考虑异质车流的信号交叉口实时排队估计  

Real Time Queue Estimation for Signalized Intersection Considering Heterogeneous Traffic Flow

作　　者：陈之锴 成卫[1] 李黎山[1] 李冰[1,2] CHEN Zhikai;CHENG Wei;LI Lishan;LI Bing(School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China;Key Laboratory of Urban ITS Technology Optimization and Integration Ministry of Public Security, Hefei 230088, China)

机构地区：[1]昆明理工大学交通工程学院,云南昆明650500 [2]城市交通管理集成与优化技术公安部重点实验室,安徽合肥230088

出　　处：《物流科技》2019年第5期73-78,共6页Logistics Sci-Tech

基　　金：公安部重点实验室课题项目(2017KFKT04)

摘　　要：信号交叉口排队长度是交通信号控制的一项重要指标,为了提高排队估计精度以及实用性,文章以相邻信号交叉口的下游交叉口排队长度作为研究对象,在关联信号交叉口排队长度计算模型基础上,提出了一种考虑异质车流的影响下的实时排队长度的估计方法。通过异质车流的行程时间,拟合双峰模型得到车辆离散模型,研究表明:利用该模型分析上游车辆离去流量,得到下游到达流率;通过信号控制方案以及停车波波速,估计交叉口车辆排队长度。大量实证研究检验平均绝对误差在0.3veh/lane/s内,提高了信号交叉口排队长度估计的有效性与实用性,从而为后续异质交通环境下信号配时优化提供理论依据。Queuing length at signalized intersections is an important index of traffic signal control. In order to improve the accuracy and practicability of queuing estimation, the queuing length at downstream intersections of adjacent signalized intersections was taken as the research object, and a real-time queue length estimation method considering heterogeneous vehicle flow was proposed based on the queuing length calculation model of signalized intersection. With the travel time of heterogeneous traffic flow, the vehicle discrete model was built by fitting the bimodal model;the upstream vehicle departure flow and the downstream arrival flow rate were analyzed by using this model;the length of vehicle queue at intersections was estimated through the signal control scheme and the parking wave speed;a large number of empirical studies verify that the average absolute error is within 0.3 veh/lane/s. The validity and practicability of queuing length estimation at signalized intersections are improved, which provides basis for subsequent signal control and traffic carding.

关 键 词：智能交通 排队实时估计 离散 异质车流 

分 类 号：U491.2[交通运输工程—交通运输规划与管理]