基于浮动车的交通信息服务系统关键技术研究
发布时间：2018-11-19 08:32
【摘要】：拥堵已成为城市生活中的突出问题，交通拥堵除了引起空气污染、噪声污染、能源浪费和交通事故外，还降低了城市生活效率，恶化城市居住环境，降低了城市吸引力。基于浮动车的交通信息服务系统是解决拥堵问题的有效途径，系统通过浮动车提供的位置和时间信息，，估计当前路网中每条道路的交通状态，出行者依据这些信息选择出行路径和出行时间，从而在空间和时间上缓解高峰流量，减少不必要的交通延误和拥堵。 本文旨在解决基于浮动车的交通信息服务系统的关键技术问题，具体包含两方面内容：高密度路网和低采样频率条件下的在线地图匹配算法，以及基于路网分层的人性化实时高效的路径搜索算法。 主辅路并行是深圳市路网的突出特点，这增加了地图匹配算法的难度。本文将在线地图匹配问题转化为局部路网中的最短路径问题。为了建立一种适合深圳市路网特征的地图匹配算法，本文提出了道路交叉口的交换点模型和多点匹配模式，前者提高了算法在道路交叉口的匹配精度，后者对主辅路旁边的定位点有良好的区分度。为了验证算法的精度和效率，在Visual Studio平台上实现了本文算法，在深圳市采集了一组浮动车定位数据并进行了验证。验证结果表明，本文算法不仅具有较高的精度和效率，还能够确定主辅路旁边的定位点的实际位置，具有极大的实际应用价值。 实时性和人性化是路径诱导算法的鲜明要求。为了使输出结果符合人们的选择偏好以及提高路径搜索的效率，本文建立了一种分层的路径搜索模型。模型把起点区域和终点区域的道路合并到顶层路网中，然后在新的顶层路网中计算起点和终点的最短路径。在针对深圳市具体路网的实例分析中，以快速路以及主干道为基础，对深圳市路网进行分区和分层处理，对比了本文分层算法和不分层的一般算法在效率和路径结果方面的差异。结果表明，分层算法效率明显高于一般算法，分层算法结果路径中顶层道路比例高，不同层级道路切换次数少，更符合人们的路径选择偏好。 本文两类算法针对深圳市路网实际状况而提出，并用实际采集或是随机选取的数据进行了验证，算法有较好的精度和效率，具有很高的使用价值。
[Abstract]:Congestion has become a prominent problem in urban life. Traffic congestion not only causes air pollution, noise pollution, energy waste and traffic accidents, but also reduces the efficiency of urban life, deteriorates the urban living environment and reduces the attraction of the city. The traffic information service system based on the floating vehicle is an effective way to solve the congestion problem. The system estimates the traffic state of every road in the current road network through the location and time information provided by the floating vehicle. Based on these information, travelers choose the travel path and travel time to reduce the peak flow in space and time, and reduce unnecessary traffic delays and congestion. This paper aims to solve the key technical problems of traffic information service system based on floating vehicle, including two aspects: the online map matching algorithm under the condition of high density road network and low sampling frequency. And the real-time and efficient path search algorithm based on road network stratification. The parallelism of main and auxiliary roads is a prominent feature of Shenzhen road network, which increases the difficulty of map matching algorithm. In this paper, the online map matching problem is transformed into the shortest path problem in the local road network. In order to establish a map matching algorithm suitable for the road network characteristics of Shenzhen City, this paper proposes a switching point model and a multi-point matching model for road intersections. The former improves the matching accuracy of the algorithm at road intersections. The latter has a good degree of differentiation for the positioning points next to the main and auxiliary roads. In order to verify the accuracy and efficiency of the algorithm, the algorithm is implemented on Visual Studio platform, and a set of floating vehicle positioning data are collected and verified in Shenzhen. The results show that the proposed algorithm not only has high accuracy and efficiency, but also can determine the actual position of the location point next to the main and auxiliary road, which is of great practical value. Real-time and humanization are the distinct requirements of path guidance algorithm. In order to make the output result accord with people's choice preference and improve the efficiency of path search, a hierarchical path search model is established in this paper. The model combines the starting and end roads into the top road network, and then calculates the shortest path of the starting and ending points in the new top road network. In the analysis of the concrete road network in Shenzhen, based on the expressway and the main road, the road network in Shenzhen is divided and stratified. The differences in efficiency and path results between the hierarchical algorithm and the non-delamination algorithm are compared. The results show that the efficiency of the hierarchical algorithm is obviously higher than that of the general algorithm. The result of the hierarchical algorithm results in a high proportion of the top-level roads and less road switching at different levels, which is more in line with people's preference for path selection. In this paper, two kinds of algorithms are proposed according to the actual situation of Shenzhen road network, and verified by the actual data collected or randomly selected. The algorithm has good accuracy and efficiency, and has a high use value.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495
本文编号：2341714
[Abstract]:Congestion has become a prominent problem in urban life. Traffic congestion not only causes air pollution, noise pollution, energy waste and traffic accidents, but also reduces the efficiency of urban life, deteriorates the urban living environment and reduces the attraction of the city. The traffic information service system based on the floating vehicle is an effective way to solve the congestion problem. The system estimates the traffic state of every road in the current road network through the location and time information provided by the floating vehicle. Based on these information, travelers choose the travel path and travel time to reduce the peak flow in space and time, and reduce unnecessary traffic delays and congestion. This paper aims to solve the key technical problems of traffic information service system based on floating vehicle, including two aspects: the online map matching algorithm under the condition of high density road network and low sampling frequency. And the real-time and efficient path search algorithm based on road network stratification. The parallelism of main and auxiliary roads is a prominent feature of Shenzhen road network, which increases the difficulty of map matching algorithm. In this paper, the online map matching problem is transformed into the shortest path problem in the local road network. In order to establish a map matching algorithm suitable for the road network characteristics of Shenzhen City, this paper proposes a switching point model and a multi-point matching model for road intersections. The former improves the matching accuracy of the algorithm at road intersections. The latter has a good degree of differentiation for the positioning points next to the main and auxiliary roads. In order to verify the accuracy and efficiency of the algorithm, the algorithm is implemented on Visual Studio platform, and a set of floating vehicle positioning data are collected and verified in Shenzhen. The results show that the proposed algorithm not only has high accuracy and efficiency, but also can determine the actual position of the location point next to the main and auxiliary road, which is of great practical value. Real-time and humanization are the distinct requirements of path guidance algorithm. In order to make the output result accord with people's choice preference and improve the efficiency of path search, a hierarchical path search model is established in this paper. The model combines the starting and end roads into the top road network, and then calculates the shortest path of the starting and ending points in the new top road network. In the analysis of the concrete road network in Shenzhen, based on the expressway and the main road, the road network in Shenzhen is divided and stratified. The differences in efficiency and path results between the hierarchical algorithm and the non-delamination algorithm are compared. The results show that the efficiency of the hierarchical algorithm is obviously higher than that of the general algorithm. The result of the hierarchical algorithm results in a high proportion of the top-level roads and less road switching at different levels, which is more in line with people's preference for path selection. In this paper, two kinds of algorithms are proposed according to the actual situation of Shenzhen road network, and verified by the actual data collected or randomly selected. The algorithm has good accuracy and efficiency, and has a high use value.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495
【参考文献】
相关期刊论文 前10条
1 曹敏晖;;城市交通存在的问题及对策分析[J];重庆交通大学学报(社会科学版);2009年01期
2 刘宇环;姚恩建;李欣;李建军;钱越;;浮动车系统地图的匹配精度[J];公路交通科技;2011年S1期
3 余志文,邬永宏;浅谈智能交通中的电子地图[J];海南师范学院学报(自然科学版);2002年02期
4 廖孝勇;孙棣华;周舒杰;田川;古曦;;混合浮动车交通状态估计的样本数量研究[J];计算机工程与应用;2011年24期
5 付梦印,李杰,邓志红;基于分层道路网络的新型路径规划算法[J];计算机辅助设计与图形学学报;2005年04期
6 杨殿阁;罗禹贡;丁捷;李克强;连小珉;;基于单线模型的交通车载电子地图[J];清华大学学报(自然科学版);2006年05期
7 王晓华,郭敏;GPS卫星定位误差分析[J];全球定位系统;2005年01期
8 钟慧玲;邝朝剑;黄晓宇;蔡文学;;基于历史频繁模式的交通流预测算法[J];计算机工程与设计;2012年04期
9 刘春,姚连璧;车载导航电子地图中道路数据的空间逻辑描述[J];同济大学学报(自然科学版);2002年03期
10 李慧兵;杨晓光;;面向行程时间预测准确度评价的数据融合方法[J];同济大学学报(自然科学版);2013年01期
本文编号：2341714
本文链接：https://www.wllwen.com/kejilunwen/jiaotonggongchenglunwen/2341714.html
