基于融合技术的道路交通状态判别模型

为了提高道路交通状态判别精度,提出基于浮动车检测与感应线圈融合技术的道路交通状态判别模型。该模型包括3部分:1)浮动车模块:利用安装在出租车上的GPS定位设备得到道路3部分的行程时间;2)感应线圈模块:利用安装在道路上的感应线圈以及交通信号参数得到道路的行程时间;3)数据融合模块。利用神经网络将以上两模块的结果作为输入从而提高道路交通状态判别的精度。该文利用7 000多辆装有GPS模块的出租车、100个安装在广州市主要道路口上的固定检测器以及广州市电子地图,对提出的模型进行了试验,试验结果表明该模型是有效的,50个数据的均方误差为0.1 s。

Traffic state classification model of travel times based on the fusion technique

Traffic state classification accuracy was improved by a traffic state classification model based on the fusion of travel time studies using a probe vehicle and crossing data from fixed detectors.The model includes a mobile detection module which measures travel times over a link using taxies equipped with differential global positioning system receivers,a loop detection module which measures travel times using fixed detectors in roads and traffic signal timing parameters,and a data fusion module which uses a neural network to combine outputs from the first two modules to improve the traffic state classification accuracy. Tests using probe vehicle data from 7 000 taxies equipped with GPS receivers,100 fixed detectors in main roads in Guangzhou and an electronic map of Guangzhou containing 10 000 roads indicate that the model is efficient and accurate.