基于浮动车数据的城市交通流信息感知方法研究
发布时间：2018-10-12 13:23
【摘要】：当今社会,城市化进程的速度已经达到难以置信的程度,越来越多的国家已经意识到了交通对市场经济发展的重要性,各国纷纷将交通建设作为国家重点建设的内容,更为重要的是,人民生活水平的提高以及城市的经济发展都与交通有着非常大的关系,伴随着国民经济的快速发展,人民生活质量稳步提升,也就意味着社会生产生活成本的降低。在此同时,也带来了众多严重的社会问题,包括大气污染、水污染等等,其中道路交通拥堵现象日益严峻,已经成为阻碍经济持续发展的命脉。随着交通难题的到来,世界各国纷纷意识到单纯依靠传统的交通控制和交通诱导方式已经很难应对越来越复杂多样化的交通问题。智能交通系统正是在这种大背景下应运而生,各国投入智能交通系统上的研究力度越来越大,在解决交通问题方面提出了许多创新高效的新理论与实际结果。针对以往交通数据采集技术的瓶颈,本文首先介绍了浮动车数据作为当今智能交通系统首选的数据源的优点,提出由程序方法获取城市道路网络数据的方法。在数据预处理和分析的基础上,首先对居民出行规律进行统计,分析得出基本符合无标度特性的出行规律。针对现有浮动车地图匹配算法在浮动车数据采样率较低时错误率较高的情形,本文提出一种全局投票地图匹配算法。该算法在浮动车GPS轨迹数据的基础上,考虑道路网络的拓扑结构和不同距离的相邻的GPS轨迹点对地图匹配过程的影响,在低采样率的数据输入能够取得较高的准确率。最后,针对浮动车本身的波动特性,本文在分析交通流信息中三大参数之间关系的基础上,提出一种考虑加权的路段平均速度估计模型,模型避免了浮动车数据的波动性和少量异常数据对路段平均速度的错误影响,得出更能反映道路使用情况的路段平均速度。
[Abstract]:In today's society, the pace of urbanization has reached an incredible degree. More and more countries have realized the importance of traffic to the development of the market economy. More importantly, the improvement of the people's living standard and the economic development of the city have very great relations with the traffic. With the rapid development of the national economy, the people's quality of life has been steadily improved. It also means that the cost of living and production of society is reduced. At the same time, it also brings many serious social problems, including air pollution, water pollution and so on. With the arrival of traffic problems, countries all over the world have realized that relying solely on traditional traffic control and traffic guidance methods has become difficult to cope with more and more complex and diversified traffic problems. It is against this background that the intelligent transportation system (its) has been put into the research of intelligent transportation system (its) more and more, and many new theories and practical results have been put forward to solve the traffic problems. Aiming at the bottleneck of traffic data acquisition technology, this paper first introduces the advantages of floating vehicle data as the first choice data source of intelligent transportation system, and puts forward a method of obtaining urban road network data by program method. On the basis of data preprocessing and analysis, first of all, the travel law of residents is statistically analyzed, and the travel law which basically accords with the scale-free characteristic is obtained. This paper presents a global voting map matching algorithm in view of the fact that the existing floating vehicle map matching algorithm has a high error rate when the data sampling rate of the floating vehicle is low. Based on the GPS trajectory data of floating vehicle and considering the influence of road network topology and adjacent GPS locus points at different distances on the map matching process, the algorithm can achieve high accuracy in low sampling rate data input. Finally, based on the analysis of the relationship between the three parameters in the traffic flow information, a weighted average speed estimation model is proposed for the fluctuating characteristics of the floating vehicle. The model avoids the fluctuation of floating vehicle data and the wrong influence of a few abnormal data on the average speed of road section, and obtains the average speed of road section which can better reflect the road usage.
【学位授予单位】：浙江工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495
本文编号：2266285
[Abstract]:In today's society, the pace of urbanization has reached an incredible degree. More and more countries have realized the importance of traffic to the development of the market economy. More importantly, the improvement of the people's living standard and the economic development of the city have very great relations with the traffic. With the rapid development of the national economy, the people's quality of life has been steadily improved. It also means that the cost of living and production of society is reduced. At the same time, it also brings many serious social problems, including air pollution, water pollution and so on. With the arrival of traffic problems, countries all over the world have realized that relying solely on traditional traffic control and traffic guidance methods has become difficult to cope with more and more complex and diversified traffic problems. It is against this background that the intelligent transportation system (its) has been put into the research of intelligent transportation system (its) more and more, and many new theories and practical results have been put forward to solve the traffic problems. Aiming at the bottleneck of traffic data acquisition technology, this paper first introduces the advantages of floating vehicle data as the first choice data source of intelligent transportation system, and puts forward a method of obtaining urban road network data by program method. On the basis of data preprocessing and analysis, first of all, the travel law of residents is statistically analyzed, and the travel law which basically accords with the scale-free characteristic is obtained. This paper presents a global voting map matching algorithm in view of the fact that the existing floating vehicle map matching algorithm has a high error rate when the data sampling rate of the floating vehicle is low. Based on the GPS trajectory data of floating vehicle and considering the influence of road network topology and adjacent GPS locus points at different distances on the map matching process, the algorithm can achieve high accuracy in low sampling rate data input. Finally, based on the analysis of the relationship between the three parameters in the traffic flow information, a weighted average speed estimation model is proposed for the fluctuating characteristics of the floating vehicle. The model avoids the fluctuation of floating vehicle data and the wrong influence of a few abnormal data on the average speed of road section, and obtains the average speed of road section which can better reflect the road usage.
【学位授予单位】：浙江工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495
【共引文献】
相关期刊论文 前3条
1 罗峗骞;汤锦辉;赵钟磊;朱永文;董相均;;中国航空交通网络时变和多层次特性分析[J];复杂系统与复杂性科学;2014年04期
2 房翠翠;孙梅;张培培;高安娜;;A comparison of coal supply-demand in China and in the US based on a network model[J];Chinese Physics B;2013年10期
3 刘军;熊庆宇;石欣;王楷;石为人;;Load-redistribution strategy based on time-varying load against cascading failure of complex network[J];Chinese Physics B;2015年07期
相关博士学位论文 前2条
1 苑卫国;微博用户行为分析和网络结构演化的研究[D];北京交通大学;2014年
2 种鹏云;基于复杂网络的危险品运输网络拓扑特性、级联失效机制及抗毁性研究[D];西南交通大学;2015年
相关硕士学位论文 前1条
1 张永振;城市公交网络模型及其基于流量信息的路径搜索算法研究[D];浙江工业大学;2013年
，本文编号：2266285
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2266285.html
