基于多源数据的城市道路网络行程时间预测模型

针对基于单一数据源、利用卡尔曼滤波理论建立行程时间预测模型存在的不足,采用多源数据进行行程时间预测以提高精度。浮动车、固定检测器是常用的交通信息采集方法,在信息种类、数据精度等方面存在一定的互补性。因此,选择2种检测器的实时交通数据作为模型输入参数。利用卡尔曼滤波理论,以流量、占有率、行程时间作为输入量构成参数矩阵,建立城市道路网络行程时间预测模型。并通过Vissim仿真实验验证了模型的有效性。结果表明:基于多源数据的行程时间预测模型平均绝对相对误差为5.45%,其精度比单独采用固定检测器检测数据预测提高了14.4%,比单独采用浮动车数据预测提高了7.5%。 

Development of A Travel Time Prediction Model for Urban Road Network using Multi-source Data

In view of the deficiencies of traditional travel time prediction models developed based on Kalman filtering technique and single data source ,the multi-source data are used to improve such models and the prediction accuracy of travel time .Floating cars and loop detectors are common ways for collecting travel time ,and the two are complementary to each other in many ways .Therefore ,the real-time traffic data from the two sources are used as the inputs of the pre-diction model .Through Kalman filtering ,flow ,occupancy and travel time are used as inputs of the proposed travel time prediction model .Finally ,the model is verified through a simulation from Vissim .The simulation results show that the average absolute relative error of the estimated travel time based on the model developed based on the multi-source data is 5 .45% ,which is 14 .4% lower than those estimated based on the loop detector data only and 7 .5% lower than those esti-mated based on the floating car data alone .