城市道路路段行程时间估计及融合方法研究
|
摘要
路段行程时间是反映交通流运行的重要指标,是构建动态交通信息服务系统、信号协调控制系统以及交通诱导系统等ITS子系统的重要基础,因此,其估计的准确与否对城市道路运行系统有着重要意义。
本文对城市路段行程时间的估计从固定检测信息和GPS浮动车检测信息两种检测方式入手。首先利用两种检测方式检测到的基础交通信息分别估计路段行程时间,然后再将本文提出的两系统行程时间估计方法的影响条件组合分类,把两系统估计结果进行有效的融合处理。
在利用固定检测器信息估计路段行程时间的方法研究中,将路段行程时间分为路段行驶时间、车辆通过交叉口时间,交叉口延误时间三部分分别进行估计。研究重点放在路段行驶时间和交叉口延误时间的估计上,对前者的研究,主要是对路段下游检测速度进行修正处理,并利用修正的速度估计路段行驶速度,从而估计路段行驶时间;对后者的研究,以HCM方法为主体方法,根据本文的研究内容进行相应的处理;最后深入分析了固定检测数据估计路段行程时间的影响因素。在利用GPS浮动车信息估计路段行程时间研究中,主要是对其中重要环节地图匹配和路段边界点通过时间进行研究。在行程时间融合方法研究中,根据不同的估计影响因素分类组合,在不同的融合条件下采用自适应加权平均融合方法进行融合处理。对文中提出的估计方法,均采用了VISSIM4.20仿真软件建立了仿真路网进行了模拟验证,结果令人满意。
With the development of society, the problem of traffic, such as traffic jam, traffic block, traffic accident, traffic pollution, has become bottleneck for the urban development, more and more. This matter not just trouble our country, even in the developed countries, the traffic problem is a very practical and serous problem. Just traffic jam costs the United States an estimated 100 billion per year. To confront with the traffic problem, people have been scratching our heads for a solution to the problem. Intelligent transportation system is the best acknowledged method to solve problems in traffic field at present. Travel time is a core parameter in intelligent transportation system. Real.time or quasi.Real.time travel time is a foundational base to subsystems of Intelligent Transport System (ITS), such as dynamic traffic information service system, signals coordination and control system and traffic guidance system, so, the effective estimation of travel time is a key problem. The accuracy of travel time estimation plays an important role in architecture of subsystems of Intelligent Transport System, the decrease of the losses of traffic problem and to improve the efficiency of traffic network. The estimation of travel time has been attracted more and more attention. This thesis briefly presents the present situation of the research on the estimation of travel time home and abroad,then, Upon the station of our urban traffic and the shortages of estimation methods nowadays,launches research.
This dissertation is supported by“Technology of Feature Extraction and Integration of Regional Traffic System State”launched by the Chinese National Programs for High Technology Research and Development. Based on the characters of our urban traffic and The detection of traffic information, and integrating the estimation methods nowadays,development new method fitting our cities qualifications. In addition,to increase the accuracy of estimation,Research on the fusion of the estimated result. This research consists of five chapters. The main works and contributions are described as follows:
Chapter 1: Introduction. This part firstly introduce the project the research based on and the research backgrounds of the dissertation, then looks back to the studied history of the estimation methods f travel time, and analyses the station of our urban traffic and the shortages of estimation methods nowadays, and illuminates the research purpose and significance of this dissertation. At last, provides the main contents and chapter arrangement of this paper.
Chapter 2: the Study on the Estimation of Travel Time Based on Fixed Detectors. Before the estimation, there is a pivotal step, pre.processing of the traffic information Detected by fixed detectors to do first. Then,based on the need of the study in this chapter, defining the urban Road Section. In the estimation of travel time in this chapter, the travel time was divided into three part, running time, queue delay and crossing intersection time. Each part uses corresponding estimations. At last, using the simulator software, VISSIM4.20 makes a simulation to validate the effectiveness of the estimation in this Chapter, and making a profound analysis of the factors that influence the estimation of travel time.
Chapter 3: the Study on the Estimation of Travel Time Based on Floating Cars. In this Chapter, the author introduces briefly Global Positioning System and Geographical Information System, and mainly analyses the system error of this two systems. Before estimating travel time using GPS information, pre.processing of GPS information based on floating cars should be do first,to weed out the fault information. The emphasis of this Chapter was map matching, estimation of the passed-time of road boundary point, and interrupted stop, and proposed new methodes to ameliorate this key steps. Using the actual taxi data proved the reasonableness and effectiveness of individual vehicle travel-time estimation method. At last, this part made a profound analysis of the factors,the sample size of floating cars and traffic state that influence the estimation of travel time by floating cars.
Chapter 4: the Study on Estimation of Travel Time Based on Information Fusion. At the first, this Chapter analyzed the characteristic of estimation method presented in the paper and the necessity of data fusion, and improved the data fusion method of travel time estimating. In this paper, using adaptive weighted averaging fusion method to fuse the travel time that was estimated by information of fixed detectors and GPS floating cars. The paper ensured the weight factors using improved optimal estimation method. And based on the different factors, the travel time fusion was divided into eleven distribution types. Finally, this part validated the effectiveness of the fusion method.
Chapter 5: Summary and Prospect. A summation was made to generalize the work in this paper. And suggestions were proposed for further research and improvement.
本文对城市路段行程时间的估计从固定检测信息和GPS浮动车检测信息两种检测方式入手。首先利用两种检测方式检测到的基础交通信息分别估计路段行程时间,然后再将本文提出的两系统行程时间估计方法的影响条件组合分类,把两系统估计结果进行有效的融合处理。
在利用固定检测器信息估计路段行程时间的方法研究中,将路段行程时间分为路段行驶时间、车辆通过交叉口时间,交叉口延误时间三部分分别进行估计。研究重点放在路段行驶时间和交叉口延误时间的估计上,对前者的研究,主要是对路段下游检测速度进行修正处理,并利用修正的速度估计路段行驶速度,从而估计路段行驶时间;对后者的研究,以HCM方法为主体方法,根据本文的研究内容进行相应的处理;最后深入分析了固定检测数据估计路段行程时间的影响因素。在利用GPS浮动车信息估计路段行程时间研究中,主要是对其中重要环节地图匹配和路段边界点通过时间进行研究。在行程时间融合方法研究中,根据不同的估计影响因素分类组合,在不同的融合条件下采用自适应加权平均融合方法进行融合处理。对文中提出的估计方法,均采用了VISSIM4.20仿真软件建立了仿真路网进行了模拟验证,结果令人满意。
With the development of society, the problem of traffic, such as traffic jam, traffic block, traffic accident, traffic pollution, has become bottleneck for the urban development, more and more. This matter not just trouble our country, even in the developed countries, the traffic problem is a very practical and serous problem. Just traffic jam costs the United States an estimated 100 billion per year. To confront with the traffic problem, people have been scratching our heads for a solution to the problem. Intelligent transportation system is the best acknowledged method to solve problems in traffic field at present. Travel time is a core parameter in intelligent transportation system. Real.time or quasi.Real.time travel time is a foundational base to subsystems of Intelligent Transport System (ITS), such as dynamic traffic information service system, signals coordination and control system and traffic guidance system, so, the effective estimation of travel time is a key problem. The accuracy of travel time estimation plays an important role in architecture of subsystems of Intelligent Transport System, the decrease of the losses of traffic problem and to improve the efficiency of traffic network. The estimation of travel time has been attracted more and more attention. This thesis briefly presents the present situation of the research on the estimation of travel time home and abroad,then, Upon the station of our urban traffic and the shortages of estimation methods nowadays,launches research.
This dissertation is supported by“Technology of Feature Extraction and Integration of Regional Traffic System State”launched by the Chinese National Programs for High Technology Research and Development. Based on the characters of our urban traffic and The detection of traffic information, and integrating the estimation methods nowadays,development new method fitting our cities qualifications. In addition,to increase the accuracy of estimation,Research on the fusion of the estimated result. This research consists of five chapters. The main works and contributions are described as follows:
Chapter 1: Introduction. This part firstly introduce the project the research based on and the research backgrounds of the dissertation, then looks back to the studied history of the estimation methods f travel time, and analyses the station of our urban traffic and the shortages of estimation methods nowadays, and illuminates the research purpose and significance of this dissertation. At last, provides the main contents and chapter arrangement of this paper.
Chapter 2: the Study on the Estimation of Travel Time Based on Fixed Detectors. Before the estimation, there is a pivotal step, pre.processing of the traffic information Detected by fixed detectors to do first. Then,based on the need of the study in this chapter, defining the urban Road Section. In the estimation of travel time in this chapter, the travel time was divided into three part, running time, queue delay and crossing intersection time. Each part uses corresponding estimations. At last, using the simulator software, VISSIM4.20 makes a simulation to validate the effectiveness of the estimation in this Chapter, and making a profound analysis of the factors that influence the estimation of travel time.
Chapter 3: the Study on the Estimation of Travel Time Based on Floating Cars. In this Chapter, the author introduces briefly Global Positioning System and Geographical Information System, and mainly analyses the system error of this two systems. Before estimating travel time using GPS information, pre.processing of GPS information based on floating cars should be do first,to weed out the fault information. The emphasis of this Chapter was map matching, estimation of the passed-time of road boundary point, and interrupted stop, and proposed new methodes to ameliorate this key steps. Using the actual taxi data proved the reasonableness and effectiveness of individual vehicle travel-time estimation method. At last, this part made a profound analysis of the factors,the sample size of floating cars and traffic state that influence the estimation of travel time by floating cars.
Chapter 4: the Study on Estimation of Travel Time Based on Information Fusion. At the first, this Chapter analyzed the characteristic of estimation method presented in the paper and the necessity of data fusion, and improved the data fusion method of travel time estimating. In this paper, using adaptive weighted averaging fusion method to fuse the travel time that was estimated by information of fixed detectors and GPS floating cars. The paper ensured the weight factors using improved optimal estimation method. And based on the different factors, the travel time fusion was divided into eleven distribution types. Finally, this part validated the effectiveness of the fusion method.
Chapter 5: Summary and Prospect. A summation was made to generalize the work in this paper. And suggestions were proposed for further research and improvement.
引文
[1]蒋捷,陈军.导航地图数据库[M].北京:科学出版社, 2005:1-7.
[2] LELITHA DEVI VANAJAKSHI.Estmation and prediction of travel time loop detector data for intelligent transportation systems applications [D].Texas:A&M University,2004.
[3]何兆成,赵建明,王振波,韦清波.城市信号控制网络中的路段行程时间估计法[J].交通运输工程学报,2008,8(4):95-98.
[4]杨兆升,董升,李善梅,刘雪杰,朱伟权.基于占有率数据的模式匹配行程时间预测模型研究[C].2006第二节智能交通年会论文集,2006.
[5]杨少辉,王殿海,王英平,董斌.最小二乘拟合法确定行程时间[J]公路交通科技,2006 ,23(9):90-93.
[6]初连禹,杨兆升.基于模糊神经网络的实时路段行程时间估计[J].系统工程理论与实践,2000:11,111-116.
[7]张和生,张毅,胡东成.路段平均行程时间估计方法[J].交通运输工程学报. 2008:8(1),89-96.
[8]姚智胜,邵春福,熊志华.支持向量机在路段行程时间预测中的应用研究[J].公路交通科技,2007,24(9),96-99.
[9]张保存,杨晓光,严新平.基于浮动车的交通信息采集系统研究[J].交通与计算机,2006:24(5),31-34。
[10]邹亮,徐建闽,朱玲湘,温慧英.基于浮动车移动检测与感应线圈融合技术的行程时间估计模型[J].公路交通科技,2007,24(6),114-117.
[11]张和生,张毅,温慧敏,胡成东.利用GPS数据估计路段的平均行程时间[J].吉林大学学报(工学版):2007, 37(3),533-537.
[12]王殿海.交通流理论[M].北京:人民交通出版社,2002.
[13]杨兆升.基础交通信息融合技术及应用[M].北京:中国铁道出版社,2005:39-42,107-110.
[14]姜桂艳.道路交通状态判别技术与应用[M].北京:人民交通出版社,2004:103-122,178—212.
[15]李江等.交通工程学[M].北京:人民交通出版社,2002:44-48,129-132.
[16]宫兴斌,姚丹亚.一种基于单线圈速度估计的简单模型方法[J].计算机工程与应用,2004,35:226-228.
[17]余田,刘建伟,曹泉,向怀坤.基于单线圈的车速检测算法研究[J].交通与计算机,2006,24(6):68-70.
[18] Athol P.Interdependence of certain operational characteristics within a moving traffic stream[J].Highway Research Record, 1965, 72: 58-87.
[19] Wang Yin hai, Nihan N L. Can single-loop detectors do the work of dual-loop detectors?[J]. Journal of Transportation Engineering, 2003, 129(2):169-176.
[20]杨晓光,庄斌,李克平.信号交叉口饱和流率和启动延误的影响分析[J].同济大学学报(自然科学版),2006,34(6):738-743.
[21]荣建,何民,陈春妹.信号交叉口排队长度动态计算方法研究[J].中国公路学报,2002,15(3):101-103.
[22]陈绍宽,郭谨一,王璇,毛宝华.交叉口延误计算方法的比较[J].北京交通大学学报,2005,29(3): 77-80.
[23]庄焰,曾文佳.信号交叉口延误计算模型研究[J].深圳大学学报理工版,2006 23(4):309-313.
[24]王嘉祺,程建川,王昊.信号交叉口增量延误分析[J].交通与计算机, 2005,23(5):13-16.
[25] Transportation Research Board.Highway Capacity Manual 2000 [M].Washi- ngton:National Research Council,2000:14-149.
[26]杜歆,李宏东,顾伟康.一种提高差分GPS基准站定位精度的新方法[J].浙江大学学报(理学版), 2002,29(6):637-642.
[27] NRC, National Research Council, 2002, Collecting, processing and integrating PS data into GIS[C].NCHRP Synthesis 301.National Academy Press; Washington D.C.
[28]陈述彭,鲁学军.地理信息系统导轮[M].北京:科学出版社,1999.
[29]姚娜.GIS、MapInfo与MapBasic学习教程[M].北京:北京大学出版社,2000:2-5.
[30]杨兆升.城市交通流诱导系统[M].北京:中国铁道出版社2004 :109-116.
[31]赵亦林.车辆定位与导航系统[M]北京:电子工业出版社,1998.
[32]李世平.地理信息系统空间数据的误差分析与质量控制[J].科学管理研究,2004,22(2):164-165。
[33]杜景龙,姜俐平.地理信息系统的数据精度分析[J].东北测绘(季刊)2001,24(1):27-28.
[34]张振辉,崔铁军,姚惠敏.车辆导航系统中地图匹配新算法[J].海洋测绘, 2006,26(2):55-58.
[35]李广峰.GPS、GIS在车辆定位导航中的应用研究[D].西安:西北工业大学,2003.
[36]李欣,俞斌.一种改进的地图匹配技术在车载导航系统中的应用[J].测控系统2008,5:60-65.
[37]周璞,刘卫宁,孙棣华.基于路网拓扑结构的无方向参数地图匹配算法[J].计算机工程与应用,2006,33:188-190.
[38]刘召芹、肖成刚、谢彩香. GPS/DR导航中的智能地图匹配方法研究[J].地理空间信息,2006,4(5):64-65.
[39]付梦印,李杰,邓志红.一种适用于车辆导航系统的快速地图匹配算法[J].北京理工大学学报, ,25(3):225-229.
[40] BERNSTEIN D , KORNHAUSER A. An introduction to map matching for personalnavigation assistants[EB/OL].http://ntl.bts.gov/cardview.cfm? Docid=11085.
[41]董均宇.基于GPS浮动车的城市路段平均速度估计技术研究[D].重庆:重庆大学自动化学院,2006.
[42]全永燊.城市交通控制[M].北京:人民交通出版社,1989:64-69.
[43] Lawrence A , Klein . Sensor and Data Fusion Concepts and Applications[M].Washington:SPIE Optical Engineering Press,1999.
[44]贾沛璋,朱征桃.最优估计及其应用[M].北京:科学出版社,1984.
[45]滕召胜.智能检测系统与数据融合[M].北京:机械工业出版社,2000.
[46]杨兆升,冯金巧.基于模糊综合推理的道路交通时间检测算法[C].2003全国智能交通系统交通信息采集与融合技术研讨会,2003:61-66.
[47]唐汝磉,钱寒峰.智能交通信息采集中数据融合技术探讨[J].科技咨询导报,2007,26:17-18.
[48]刘智勇,李水友.基于信息融合技术的交通量检测方法[J].公路交通科技,2003(2):81-84.
[49]王炯琦,周海银,吴翊.基于最优估计的数据融合理论[J].应用数学,2007,20(2):392-399.
[2] LELITHA DEVI VANAJAKSHI.Estmation and prediction of travel time loop detector data for intelligent transportation systems applications [D].Texas:A&M University,2004.
[3]何兆成,赵建明,王振波,韦清波.城市信号控制网络中的路段行程时间估计法[J].交通运输工程学报,2008,8(4):95-98.
[4]杨兆升,董升,李善梅,刘雪杰,朱伟权.基于占有率数据的模式匹配行程时间预测模型研究[C].2006第二节智能交通年会论文集,2006.
[5]杨少辉,王殿海,王英平,董斌.最小二乘拟合法确定行程时间[J]公路交通科技,2006 ,23(9):90-93.
[6]初连禹,杨兆升.基于模糊神经网络的实时路段行程时间估计[J].系统工程理论与实践,2000:11,111-116.
[7]张和生,张毅,胡东成.路段平均行程时间估计方法[J].交通运输工程学报. 2008:8(1),89-96.
[8]姚智胜,邵春福,熊志华.支持向量机在路段行程时间预测中的应用研究[J].公路交通科技,2007,24(9),96-99.
[9]张保存,杨晓光,严新平.基于浮动车的交通信息采集系统研究[J].交通与计算机,2006:24(5),31-34。
[10]邹亮,徐建闽,朱玲湘,温慧英.基于浮动车移动检测与感应线圈融合技术的行程时间估计模型[J].公路交通科技,2007,24(6),114-117.
[11]张和生,张毅,温慧敏,胡成东.利用GPS数据估计路段的平均行程时间[J].吉林大学学报(工学版):2007, 37(3),533-537.
[12]王殿海.交通流理论[M].北京:人民交通出版社,2002.
[13]杨兆升.基础交通信息融合技术及应用[M].北京:中国铁道出版社,2005:39-42,107-110.
[14]姜桂艳.道路交通状态判别技术与应用[M].北京:人民交通出版社,2004:103-122,178—212.
[15]李江等.交通工程学[M].北京:人民交通出版社,2002:44-48,129-132.
[16]宫兴斌,姚丹亚.一种基于单线圈速度估计的简单模型方法[J].计算机工程与应用,2004,35:226-228.
[17]余田,刘建伟,曹泉,向怀坤.基于单线圈的车速检测算法研究[J].交通与计算机,2006,24(6):68-70.
[18] Athol P.Interdependence of certain operational characteristics within a moving traffic stream[J].Highway Research Record, 1965, 72: 58-87.
[19] Wang Yin hai, Nihan N L. Can single-loop detectors do the work of dual-loop detectors?[J]. Journal of Transportation Engineering, 2003, 129(2):169-176.
[20]杨晓光,庄斌,李克平.信号交叉口饱和流率和启动延误的影响分析[J].同济大学学报(自然科学版),2006,34(6):738-743.
[21]荣建,何民,陈春妹.信号交叉口排队长度动态计算方法研究[J].中国公路学报,2002,15(3):101-103.
[22]陈绍宽,郭谨一,王璇,毛宝华.交叉口延误计算方法的比较[J].北京交通大学学报,2005,29(3): 77-80.
[23]庄焰,曾文佳.信号交叉口延误计算模型研究[J].深圳大学学报理工版,2006 23(4):309-313.
[24]王嘉祺,程建川,王昊.信号交叉口增量延误分析[J].交通与计算机, 2005,23(5):13-16.
[25] Transportation Research Board.Highway Capacity Manual 2000 [M].Washi- ngton:National Research Council,2000:14-149.
[26]杜歆,李宏东,顾伟康.一种提高差分GPS基准站定位精度的新方法[J].浙江大学学报(理学版), 2002,29(6):637-642.
[27] NRC, National Research Council, 2002, Collecting, processing and integrating PS data into GIS[C].NCHRP Synthesis 301.National Academy Press; Washington D.C.
[28]陈述彭,鲁学军.地理信息系统导轮[M].北京:科学出版社,1999.
[29]姚娜.GIS、MapInfo与MapBasic学习教程[M].北京:北京大学出版社,2000:2-5.
[30]杨兆升.城市交通流诱导系统[M].北京:中国铁道出版社2004 :109-116.
[31]赵亦林.车辆定位与导航系统[M]北京:电子工业出版社,1998.
[32]李世平.地理信息系统空间数据的误差分析与质量控制[J].科学管理研究,2004,22(2):164-165。
[33]杜景龙,姜俐平.地理信息系统的数据精度分析[J].东北测绘(季刊)2001,24(1):27-28.
[34]张振辉,崔铁军,姚惠敏.车辆导航系统中地图匹配新算法[J].海洋测绘, 2006,26(2):55-58.
[35]李广峰.GPS、GIS在车辆定位导航中的应用研究[D].西安:西北工业大学,2003.
[36]李欣,俞斌.一种改进的地图匹配技术在车载导航系统中的应用[J].测控系统2008,5:60-65.
[37]周璞,刘卫宁,孙棣华.基于路网拓扑结构的无方向参数地图匹配算法[J].计算机工程与应用,2006,33:188-190.
[38]刘召芹、肖成刚、谢彩香. GPS/DR导航中的智能地图匹配方法研究[J].地理空间信息,2006,4(5):64-65.
[39]付梦印,李杰,邓志红.一种适用于车辆导航系统的快速地图匹配算法[J].北京理工大学学报, ,25(3):225-229.
[40] BERNSTEIN D , KORNHAUSER A. An introduction to map matching for personalnavigation assistants[EB/OL].http://ntl.bts.gov/cardview.cfm? Docid=11085.
[41]董均宇.基于GPS浮动车的城市路段平均速度估计技术研究[D].重庆:重庆大学自动化学院,2006.
[42]全永燊.城市交通控制[M].北京:人民交通出版社,1989:64-69.
[43] Lawrence A , Klein . Sensor and Data Fusion Concepts and Applications[M].Washington:SPIE Optical Engineering Press,1999.
[44]贾沛璋,朱征桃.最优估计及其应用[M].北京:科学出版社,1984.
[45]滕召胜.智能检测系统与数据融合[M].北京:机械工业出版社,2000.
[46]杨兆升,冯金巧.基于模糊综合推理的道路交通时间检测算法[C].2003全国智能交通系统交通信息采集与融合技术研讨会,2003:61-66.
[47]唐汝磉,钱寒峰.智能交通信息采集中数据融合技术探讨[J].科技咨询导报,2007,26:17-18.
[48]刘智勇,李水友.基于信息融合技术的交通量检测方法[J].公路交通科技,2003(2):81-84.
[49]王炯琦,周海银,吴翊.基于最优估计的数据融合理论[J].应用数学,2007,20(2):392-399.