车辆诱导系统及关键技术研究
|
摘要
随着经济的发展,城市化水平的提高,交通问题也日趋严重。尤其在一些大城市,交通拥挤、环境污染等已经成为制约经济发展、亟待解决的重要问题。智能交通系统是解决城市交通问题公认的最好的方法。车辆诱导系统是智能交通系统中的一个重要方面,旨在通过向驾驶员提供基于实时交通信息的最佳行驶路线来达到诱导出行行为,减少车辆在道路上的逗留时间,避免交通拥挤、阻塞现象,从而改善城市交通状况的目的。文章主要对车辆诱导系统的结构框架、实施框架以及若干关键技术进行了研究。
本文的主要工作如下:
1.介绍了国内外几种具有代表性的车辆诱导系统,总结了当前的研究现状,并对车辆诱导系统的研究内容、构成和分类进行了讨论。在对各个功能模块进行分析的基础上给出了符合中国国情的车辆诱导系统结构框架及实施框架。
2.行程时间预测是车辆诱导系统中最重要的问题之一。对现有的几种行程时间预测方法及近年来所做出的一些研究进行了分析,指出这些方法普遍过于追求预测的准确性,而忽视了算法的实用性。为此,给出一种基于车牌自动识别系统的行程时间预测模型。该模型使用四分位异常数据剔除的方法对观测到的行程时间进行预处理;在对指数平滑方法进行分析的基础上,给出了等参数模型和等参数比例模型两种行程时间预测方法。最后,用某城市的交通实测数据对该模型进行验证,实验结果表明该模型有较好的预测效果,算法简单实用、易于推广。
3.路径优化是车辆诱导系统中的一个关键功能,路径选择的好坏直接影响着诱导的效果。城市交通网络中路径行程时间是随着时间的变化而变化的,因此最小时间路径求解比较困难,传统的一些最短路径求解方法都无法直接应用。为此,提出把交通网络抽象为时间依赖的网络模型的解决方法。在对时间依赖网络模型和理论基础进行分析的基础上给出最小时间路径算法,并通过实验验证了算法的正确性和有效性。
Along with the economical development, the urbanized level enhancement, the transportation question day by day is also serious. Especially in some big cities, traffic congestion, the environmental pollution and so on, the questions which caused already becomes urgently awaited to be solved which restricts economy development. The study on the vehicles guidance system is an important research in the field of Intelligent Transportation System, which guide the behaviors of travelers by providing them with optimal route based on real-time traffic information. As the result the travel time can be saved and the traffic congestion can be avoid. The article studied on the framework and the key technology of the vehicles guidance system.
The main working content as follows:
1. Several kind representative vehicles guidance system in the domestic and foreign was introduced and the current study present situation was summarized. The working content, constitution and classification of the vehicles guidance system were discussed. Based on the analysis to each function module, conformed to the Chinese national condition vehicles guidance system framework was presented.
2. The prediction of link travel time is one of the most important questions of the vehicles guidance system. The accuracy vehicles guide need the real-time prediction to link travel time, and the accuracy vehicles guide depend on the prediction accurately. Several kinds of prediction method was introduction and some study based on which was analyzed. Based on the data collected by the license auto-recognition system, an algorithm for predicting travel time is proposed. The algorithm makes the moving exponential average model as the foundation, depend on the history travel time data to predict the link travel time in future times. The method is proved to be sample and efficient by experiments and practical application.
3. The route optimization is an essential function of the vehicles guidance system. As the time is changing, the travel time is also changing in traffic network. So shortest path search becomes considerably more difficult. Some traditional method for searching the shortest path was disabled in time-dependent networks. For solve the problems, many transportation systems can be represented by networks with travel times that are time-dependent. Based on the analyses to time-dependent networks and its theoretical foundations, a new shortest path algorithm is presented. The algorithm is proved to be correct and efficient by experiments and practical application. The shortest path algorithm in time-dependent networks has a broad application fields.
本文的主要工作如下:
1.介绍了国内外几种具有代表性的车辆诱导系统,总结了当前的研究现状,并对车辆诱导系统的研究内容、构成和分类进行了讨论。在对各个功能模块进行分析的基础上给出了符合中国国情的车辆诱导系统结构框架及实施框架。
2.行程时间预测是车辆诱导系统中最重要的问题之一。对现有的几种行程时间预测方法及近年来所做出的一些研究进行了分析,指出这些方法普遍过于追求预测的准确性,而忽视了算法的实用性。为此,给出一种基于车牌自动识别系统的行程时间预测模型。该模型使用四分位异常数据剔除的方法对观测到的行程时间进行预处理;在对指数平滑方法进行分析的基础上,给出了等参数模型和等参数比例模型两种行程时间预测方法。最后,用某城市的交通实测数据对该模型进行验证,实验结果表明该模型有较好的预测效果,算法简单实用、易于推广。
3.路径优化是车辆诱导系统中的一个关键功能,路径选择的好坏直接影响着诱导的效果。城市交通网络中路径行程时间是随着时间的变化而变化的,因此最小时间路径求解比较困难,传统的一些最短路径求解方法都无法直接应用。为此,提出把交通网络抽象为时间依赖的网络模型的解决方法。在对时间依赖网络模型和理论基础进行分析的基础上给出最小时间路径算法,并通过实验验证了算法的正确性和有效性。
Along with the economical development, the urbanized level enhancement, the transportation question day by day is also serious. Especially in some big cities, traffic congestion, the environmental pollution and so on, the questions which caused already becomes urgently awaited to be solved which restricts economy development. The study on the vehicles guidance system is an important research in the field of Intelligent Transportation System, which guide the behaviors of travelers by providing them with optimal route based on real-time traffic information. As the result the travel time can be saved and the traffic congestion can be avoid. The article studied on the framework and the key technology of the vehicles guidance system.
The main working content as follows:
1. Several kind representative vehicles guidance system in the domestic and foreign was introduced and the current study present situation was summarized. The working content, constitution and classification of the vehicles guidance system were discussed. Based on the analysis to each function module, conformed to the Chinese national condition vehicles guidance system framework was presented.
2. The prediction of link travel time is one of the most important questions of the vehicles guidance system. The accuracy vehicles guide need the real-time prediction to link travel time, and the accuracy vehicles guide depend on the prediction accurately. Several kinds of prediction method was introduction and some study based on which was analyzed. Based on the data collected by the license auto-recognition system, an algorithm for predicting travel time is proposed. The algorithm makes the moving exponential average model as the foundation, depend on the history travel time data to predict the link travel time in future times. The method is proved to be sample and efficient by experiments and practical application.
3. The route optimization is an essential function of the vehicles guidance system. As the time is changing, the travel time is also changing in traffic network. So shortest path search becomes considerably more difficult. Some traditional method for searching the shortest path was disabled in time-dependent networks. For solve the problems, many transportation systems can be represented by networks with travel times that are time-dependent. Based on the analyses to time-dependent networks and its theoretical foundations, a new shortest path algorithm is presented. The algorithm is proved to be correct and efficient by experiments and practical application. The shortest path algorithm in time-dependent networks has a broad application fields.
引文
[1]贺国光.ITS系统工程导论[M].北京:中国铁道出版社,2004.
[2]杨兆升.城市交通流诱导系统理论与模型[M].人民交通出版社,2000,1.
[3]杨兆升.城市交通流诱导系统[M].中国铁道出版社,2004,11.
[4]交通部公路科学研究所.中国智能交通系统体系框架研究.
[5]The U.S Department of Transportation,ITS AMERICA,ITS Architecture Development Program,phase 1,1994.
[6]Transportation Research Record,No.1453,Intelligent Transportation System:Evaluation,Driver Behavior,and Artificial Intelligence,1994.
[7]王川久,杨兆升.城市交通流诱导系统的框架研究[J].吉林大学学报(工学版),2003,33(3):14-18.
[8]杨兆升,王爽等.基础交通信息融合方法综述[J].公路交通科技.2006(3):111-116.
[9]杨兆升.关于智能运输系统的关键理论[J].交通运输工程学报:2001(1)65-68.
[10]向红艳,朱顺应.路段短时交通量预测自适应控制法[J].公路交通科技.2006(1):116-119.
[11]Jessica Anderson,Michael Bell.Travel time estimation in urban road networks[J].IEEE.1998(3):111-116.
[12]姚丽亚等.基于实时交通信息的行程时间估算和路径选择分析[J].公路交通科技.2006(11):86-89.
[13]Hao Liu,Henk J.Prediction of Urban Travel Times with Intersection Delays[J].IEEE Conference on Intelligent Transportation Systems.2005(9):1062-1067.
[14]John Rice,Erik van Zwet.A Simple and Effective Method for Predicting Travel Times on Freeways.IEEE TRANSACTION ON INTELLIGENT TRANSPORTATION SYSTEM.2004(5)200-207.
[15]Kim,Choi.Evaluation of Estimated Shortest Travel Time Using Traffic Network Simulation.1999IEEE,Ⅳ:586-588.
[16]朱中,杨兆升.基于卡尔曼滤波理论的实时行程时间预测模型[J].系统工程理论与实践,1999,9:74-78.
[17]温惠英等.基于灰色关联分析的路段行程时间卡尔曼滤波预测算法[J].华南理工大学学报.2006(9):66-70.
[18]Brian L Smith and Michael J Demertsky.Short-term Traffic Flow Prediction:Neural Network Approach.Transportation Research Record 1453,Washington,D.C.TRB,1993.
[19]杨兆升,朱中.基于BP神经网络的路径行程时间实时预测模型[J].系统工程理论与实践,1999,8:59-64.
[20]温凯歌,曲什茹,王娇.系统灰理论在交通量预测中的应用[J].沈阳理工大学学报,2006,25(2):11-14.
[21]翁剑成等.基于非参数回归的快速路行程速度短期预测算法[J].公路交通科技.2007(3):93-98.
[22]徐天东,郝媛,孙立军.一种新的城市快速路行程时间预测技术[J].交通与计算机,2007,25(2):83-87.
[23]刘廷新,李振宇.指数平滑法在交通参数短期预测中的应用[J].山东交通学院学报,2002,10(3):29-32.
[24]胡佩锋,袁振洲.一种基于周期相似特性的实时交通量预测模型[J].公路交通科技,2007,1(2):145-149.
[25]杨晓光,曾松.行程时间信息质量与接受水平关系分析[J].中国公路学报,2002,15(4):78-81.
[26]杨晓光,蔡润林,庄斌.基于车牌自动识别系统的城市道路行程时间预测算法[J].交通与计算机,2005,23(3):29-32.
[27]高作刚,朱健等.一种行程时间检测和预测的实现方法[J].交通与计算机,2004,22(4):37-39.
[28]胡上序,陈德钊.观测数据的分析与处理[M].浙江大学出版社,1996,3.
[29]李时.应用统计学[M].清华大学出版社&北京交通大学出版社,2005,9。
[30]张可,刘小明,王笑京.车辆自动导航的路线优化系统研究[J].系统工程,2001,19(2):48-53
[31]John Wiley,Sons Ltd.Advanced Vehicle and Infrastructure System.C.O.Nwagboso,1997.
[32]Fisher P F.A Primer of Geographic Search Using Artificial Intelligence[J].Computer &Geosciences,1990,16(6):753-776.
[33]Deon,Pang C Y.Shortest Path Algorithms:Taxonomy and Annotation[J].Networks,1984,14:275-323.
[34]Zhan F B,Noon C E.Shortest Path Algorithms:An Evaluation Using Real Road Networks[J].Transportation Science,1998,32(1):65-73.
[35]CherkassKy B V,Goldberg A V,Radzik T.Shortest paths algorithms:Theory and experimental evaluation.Mathematical Programming,1996,73:129-174.
[36]王峰,游志胜等.Dijkstra及基于Dijkstra的前N条最短路径算法在智能交通系统中的应用[J].计算机应用研究,2006,9:203-205.
[37]乐阳,龚健雅.Dijkstra最短路径算法的一种高效率实现[J].武汉测绘科技大学学报,1999,24(3):209-212.
[38]Pallottion S.Shortest Path Methods:Complexity,Interrelation and New Propositions[J].Networks,1984,14:257-267.
[39]陈洁,陆峰.一种基于双端队列的交通网络最短路径Pallottino优化算法[J].中国图像图形学报,2006,11(3):419-424.
[40]邹亮,徐建闽,朱玲湘..A*算法改进及其在动态最短路径问题中的应用[J].深圳大学学报理工版,2007,24(1):32-35.
[41]段莉琼,朱建军等.改进的最短路径搜索A*算法的高效实现[J].海洋测绘,2004,24(5):20-22.
[42]谭国真,高文.时间依赖的网络中最小时间路径算法[J].计算机学报,2002,25(2):165-172
[43]Kaufman D E,Smith R L.Fastest path in time-dependent networks for intelligent vehicle-highway systems application,ⅣHS Journal,1993,11(1):1-11.
[44]Kiseko Sung,Michael G H Bell,Seong.Shortest path in a network with time-dependent flow speed[J].European Journal of Operational Research,2000,121(1):32-39.
[45]陆峰.最短路径算法:分类体系与研究进展[J].测绘学报,2001,30(3):269-275.
[46]隽志才,倪安宁等.两种策略下的最短路径并行算法研究与实现[J].系统工程理论方法应用,2006,15(2):123-127.
[47]林澜,闫春钢等.动态网络最短路问题的复杂性与近似算法[J].计算机学报,2007,30(4):608-614.
[48]Jinsong Tan,Hon.Least-Cost Path in Public Transportation Systems with Fare Rebates that are Path and Time-Dependent.IEEE Intelligent Transportation Systems Conference Washington.2004,10:1000-1005.
[49]杨兆升,初连禹.动态路径诱导系统的研究进展[J].公路交通科技,2000,17(1):34-38.
[50]朱文兴,贾磊等.城市交通网络中的路径优化研究[J].山东大学学报,2005,35(1):74-77.
[2]杨兆升.城市交通流诱导系统理论与模型[M].人民交通出版社,2000,1.
[3]杨兆升.城市交通流诱导系统[M].中国铁道出版社,2004,11.
[4]交通部公路科学研究所.中国智能交通系统体系框架研究.
[5]The U.S Department of Transportation,ITS AMERICA,ITS Architecture Development Program,phase 1,1994.
[6]Transportation Research Record,No.1453,Intelligent Transportation System:Evaluation,Driver Behavior,and Artificial Intelligence,1994.
[7]王川久,杨兆升.城市交通流诱导系统的框架研究[J].吉林大学学报(工学版),2003,33(3):14-18.
[8]杨兆升,王爽等.基础交通信息融合方法综述[J].公路交通科技.2006(3):111-116.
[9]杨兆升.关于智能运输系统的关键理论[J].交通运输工程学报:2001(1)65-68.
[10]向红艳,朱顺应.路段短时交通量预测自适应控制法[J].公路交通科技.2006(1):116-119.
[11]Jessica Anderson,Michael Bell.Travel time estimation in urban road networks[J].IEEE.1998(3):111-116.
[12]姚丽亚等.基于实时交通信息的行程时间估算和路径选择分析[J].公路交通科技.2006(11):86-89.
[13]Hao Liu,Henk J.Prediction of Urban Travel Times with Intersection Delays[J].IEEE Conference on Intelligent Transportation Systems.2005(9):1062-1067.
[14]John Rice,Erik van Zwet.A Simple and Effective Method for Predicting Travel Times on Freeways.IEEE TRANSACTION ON INTELLIGENT TRANSPORTATION SYSTEM.2004(5)200-207.
[15]Kim,Choi.Evaluation of Estimated Shortest Travel Time Using Traffic Network Simulation.1999IEEE,Ⅳ:586-588.
[16]朱中,杨兆升.基于卡尔曼滤波理论的实时行程时间预测模型[J].系统工程理论与实践,1999,9:74-78.
[17]温惠英等.基于灰色关联分析的路段行程时间卡尔曼滤波预测算法[J].华南理工大学学报.2006(9):66-70.
[18]Brian L Smith and Michael J Demertsky.Short-term Traffic Flow Prediction:Neural Network Approach.Transportation Research Record 1453,Washington,D.C.TRB,1993.
[19]杨兆升,朱中.基于BP神经网络的路径行程时间实时预测模型[J].系统工程理论与实践,1999,8:59-64.
[20]温凯歌,曲什茹,王娇.系统灰理论在交通量预测中的应用[J].沈阳理工大学学报,2006,25(2):11-14.
[21]翁剑成等.基于非参数回归的快速路行程速度短期预测算法[J].公路交通科技.2007(3):93-98.
[22]徐天东,郝媛,孙立军.一种新的城市快速路行程时间预测技术[J].交通与计算机,2007,25(2):83-87.
[23]刘廷新,李振宇.指数平滑法在交通参数短期预测中的应用[J].山东交通学院学报,2002,10(3):29-32.
[24]胡佩锋,袁振洲.一种基于周期相似特性的实时交通量预测模型[J].公路交通科技,2007,1(2):145-149.
[25]杨晓光,曾松.行程时间信息质量与接受水平关系分析[J].中国公路学报,2002,15(4):78-81.
[26]杨晓光,蔡润林,庄斌.基于车牌自动识别系统的城市道路行程时间预测算法[J].交通与计算机,2005,23(3):29-32.
[27]高作刚,朱健等.一种行程时间检测和预测的实现方法[J].交通与计算机,2004,22(4):37-39.
[28]胡上序,陈德钊.观测数据的分析与处理[M].浙江大学出版社,1996,3.
[29]李时.应用统计学[M].清华大学出版社&北京交通大学出版社,2005,9。
[30]张可,刘小明,王笑京.车辆自动导航的路线优化系统研究[J].系统工程,2001,19(2):48-53
[31]John Wiley,Sons Ltd.Advanced Vehicle and Infrastructure System.C.O.Nwagboso,1997.
[32]Fisher P F.A Primer of Geographic Search Using Artificial Intelligence[J].Computer &Geosciences,1990,16(6):753-776.
[33]Deon,Pang C Y.Shortest Path Algorithms:Taxonomy and Annotation[J].Networks,1984,14:275-323.
[34]Zhan F B,Noon C E.Shortest Path Algorithms:An Evaluation Using Real Road Networks[J].Transportation Science,1998,32(1):65-73.
[35]CherkassKy B V,Goldberg A V,Radzik T.Shortest paths algorithms:Theory and experimental evaluation.Mathematical Programming,1996,73:129-174.
[36]王峰,游志胜等.Dijkstra及基于Dijkstra的前N条最短路径算法在智能交通系统中的应用[J].计算机应用研究,2006,9:203-205.
[37]乐阳,龚健雅.Dijkstra最短路径算法的一种高效率实现[J].武汉测绘科技大学学报,1999,24(3):209-212.
[38]Pallottion S.Shortest Path Methods:Complexity,Interrelation and New Propositions[J].Networks,1984,14:257-267.
[39]陈洁,陆峰.一种基于双端队列的交通网络最短路径Pallottino优化算法[J].中国图像图形学报,2006,11(3):419-424.
[40]邹亮,徐建闽,朱玲湘..A*算法改进及其在动态最短路径问题中的应用[J].深圳大学学报理工版,2007,24(1):32-35.
[41]段莉琼,朱建军等.改进的最短路径搜索A*算法的高效实现[J].海洋测绘,2004,24(5):20-22.
[42]谭国真,高文.时间依赖的网络中最小时间路径算法[J].计算机学报,2002,25(2):165-172
[43]Kaufman D E,Smith R L.Fastest path in time-dependent networks for intelligent vehicle-highway systems application,ⅣHS Journal,1993,11(1):1-11.
[44]Kiseko Sung,Michael G H Bell,Seong.Shortest path in a network with time-dependent flow speed[J].European Journal of Operational Research,2000,121(1):32-39.
[45]陆峰.最短路径算法:分类体系与研究进展[J].测绘学报,2001,30(3):269-275.
[46]隽志才,倪安宁等.两种策略下的最短路径并行算法研究与实现[J].系统工程理论方法应用,2006,15(2):123-127.
[47]林澜,闫春钢等.动态网络最短路问题的复杂性与近似算法[J].计算机学报,2007,30(4):608-614.
[48]Jinsong Tan,Hon.Least-Cost Path in Public Transportation Systems with Fare Rebates that are Path and Time-Dependent.IEEE Intelligent Transportation Systems Conference Washington.2004,10:1000-1005.
[49]杨兆升,初连禹.动态路径诱导系统的研究进展[J].公路交通科技,2000,17(1):34-38.
[50]朱文兴,贾磊等.城市交通网络中的路径优化研究[J].山东大学学报,2005,35(1):74-77.