以切换定位技术作为手机探测车的定位技术,针对切换定位技术低定位精度的特征,对手机探测车异常切换位置修正问题进行了研究.根据问题描述,提出一个两阶段手机探测车异常切换位置修正方法,分别为异常切换路段速度修正和异常切换位置重置.在第一阶段中,结合桥接切换路段构建一个双层数据融合模型,对异常切换路段速度修正值进行求解.在第二阶段中,采用遗传算法求解异常切换位置重置问题得到异常切换位置修正结果.通过实证数据,评估分析了基于本文修正方法计算得到的手机探测车切换路段速度估计绝对误差,与原始数据比较表明,本文提出的方法是可行有效的,可以应用于道路交通信息提取的研究中.
In this paper,handover location technique is employed for positioning mobile probes.In view of low location accuracy of handover location technique,an outlier handover location correction approach was presented that solved the problem in two phases,travel speed correction of outlier handover link and outlier handover location reposition.In the first phase,a two-layer data fusion model based on the definition of bridge handover link was proposed for recalculating the travel speeds of outlier handover links.In the second phase,the problem of outlier handover location reposition was solved by genetic algorithm.Based on the empirical data,absolute errors of travel speeds of handover links determined by the corrected results are analyzed and compared with results without outlier handover location correction.Evaluation results demonstrate that the proposed approach is feasible and effective,which can be used in the application of traffic information extraction.