智能公交信息的采集处理及应用研究

发布时间：2018-04-24 20:16

本文选题：智能公交 + 信息采集　；参考：《重庆交通大学》2014年硕士论文

【摘要】：随着经济和社会的发展，城市车辆拥有量不断增加，交通问题也变的日益严重。如何解决城市交通问题已经成为城市发展的重要课题，在加大城市交通建设和管理的同时，人们也意识到发展城市公共交通是改善城市交通问题的重要手段之一。随着信息技术不断发展，以及在城市交通运输系统中有效地应用，形成以应用交通信息为中心的智能交通运输系统（ITS），智能公交作为ITS中最重要的部分，完善智能公交信息技术，提高采集、处理及应用公交信息能力是公交系统高效运行的保障，是提升公交服务水平和增加企业经营效益的前提。本文从智能公交信息采集的实现技术出发，首先对当前使用的智能公交信息采集技术及特点进行了分析，其次智能公交系统的运作，不仅需要信息采集子系统提供的信息，还必须实现从其它信息化子系统获取相关的公交信息。现有的智能公交系统的各信息化子系统是在不同时间、针对不同业务要求而开发的，导致业务请求在不同平台之间难以实现调用，形成信息孤岛，将大量有用公交信息封闭在一个子系统当中。针对这一问题，研究了基于Web服务的SOA异构信息系统集成，它将信息化子系统的业务功能进行封装，划分为粒度不同的服务，对业务功能的调用转变为对服务的调用，实现各信息化子系统间的信息共享。要实现智能公交系统对公交车辆的智能化管理，还必须获取大量原始信息后面隐藏着的有价值信息，需要对获得的公交信息采取进一步处理。对实时客流量和公交车辆行程时间的预测能满足出行者需求和有利于车辆的实时调度，当前客流量获取基本都是靠人工调查，效率低下，采集的信息单一；公交车辆行程时间采用路程长度与全程平均速度比值获得，与实际值误差大。针对这些问题，研究了公交IC卡和RBF神经网络结合的客流量预测模型，进而对车辆行程时间预测也进行了改善。通过对IC卡信息的获取和分析可以获得比较全面的公交客流量信息，采用RBF神经网络对客流量的预测，得到较高精度的预测结果。再将预测的客流量用于车辆行程时间预测模型，完成预测。最后针对当前智能公交信息服务模式下，服务系统不能提供具有决策性的信息，探索性的建立了智能化公交信息服务系统，将神经网络和专家系统结合使用在信息服务系统中，，在神经系统具有反馈学习和专家系统能推理给出最优解共同处理模式下，为出行者提供决策信息。
[Abstract]:With the development of economy and society, the quantity of urban vehicles is increasing and traffic problems become increasingly serious. How to solve the problem of urban traffic has become an important issue for urban development. While increasing the construction and management of urban traffic, people also realize that the development of urban public transport is an important means to improve urban traffic problems. With the continuous development of information technology and the effective application of the information technology in the urban transportation system, the intelligent transportation system (ITS) which uses traffic information as the center is formed. As the most important part of the ITS, intelligent public transportation is the most important part of the public transport information technology to improve the acquisition, processing and application of public transportation information ability is high. The guarantee of effective operation is the premise to improve the level of public transport service and increase the business efficiency of enterprises.
Starting from the realization technology of intelligent bus information collection, this paper first analyzes the current information acquisition technology and characteristics of the intelligent public transport. Secondly, the operation of the intelligent bus system needs not only information provided by the information collection subsystem, but also the relevant information of public transportation from other information subsystems. The information subsystem of the public transportation system is developed at different time and for different business requirements. It causes the business requests to be difficult to call between different platforms, form information island and close a large number of useful information in a subsystem. In this question, the SOA heterogeneous information system based on Web service is studied. Integration, it encapsulates the business functions of the information subsystem, divides into different services, calls the call of business functions to service, and realizes information sharing among the information subsystems.
In order to realize intelligent management of public transport system to bus vehicles, it is necessary to obtain valuable information hidden behind a large number of original information, and need to take further processing of the obtained bus information. The prediction of real-time passenger traffic and bus travel time can meet the needs of the travelers and facilitate the real-time scheduling of vehicles. The passenger flow acquisition is basically based on artificial investigation, low efficiency and single collection of information; bus travel time is obtained by the ratio of road length to the average speed, and the error is large with the actual value. In view of these problems, the passenger traffic prediction model combined with bus IC card and RBF neural network is studied, and then the vehicle travel time is predicted. Through the acquisition and analysis of IC card information, more comprehensive traffic information can be obtained, the RBF neural network is used to predict the passenger traffic, and the prediction results of high precision are obtained. Then the predicted passenger flow is used in the vehicle travel time prediction model, and the prediction is finished. Finally, the information service of the current intelligent bus is used. In the mode, the service system can not provide information with decision making, and establish an intelligent bus information service system, which combines the neural network and expert system in the information service system, and provides the decision for the travelers in the neural system with feedback learning and the expert system can reasoning and giving the best solution together. Information.

【学位授予单位】：重庆交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;U491.17

【参考文献】