综合考虑到浮动车检测技术与感应线圈检测技术的优缺点,为了提高道路行程时间估计的精度及完备性,提出基于浮动车与感应线圈的融合检测技术的行程时间估计模型。该模型利用神经网络技术对两种检测技术同一路段的检测数据进行融合,从而达到提高道路行程时间估计精度和完备性的目的。最后,以广州市7000多辆装有GPS装置的出租车所提供的浮动车数据、100多个安装在广州市各主要道路口上的感应线圈检测器提供的感应线圈数据以及广州市交通电子地图为基础,在10条道路上分别随机选取的500个两种检测数据对提出的模型进行了验证,试验结果表明,此模型在道路行程时间估计的精度方面较浮动车移动检测技术与感应线圈技术有较大提高。
Considering the advantages and disadvantages of both probe vehicle and loop techniques, to improve the accuracy and completeness of estimating travel time, a new estimation model of travel time is described based on fusion technique performing travel time studies using probe vehicle and loop detectors. This model uses neural network to fuse the same road detecting data of two detecting techniques to improve the accuracy and completeness of estimating travel time. Finally, the test of the model is verified using random 500 data in 10 roads based on probe vehicle data from 7 000 taxies equipped with GPS receivers, 100 fixed detectors fixed in main roads in Guangzhou and electronic map of Guangzhou City.The results indicate that the model is more efficient than probe vehicle and loop technique on estimating travel time.