车联网环境下行车主动服务需求感知模型及关键算法研究

发布时间：2018-05-15 15:44

本文选题：车联网 + 主动数据库　；参考：《江苏大学》2017年硕士论文

【摘要】：随着我国汽车保有量的持续增加,有限的道路资源和持续增长的车辆构成了难以调和的矛盾,交通事故、交通拥堵频发。行车服务系统逐渐出现为车辆提供服务,如客车上使用的G-BOS(客车智慧运营系统)和货车上使用的天行健车联网服务系统,能够在一定程度上促进驾驶员的安全驾驶及车辆的维修保养。但是,这些行车服务系统都有着特殊的应用对象,如G-BOS可以为客车提供行车服务,天行健能够为货车提供服务,对于其他类型的车辆则不适用。行车主动服务系统(Driving Active Service System,DASS)就是要通过采集车辆在车联网环境(Internet of Vehicles,IV)下的行驶信息,通过信息融合对行车服务进行有效的自动感知与辨识,为车联网环境下的车辆行驶提供实时、主动、高效的服务。论文在传统的行车服务系统(Traditional Traffic Service Systems,TTSS)的基础上,以江苏省六大人才高峰项目“行车主动服务系统关键技术研究(项目编号:DZXX-048)”为依托,对DASS系统需求感知模型及关键算法进行了研究。DASS需求感知模型是整个DASS系统最基础、但同时又是最重要的部分,是DASS系统基础理论的研究及关键技术探索。论文的主要研究工作包含以下4部分内容:(1)系统阐述了车联网与DASS的研究及发展现状,重点分析了车联网环境下DASS的可行性、详细分析了车联网环境下的DASS行车服务需求,并对车联网环境下的车辆行驶信息进行了系统性分类、对DASS行车服务进行了细致的划分。(2)提出了基于主动数据库的DASS需求感知模型,研究了以ECA(事件-条件-动作)模型为基础的DASS需求感知模型,构建了识别交通拥堵的ECA规则,并建立起一套数据库管理系统的界面。(3)针对上述DASS需求感知模型服务实现的过程,分别提出了基于Multiagent的信息融合算法、不需借助于检测设备的交通拥堵感知算法,以及DASS服务类型、服务能力匹配算法等三类关键算法,并对算法的复杂度进行了分析和实验验证。(4)自主搭建了基于Prescan-Matlab/simulink软件和罗技G29的驾驶员硬件在环DASS仿真平台,以DASS感知模型提供追尾预警安全类行车服务和实时动态的路径规划服务为例,验证了上述模型与关键算法的有效性和优越性。
[Abstract]:With the continuous increase of vehicle ownership in China, the limited road resources and the continuous growth of vehicles constitute an irreconcilable contradiction, traffic accidents and traffic congestion occur frequently. Train service systems have emerged to provide services for vehicles, such as the G-BOS (bus Intelligence Operation system) used on passenger cars and the connected service system for skyline vehicles used on trucks. Can promote driver's safe driving and vehicle maintenance to a certain extent. However, these vehicle service systems have special application objects, such as G-BOS can provide driving services for passenger cars, days can provide services for trucks, but not for other types of vehicles. The driving Active Service system (DASS) is to collect the driving information of the vehicle under the environment of the vehicle network, and to realize the effective automatic perception and identification of the driving service through the information fusion. It provides real-time, active and efficient service for vehicle running in networked environment. On the basis of the traditional Traffic Service system TTSS, this paper is based on the research on the key technology of driving active service system (project No.: DZXX-048), which is the six talents peak project in Jiangsu Province. This paper studies the DASS system requirement awareness model and its key algorithms. The Dass requirement perception model is the most basic but also the most important part of the whole DASS system. It is the basic theory research and key technology exploration of the DASS system. The main research work of this paper includes the following four parts: 1) the research and development status of DASS and vehicle networking are described, the feasibility of DASS under the environment of vehicle networking is analyzed, and the requirement of DASS driving service in the networked environment is analyzed in detail. The vehicle driving information in the networked environment is systematically classified, and the DASS driving service is divided into two parts. (2) the DASS requirement perception model based on active database is proposed. Based on the ECA-based event-conditional action model, the DASS requirement awareness model is studied, and the ECA rules to identify traffic congestion are constructed. An interface of a database management system is established. (3) aiming at the process of implementing the above DASS requirement perception model service, the information fusion algorithm based on Multiagent is proposed, and the traffic congestion awareness algorithm without the aid of detection equipment is proposed. And three kinds of key algorithms, such as DASS service type, service ability matching algorithm, etc., and the complexity of the algorithm is analyzed and verified by experiments. (4) the driver hardware in loop DASS simulation platform based on Prescan-Matlab/simulink software and Logitech G29 is built independently. The effectiveness and superiority of the above model and the key algorithms are verified by taking the DASS perception model as an example to provide the rear-end warning safety traffic service and the real-time dynamic path planning service.
【学位授予单位】：江苏大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495

【参考文献】