基于GPS的公交车辆到站预测模型设计与应用

发布时间：2018-10-16 15:11

【摘要】：随着城市现代化的快速发展,城市人口不断增加,市民的生活范围不断扩大,城市公共交通问题逐渐成为国内外关注的重大问题。近些年科技的不断发展,使得智能公交系统正以其智能化服务公共交通的特性越来越受到政府和社会的广泛重视。智能公交系统必将成为衡量未来城市发展状况不可或缺的一项重要指标。而公交车辆到站时间预测作为智能公交系统中的重要组成部分,预测的精确程度也直接影响着智能公交系统的整体水平。本文主要研究了以时间段、天气、节假日为参考因素的支持向量机与实时预测相结合的混合公交到站时间预测模型的分析设计与实现。首先对于本文依托的玉溪智能公交系统的特点、影响到站时间因素进行分析,同时对几种常用的时间预测模型进行选型,选择了具有自适应性,鲁棒性以及小样本预测的支持向量机模型(SVM)作为基础预测模型,并确立了时间段、天气、节假日为参考因素。又由于道路交通的复杂性大,单一的历史数据预测,无法应对突发状况。而本项目中利用GPS技术获得的实时定位信息,RFID技术与GPS技术交互获得实时到站信息,为预测模型的提供了实时可靠的预测数据,有效的应对了突发状况。技术方案确定后,通过前期的分析、设计,完成了模型的实现。并采用现场GPS采集的真实数据进行测试,达到了良好的预测结果。最后,在到站时间预测功能设计实现的基础上,运用了手机、网络、电子站牌多种方式进行了功能的发布,使本文的研究具有了更大的实用价值。
[Abstract]:With the rapid development of urban modernization, the urban population is increasing and the living range of the citizens is expanding. The urban public transport problem has gradually become a major issue of concern at home and abroad. With the development of science and technology in recent years, the intelligent public transport system has been paid more and more attention by the government and the society because of its intelligent service to public transportation. Intelligent public transport system will become an indispensable index to measure the future urban development. As an important part of intelligent bus system, bus arrival time prediction has a direct impact on the overall level of intelligent bus system. This paper mainly studies the analysis, design and implementation of the mixed bus arrival time prediction model which combines the support vector machine with real-time prediction, which takes time, weather and holiday as reference factors. First of all, the characteristics of Yuxi Intelligent bus system, which is based on this paper, are analyzed, and the factors affecting the arrival time are analyzed. At the same time, several commonly used time prediction models are selected to be adaptive. The robustness and the support vector machine (SVM) model of small sample prediction are used as the basic prediction model, and the time period, weather and holidays are established as reference factors. Because of the complexity of road traffic, a single historical data forecast, can not cope with emergencies. In this project, the real-time location information obtained by GPS technology, the real-time arrival information obtained by RFID technology and GPS technology, provide real-time and reliable prediction data for the prediction model, and effectively deal with the unexpected situation. After the technical scheme is determined, the model is realized through the analysis and design. The real data collected by field GPS are used to test, and good prediction results are achieved. Finally, on the basis of the design and realization of the arrival time prediction function, the functions of mobile phone, network and electronic station are released, which makes the research of this paper have more practical value.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：U495;P228.4

【相似文献】