首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于大规模浮动车数据的交叉口转向规则自动提取算法
引用本文:庄立坚,何兆成,杨文臣,叶伟佳,邓玲丽.基于大规模浮动车数据的交叉口转向规则自动提取算法[J].武汉水运工程学院学报,2013(5):1084-1088.
作者姓名:庄立坚  何兆成  杨文臣  叶伟佳  邓玲丽
作者单位:[1]中山大学智能交通研究中心,广州510275 [2]同济大学道路与交通工程教育部重点实验室,上海201804
基金项目:广东省粤港关键领域重点突破项目资助(批准号:2011A011305002)
摘    要:针对传统电子地图在生成时只包含路网拓扑关系,不具备路口转向规则的自动生成及更新能力,提出了利用大规模浮动车数据自动生成交叉口转向规则的算法.该方法建立转向规则数学表达和存储模型,并依托大规模浮动车数据分析和处理,引入置信点概念,提出基于首尾置信点控制的转向规则自动提取算法;以广州市为例,选取1d的浮动车GPS数据,对提出的算法进行效用评价,实验结果表明提出的算法准确率达90.4%,可准确实现多数交叉口转向规则的自动提取.

关 键 词:智能交通  转向规则  浮动车  置信点  判定准则  交叉口

A Large-scale Floating Car Data-based Algorithm of Turning Rule Extraction at Intersections
Authors:ZHUANG Lijian  HE Zhaocheng  YANG Wenchen  YE Weijia  DENG Lingli
Institution:1.Research Center of Intelligent Transportation System, Sun Yat-sen University, Guangzhou 510275, China; 2.The Key Laboratory of Road and Traffic Engineering, Ministry of Education, Tongji University, Shanghai 201804, China;)
Abstract:In the light that traditional electronic maps only contains road network topological data when generated and lacks the abilities in extracting and updating turning rules at intersections automatically,a large-scale floating car data-based algorithm of turning rule extraction (FCDTRE) at intersections is presented.Mathematical models of turning rules are first developed,then,on the basis of processing of large-scale floating car data,the concept of Confidence Point (CP) is employed and the beginning and end CPs-based automatic extraction algorithm of turning rules is developed.Taking Guangzhou as a study case,one-day floating car data (FCD) are used to evaluate the performance of the FCDTRE.Extensive experimental results have demonstrated the better potential of the FCDTRE in automatic extraction of turning rules at most intersections,with the accuracy rate of over 90.4%.
Keywords:intelligent transportation  turning rule  floating car  confidence point  judgment criterion  intersection
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号