融合出租车驾驶经验的层次路径规划方法  

Hierarchical Path Planning Method Based on Taxi Driver Experiences

作　　者：胡继华[1] 黄泽[1] 邓俊[1] 谢海莹[1] 

机构地区：[1]中山大学工学院智能交通中心,广州510006

出　　处：《交通运输系统工程与信息》2013年第1期185-192,共8页Journal of Transportation Systems Engineering and Information Technology

基　　金：国家863计划项目(2011AA110306-4);广东省2011年度安全生产专项资金项目(2011-118)

摘　　要：出租车驾驶员对城市道路交通状况较为熟悉,他们选择的路径具有代表性,因此将出租车驾驶员路径选择经验融合到路径规划算法中,对提高出行效率具有重要的意义.本文提出一种融合出租车驾驶经验的层次路径规划方法,主要包括三部分:首先,从出租车GPS数据中提取出出租车载客行驶轨迹;然后,根据各路段出租车行驶频率高低对路网进行分层,构建基于出租车经验路径的分层路网;在此基础上,使用Dijkstra算法实现层次路径规划.最后,本文以广州市为研究区域,将该方法得到的规划路径与经典路径规划算法的结果进行比较.结果表明,融合出租车驾驶经验的路径规划方法所得路径在行程时间上占有一定的优势.The route choice behaviors of taxi drivers are usually representative because they are more familiar with urban road status. This makes it possible to use the taxi drivers＇ experience to support the path planning. To make the guidance result meet the drivers＇ expectations well, this study presents a hierarchical path planning method using the taxi driver experiences. The method consists of three steps ： first, routes are recovered from the taxi trajectories; second, all roads are redefined and categorized according to the track data and the road network is classified into different experience grades using travel frequency for road segments; third, with the Dijkstra algorithm, a hierarchical path planning method is proposed. Finally, taking Guangzhou city as an example, this paper compares the paths generated by the proposed approach with the conventional algorithm＇s results. The experimental result shows that travel time of the paths planned by the proposed method has been effectively reduced.

关 键 词：智能交通 路径规划 DIJKSTRA算法 出租车驾驶经验 分层路网 

分 类 号：U491.1[交通运输工程—交通运输规划与管理]