短时交通流预测方法研究进展_陆海亭精选.pdf

交通运输工程与信息学报 第 7 卷 第 4 期 2009 年 12 月 Journal of Transportation Engineering and Information No.4 Vo1.7 Dec.2009 短时交通流预测方法研究进展 短时交通流预测方法研究进展 陆海亭 张 宁 黄 卫 夏井新 东南大学，教育部智能运输系统工程研究中心，南京 210096 摘 要：实时准确高效的交通流预测是实现交通流诱导和交通控制的关键技术之一，近年来它在智能运输 领域受到广泛关注。本文首先介绍了短时交通流的属性和预测要求，接着将现有预测方法分成 4 类：基于 线性理论的方法、基于非线性理论的方法、基于混合理论的方法和基于交通流理论的方法，并且总结评述 了现有各种预测模型的优缺点，之后探讨了当今短时交通流预测领域的研究热点，最后指出了其未来研究 方向。 关键词：短时交通流预测；方法分类；研究热点；研究方向 + 中图分类号：U491.1 12 文献标识码： 文章编号：1672-4747（2009）04-0084-08 Research Progress of Short Term Traffic Flow Prediction Methods LU Hai-ting ZHANG Ning HUANG Wei XIA Jing-xin Intelligent Transportation System Engineering Research Centre， the Ministry of Education of China, Southeast University, Nanjing 210096, China Abstract：Real-time, accurate and efficiency short term traffic flow prediction is one of the key technologies to realize traffic flow guidance and traffic control, which has being widely concerned in the domain of ITS during recent years. Firstly, the attributes and requirements of traffic flow were introduced in this paper. Then，the prediction methods were fallen into 4 categories: linear, nonlinear, hybrid, and traffic flow based. The present forecasting models’ advantages and disadvantages were summarized and commented at the same time. Subsequently，the hot topics in the field of traffic flow prediction were also discussed. Finally, the future research direction is pointed out. 收稿日期：2009-03-04. 作者简介：陆海亭（1979-），男，安徽金寨人，汉族，东南大学教育部智能运输系统工程研究中心硕士研究生，研究方向 为智能交通与控制。 84 短时交通流预测方法研究进展 陆海亭 等 Key words：Short term prediction, methods classifica