出租车GPS数据轨迹化方法研究

为提高出租车行业理论研究与实际应用的GPS轨迹数据质量,针对原始的出租车GPS数据,提出一套考虑出租车行业特性的完整的GPS数据轨迹化方法。分析了哈尔滨市出租车GPS数据的采样率、误差及与路网的匹配情况;提出一种考虑车头朝向、临近距离、路段连通性及转弯限制四种权重的拓扑地图匹配算法;提出一种出租车出行轨迹识别方法,并在此条件下建立了出租车GPS轨迹的存储结构模型。结果表明,提出的地图匹配算法有较高的准确率,能够满足理论研究及实践应用的需要;而轨迹识别方法及存储模型在实际中成功应用,为哈尔滨市出租车智能管理调度平台提供数据支撑。

Research on Taxi GPS Trajectory Composing Method

In order to improve the GPS trajectory data quality of the research and application about taxi, aimed at raw taxi GPS data, this paper proposed a set of composing taxi trajectory method by considering the taxi area feature. The sampling frequency, error, and map matching situation of taxi GPS data in Harbin were analyzed. Based on the four weights consisting of heading, proximity, link connectivity and turn restriction, a topological map matching algorithm was proposed. A method of identifying taxi trajectory was proposed, and after this, the taxi GPS trajectory storage model was built. The results show the proposed map matching algorithm had high accuracy rate and could met the research and application need. And the trajectory identifying method and storage model were successfully applied in real world application. All these work provided the basic data for Taxi management system in Harbin.