基于PHP技术的交通诱导服务平台设计与实现

发布时间：2018-11-24 18:54

【摘要】：随着城市规模的极速扩大,城市中的交通资源已经不能赶上城市的扩张速度。由于交通资源的紧张,导致了一系列社会问题：交通拥堵,交通事故频发,出行时间冗长等。随着近两年物联网、车联网、车路协同和互联网大数据的蓬勃发展,融合各种领域领先技术的智能交通诱导系统应运而生。但是目前的交通诱导服务平台存在着以下的问题：不能承载海量的用户数,开发和实验周期非常长,大数据的处理能力不能满足要求。为此,本文在调研了各个领域的研究前沿后,针对上述问题提出了一套针对性的解决方案。PHP是当今开发效率最高的语言,因此通过PHP开发诱导服务平台有得天独厚的优势。为了保证平台能够支撑海量的服务请求,本文提出了采用分布式的系统架构,使得系统能够平滑地实现横向扩容,动态适配服务资源。由于PHP自身语言性能存在瓶颈,因此采用核心算法C扩展化和HHVM化来保证系统的性能不受桎梏。同时,本文提出了新型的异构感知节点部署算法,能够很大程度上提高网络节点的部署效率。通过无缝融合RSSI、 GPS、 INS惯导系统,提出了高速道路下的多模融合差分融合定位算法。通过浮动车采集路网的OD矩阵参数,对传统的Dijkstra算法进行优化,提出了一种高效的动态诱导算法。最后,针对上述理论分析和研究,在开放的私有云主机上实现算法并部署了交通诱导服务平台的演示系统,并对一些关键功能进行了测试和分析,符合预期。
[Abstract]:With the rapid expansion of the city, the traffic resources in the city can not catch up with the expansion of the city. The shortage of traffic resources leads to a series of social problems: traffic congestion, frequent traffic accidents, long travel time and so on. With the rapid development of Internet of things, vehicle-road coordination and Internet big data, intelligent traffic guidance system (its), which integrates leading technologies in various fields, has emerged as the times require. However, the current traffic guidance service platform has the following problems: can not carry a large number of users, development and experimental period is very long, big data's processing capacity can not meet the requirements. Therefore, after investigating the research frontier in various fields, this paper puts forward a set of targeted solutions to the above problems. PHP is the language with the highest development efficiency nowadays, so it is advantageous to develop induction service platform through PHP. In order to ensure that the platform can support a large number of service requests, this paper proposes a distributed system architecture, which enables the system to smoothly achieve horizontal expansion, dynamic adaptation of service resources. Because of the bottleneck of PHP's own language performance, the core algorithms C extension and HHVM are adopted to ensure that the system performance is not shackled. At the same time, this paper proposes a new heterogeneous aware node deployment algorithm, which can greatly improve the deployment efficiency of network nodes. Through seamless fusion of RSSI, GPS, INS inertial navigation system, a multi-mode fusion differential fusion localization algorithm for high-speed road is proposed. By collecting the OD matrix parameters of the road network by floating vehicle, the traditional Dijkstra algorithm is optimized, and an efficient dynamic induction algorithm is proposed. Finally, in view of the above theoretical analysis and research, the algorithm is implemented on the open private cloud host and the demonstration system of the traffic guidance service platform is deployed, and some key functions are tested and analyzed to meet the expectations.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495

【参考文献】