基于Storm平台的路段行程时间估算与拥堵评价应用研究

发布时间：2018-03-10 17:23

本文选题：云计算　切入点：GPS浮动车　出处：《武汉理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：应用云计算技术解决交通领域的问题是当前技术研究的一个热门趋势。在交通领域中,由遍布城市道路各个角落的检测器和浮动车采集的数据是海量的,且具有覆盖范围广、信息量丰富等特点,非常适合于应用大数据技术进行挖掘和实时分析工作。在城市交通大数据技术中,利用浮动车GPS数据,进行路段行程时间的实时计算和道路拥堵状态的评价,可以为基于实时路况的路径规划、行程时间预测、道路拥堵预测等研究提供数据基础,可将有助于交通监管平台和居民及时了解当前道路交通参数状态,提前规划出行时间和出行路径。本文主要研究了应用云计算技术处理海量浮动车GPS数据,计算城市路段行程时间和交通状态评价的方法和实现流程。本文通过对GPS浮动车地图匹配技术的研究分析,提出了适用于海量数据计算的地图匹配改进方法和具体实现流程;通过对行程时间估算方法的研究,提出了一种改进的少样本估算方法,同时探讨了在Storm云平台的具体实现技术;通过对交通拥堵评价的研究分析,给出了一种基于路段行程时间的简易评价指标,同时研究了在云平台下的实现方案。本文所完成的研究工作主要如下:(1)系统研究了GPS浮动车技术中的基础技术—地图匹配技术的主要原理和方法,分析了当前主要地图匹配方法的优点和不足,提出适用于实际大数据应用的基于权重的改进匹配方法。为解决大量数据实时处理中的匹配效率问题,提出GIS路段划分方法、四网格匹配法、权重计算方法优化和利用路网拓扑结构的待匹配路段筛选方法,同时给出了具体细节技术的实现流程。(2)研究分析了城市交通路段行程时间估算的主要方法,在考虑实际应用中的情况并总结各方法优劣后,提出综合时间插值法和速度-时间积分法的改进方法:基于运行时刻和运行速度的少样本估算方法。同时在路段行程时间的估算的基础上提出了一种新的简易拥堵评价模型。(3)研究设计了路段行程时间实时估算以及道路拥堵评价方法在云平台中具体算法流程和实现方案,同时应用Storm云平台进行了实际的程序开发和应用。
[Abstract]:The application of cloud computing technology to solve traffic problems is a hot trend in the field of transportation. In the field of transportation, the data collected by detectors and floating vehicles in every corner of urban roads are massive and have a wide range of coverage. Because of its rich information, it is very suitable for mining and real time analysis by using big data technology. In the urban traffic big data technology, the floating vehicle GPS data is used. The real-time calculation of road travel time and the evaluation of road congestion can provide the data basis for the research of path planning, travel time prediction and road congestion prediction based on real-time road conditions. It will be helpful for the traffic supervision platform and residents to understand the current state of road traffic parameters and plan the travel time and route ahead of time. This paper mainly studies the application of cloud computing technology to deal with massive floating vehicle GPS data. Based on the research and analysis of GPS floating vehicle map matching technology, the paper puts forward a map matching improvement method and concrete realization flow for mass data calculation. Based on the study of travel time estimation method, an improved estimation method with less samples is proposed, and the specific implementation technology in Storm cloud platform is discussed. A simple evaluation index based on route travel time is given. At the same time, the realization scheme under the cloud platform is studied. The main research work in this paper is as follows: 1) the main principle and method of map matching technology, which is the basic technology of GPS floating vehicle technology, is studied systematically. This paper analyzes the advantages and disadvantages of the current main map matching methods, and puts forward an improved matching method based on weight, which is suitable for the practical application of big data. In order to solve the problem of matching efficiency in the real-time processing of a large amount of data, the GIS section partition method is proposed. Four grid matching method, weight calculation method optimization and road network topology structure to be used to match the section selection method, at the same time, the specific details of the implementation flow. 2) the main methods of estimating the travel time of urban traffic section are studied and analyzed. After considering the practical application and summarizing the advantages and disadvantages of each method, An improved method of integrated time interpolation and velocity-time integration is proposed, which is based on running time and running speed. Based on the estimation of travel time, a new simple method is proposed. The real time estimation of road travel time and the specific algorithm flow and implementation scheme of road congestion evaluation method in cloud platform are studied and designed. At the same time, the actual program development and application are carried out by using Storm cloud platform.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495

【参考文献】