time estimation accuracy of urban expressway sections,this paper presents a travel time estimation method based on adaptive filtering and grid virtual vehicle.The velocity data between fixed detectors are obtained by adaptive smoothing filter interpolation,and the spatiotemporal velocity field between fixed detectors is reconstructed.Then the reconstructed velocity field is subdivided into several grid elements.The road travel time is estimated by calculating the virtual trajectory of the virtual vehicle in the reconstructed space-time velocity field.Finally,the measured data of induction coil on urban expressway are selected for example verification.The experimental results show that the root mean square error of this method is 0.15,0.35 and 1.06,respectively,and the average relative error is less than 10%,so it has high estimation accuracy.

关 键 词：交通工程 行程时间估计 自适应平滑滤波 速度场重构 网格虚拟车 

分 类 号：U491[交通运输工程—交通运输规划与管理]