利用低频浮动车交叉口运行状态的单车路段行程时间估计

Single Car Travel Time Estimation Based on Running State of Low Frequency Floating Car at Intersections

针对现有的基于浮动车数据的单车路段行程时间估计方法适用高频GPS数据,以及对浮动车在路段交叉口运行状态考虑不足的问题,提出一种基于低频浮动车交叉口运行状态的单车路段行程时间估计方法。该方法主要进行了以下改进：（1）基于概率转换的地图匹配方法,用A＊算法拟合缺失路径,提高了地图匹配的效率和完整性;（2）分析车辆在交叉口区域内外的运行状态,改进了单车路段行程时间的估计方法,并对干扰停车予以剔除,提高了对路段行程时间估计的准确性。并用浮动车GPS实验数据验证了本文所提方法对低频GPS数据计算单车路段行程时间的有效性。

The existing single car travel-time calculation methods based on floating car data are suitable for high frequency GPS data but do not give sufficient consideration to the running state of floating car at intersection.In view of the problems,we put forward a method of travel-time estimation based on the running state of low frequency floating cars at intersection.This method has the main following improvements：（1） We improved the efficiency and integrity of matched path based on probability transformation map matching algorithm and A＊algorithm to fit missing path.（2） We improved the precision of travel-time estimation by modifying travel-time estimation method according to the running state of vehicle at the intersection area and eliminating interference stop.Low frequency GPS data verified the validation of our proposed method.