论文部分内容阅读
为更好地模拟城市信号交叉口交通流的运行特性,在深入分析信号交叉口减速停车过程中车辆跟驰运动特性的基础上,提出一种考虑信号灯作用的车辆跟驰模型。首先,设计一种基于视频的数据采集方法来采集信号交叉口减速停车过程中的跟驰数据;然后,对实测数据进行挖掘分析,选择相对速度、车间距、前导车加速度和跟驰车到信号灯的距离作为拟构建模型的输入变量;最后,在GM模型的基础上确定新模型的结构并进行参数标定、效果验证和比较分析。结果表明:新提出的模型能够很好地拟合实测数据,其平均绝对误差值比重新标定的GM模型减小了6.08%,拟合性能明显优于重新标定的GM模型。
In order to better simulate the operation characteristics of traffic flow at urban signalized intersections, a car-following model considering the role of traffic lights is proposed based on the analysis of the car-following motion characteristics during signalized intersection decelerations and decelerations. Firstly, a video-based data acquisition method is designed to collect the following data of decelerated parking at the signalized intersection. Then, the measured data are analyzed and the relative speed, vehicle distance, leading vehicle acceleration and car-to- As the input variable to construct the model. Finally, the structure of the new model is determined and the parameters are calibrated, validated and compared according to GM model. The results show that the proposed model can well fit the measured data and the average absolute error is reduced by 6.08% compared with the recalculated GM model, and the fitting performance is obviously better than the re-calibrated GM model.