信号交叉口集聚车辆的跟驰行为分析与建模

Analysis and Modeling for Car-following Behaviors on Arrival Traffic Flow of Signalized Intersection

为了更好地模拟城市信号交叉口集聚车辆的跟驰行为,进而应用于城市信号交叉口信号配时和交通流理论研究,采用一种基于视频的交通流数据采集方法来采集信号交叉口的微观交通行为数据.运用灰色关联分析方法对采集到的微观交通数据进行分析,挖掘出其中的有用信息,从而寻求能够最大程度反映信号交叉口集聚车辆跟驰行为的影响变量.构建城市信号交叉口车辆集聚过程中的跟驰模型并进行参数标定、效果验证和比较分析.研究表明,新提出的跟驰模型能够很好地拟合信号交叉口集聚车辆的跟驰行为实测数据,其拟合性和稳定性优于重新标定后的扩展GM模型.

In order to better simulate car-following behaviors on arrival traffic flow of signalized intersections and further to better study signal timing and traffic flow theory, the field car-following data at a signalized intersection are collected by using a new proposed data acquisition method, and then the measured data are mined to seek the endogenous variables as the input variables of car-following model. On the basis of foregoing analysis, a new car-following model is proposed, calibrated and verified. This comparative research shows that the new proposed car-following model can fit the measured data well and that the stability of traffic flow simulated by it is prior to that by the recalibrated expansion of GM model.