基于Storm云平台的地图道路匹配算法研究

发布时间：2018-07-03 01:44

本文选题：地图匹配 + 城市路网拓扑构建　；参考：《云南大学》2013年硕士论文

【摘要】：随着我国城市道路交通系统的迅猛发展,地理信息系统和卫星定位技术在解决城市智能交通方面得到了广泛的应用。地图匹配是GPS车辆定位导航系统研究的关键技术,也是影响智能交通系统(ITS)发展的主要环节。本文主要研究基于ArcGIS的地图底层数据处理和地图匹配算法,并对地图匹配算法迁移到Storm平台的可行性进行了验证。其研究成果对于实现GPS和数字地图的信息融合以及探索Storm平台在交通领域的应用都具有重要的意义。本文首先对深圳市ArcGIS电子地图进行相关预处理,在预处理的基础上提出了一种全新的道路底层数据提取方法,并根据这些道路底层数据的特点提出了一种新的城市路网拓扑结构构建算法,该算法可以快速构建出深圳市城市路网拓扑。接着,在对深圳市出租车7天行车轨迹GPS数据进行预处理的前提下,借鉴分治法的思想,提出一种基于九宫格网格的地图道路匹配算法,该算法能克服常用地图匹配算法运算效率不高的问题,具有较好的匹配精度和效率。之后结合实验,对九宫格网格的地图匹配算法进行验证。实验结果表明,该地图匹配算法在复杂路段也能有较高的匹配精度,算法有良好的适应性,能够满足车辆地图匹配的要求。最后,本文将所提的地图匹配算法迁移到Storm云平台上。由于Storm云平台的强大并行计算能力,可以使地图匹配算法的运行效率提高2-3倍。
[Abstract]:With the rapid development of urban road traffic system in China, geographic information system (GIS) and satellite positioning technology have been widely used to solve the problem of urban intelligent transportation. Map matching is the key technology of GPS vehicle positioning and navigation system, and also the main link that affects the development of Intelligent Transportation system (its). This paper mainly studies the map bottom data processing and map matching algorithm based on ArcGIS, and verifies the feasibility of migration of map matching algorithm to Storm platform. The research results are of great significance to realize the information fusion of GPS and digital map and to explore the application of Storm platform in the field of transportation. This paper first carries on the correlation preprocessing to the ArcGIS electronic map of Shenzhen City, and puts forward a new method of extracting the road bottom data based on the preprocessing. According to the characteristics of these data, a new algorithm for constructing urban road network topology is proposed, which can quickly construct the topology of Shenzhen city road network. Then, on the premise of preprocessing the GPS data of 7-day driving track of Shenzhen taxi, a map road matching algorithm based on nine-grid is proposed based on the idea of divide-and-conquer method. This algorithm can overcome the problem that the common map matching algorithm is not efficient, and has better matching accuracy and efficiency. Then the map matching algorithm of nine grid grids is verified by experiments. The experimental results show that the map matching algorithm can also have high matching accuracy in complex sections, and the algorithm has good adaptability and can meet the requirements of vehicle map matching. Finally, the proposed map matching algorithm is migrated to Storm cloud platform. Because of the powerful parallel computing ability of Storm cloud platform, the efficiency of map matching algorithm can be improved 2-3 times.
【学位授予单位】：云南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：U495;P208

【参考文献】