基于大数据量的特种车辆搜路算法优化与实现
发布时间：2018-08-21 12:04
【摘要】：车辆导航软件是目前车辆出行必备的工具。目前市场上的导航软件满足了普通用户日常出行的需要。由于车体超重、超高、超宽、超长的特种车辆对道路通行能力有特殊要求,市场的导航软件则没有根据特种车辆需求条件进行定制的功能。特种车辆出行过程中经常要进行较长路程的行车,因此需要一种面向大数据量的路网中快速完成搜路的算法。为了响应国家提倡国产化的号召,具有跨平台能力的软件具有很好的发展前景。本文首先根据客户的需求和国家政策引出了课题的背景和研究意义,通过查阅相关文献总结了国内外的Dijkstra算法的优化研究的进展和车辆导航软件在国内外的研究进展,总结了目前市场上的导航软件不能满足特种车辆特殊的需求的问题。论文其次介绍了车辆搜路分析过程中所需基础数据的结构、存储方式、为了提升搜路效率对数据的加载所用的数据组织方式和结合特种车辆自身对道路需求对路网数据中的“关键点”数据做的预处理。论文接着阐述了车辆搜路过程所需要的常规算法-Dijkstra算法,并在大数据路网环境下的Dijkstra算法的效率做了优化和提升。在Dijkstra算法的基础上结合特种车辆对道路的特殊需求设计了根据特种车辆用户对道路选择的条件实现搜路分析的算法,并进行了实现。论文通过对跨平台仿组件框架的研究以及对Qt内部图形框架的学习与研究,确定了特种车辆导航组件的设计思路,采用面向对象的方法,对大数据量的路网数据预处理,入库以及道路“关键点”数据的提取和管理,常规搜路算法以及特种车辆搜路算法进行了详细的设计与实现。最后论文通过具体的实例,成功地验证了特种车辆组件各个功能,实现了特种车辆组件能在大数据量的路网条件较短时间内完成搜路分析的特性。
[Abstract]:Vehicle navigation software is a necessary tool for vehicle travel. At present, the navigation software in the market meets the daily travel needs of ordinary users. Because the special vehicles with overweight, super-high, ultra-wide and super-long have special requirements for road capacity, the navigation software of the market does not have the function of customizing according to the requirements of special vehicles. Special vehicles often travel a long distance, so we need a fast search algorithm for large amount of data in the road network. In order to respond to the call of nationalization, cross-platform software has a good prospect. Firstly, according to the customer's demand and the national policy, this paper leads to the background and significance of the research, and summarizes the research progress of the Dijkstra algorithm optimization and the vehicle navigation software at home and abroad by consulting the relevant literature. The problems that the navigation software in the market can not meet the special needs of special vehicles are summarized. Secondly, the paper introduces the structure and storage mode of the basic data needed in the process of vehicle search analysis. In order to improve the efficiency of road search, the data organization mode used to load the data and the preprocessing of the "key point" data in the road network data based on the special vehicle itself are proposed. Then, the paper describes the conventional algorithm-Dijkstra algorithm, which is needed in the process of vehicle search, and optimizes and improves the efficiency of the Dijkstra algorithm in the big data network environment. Based on the Dijkstra algorithm and the special demand of the special vehicle to the road, the algorithm of road search analysis is designed according to the condition of the special vehicle user's choice of the road, and the algorithm is implemented. Through the research of cross-platform simulation component framework and the study and research of QT internal graphics framework, this paper determines the design idea of special vehicle navigation component, and uses the object-oriented method to preprocess the road network data of large amount of data. The data extraction and management of the "key points" in the database, the conventional search algorithm and the special vehicle search algorithm are designed and implemented in detail. Finally, through concrete examples, the paper successfully verifies each function of the special vehicle component, and realizes the characteristic that the special vehicle component can complete the road search analysis in the short time of the road network condition of large amount of data.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495
本文编号：2195663
[Abstract]:Vehicle navigation software is a necessary tool for vehicle travel. At present, the navigation software in the market meets the daily travel needs of ordinary users. Because the special vehicles with overweight, super-high, ultra-wide and super-long have special requirements for road capacity, the navigation software of the market does not have the function of customizing according to the requirements of special vehicles. Special vehicles often travel a long distance, so we need a fast search algorithm for large amount of data in the road network. In order to respond to the call of nationalization, cross-platform software has a good prospect. Firstly, according to the customer's demand and the national policy, this paper leads to the background and significance of the research, and summarizes the research progress of the Dijkstra algorithm optimization and the vehicle navigation software at home and abroad by consulting the relevant literature. The problems that the navigation software in the market can not meet the special needs of special vehicles are summarized. Secondly, the paper introduces the structure and storage mode of the basic data needed in the process of vehicle search analysis. In order to improve the efficiency of road search, the data organization mode used to load the data and the preprocessing of the "key point" data in the road network data based on the special vehicle itself are proposed. Then, the paper describes the conventional algorithm-Dijkstra algorithm, which is needed in the process of vehicle search, and optimizes and improves the efficiency of the Dijkstra algorithm in the big data network environment. Based on the Dijkstra algorithm and the special demand of the special vehicle to the road, the algorithm of road search analysis is designed according to the condition of the special vehicle user's choice of the road, and the algorithm is implemented. Through the research of cross-platform simulation component framework and the study and research of QT internal graphics framework, this paper determines the design idea of special vehicle navigation component, and uses the object-oriented method to preprocess the road network data of large amount of data. The data extraction and management of the "key points" in the database, the conventional search algorithm and the special vehicle search algorithm are designed and implemented in detail. Finally, through concrete examples, the paper successfully verifies each function of the special vehicle component, and realizes the characteristic that the special vehicle component can complete the road search analysis in the short time of the road network condition of large amount of data.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495
【参考文献】
相关期刊论文 前10条
1 李擎,宋顶立,张双江,李哲,刘建光,王志良;两种改进的最优路径规划算法[J];北京科技大学学报;2005年03期
2 王杰臣,毛海城,杨得志;图的节点-弧段联合结构表示法及其在GIS最优路径选取中的应用[J];测绘学报;2000年01期
3 刘玉海,肖江阳,张锡恩;一种新型最短路径搜索算法的研究[J];计算机工程与应用;2001年17期
4 姚亚锋;方贤进;陈代梅;;Dijkstra算法的一种高效率实现[J];计算机与数字工程;2007年07期
5 胡金星;刘允才;;面向动态导航的城市路网实时交通信息服务系统研究[J];交通与计算机;2005年06期
6 张国强,晏克非;城市道路网络交通特性仿真模型及最短路径算法[J];交通运输工程学报;2002年03期
7 朱静;Dijkstra算法在GIS中的优化实现[J];计算机与现代化;2005年09期
8 俞奕;;GIS中最短路径问题的应用研究[J];软件导刊;2007年13期
9 乐阳,龚健雅;Dijkstra最短路径算法的一种高效率实现[J];武汉测绘科技大学学报;1999年03期
10 卫小伟;;城市智能交通控制系统研究与设计[J];现代电子技术;2010年17期
相关硕士学位论文 前1条
1 蔚洁;车辆监控导航系统中最短路径的实时性研究[D];河北师范大学;2007年
，本文编号：2195663
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2195663.html
