系统仿真学报 ›› 2018, Vol. 30 ›› Issue (11): 4100-4106.doi: 10.16182/j.issn1004731x.joss.201811007

• 仿真建模理论与方法 • 上一篇    下一篇

基于深度学习的交通流量预测

刘明宇1, 吴建平1, 王钰博2, 何磊3   

  1. 1.清华大学土木系,北京 100084;
    2.浙江大学竺可桢学院,浙江 杭州 310000;
    3.北京市公安局公安交通管理局,北京 100037
  • 收稿日期:2018-05-23 修回日期:2018-06-29 发布日期:2019-01-04
  • 作者简介:刘明宇(1993-),男,吉林,博士,研究方向为智能交通、机器学习。
  • 基金资助:
    国家自然科学基金(U1509205,U1709212),清华大学自主科研计划项目(2015THZ01), 北京市朝阳区科技计划项目(CYSF1701)

Traffic Flow Prediction Based on Deep Learning

Liu Mingyu1, Wu Jianping1, Wang Yubo2, He Lei3   

  1. 1. Department of Civil Engineering, Tsinghua University, Beijing 100084, China;
    2. Chu Kochen Honors College, Hangzhou 310000, China;
    3. Beijing Traffic Management Bureau, Beijing 100037, China
  • Received:2018-05-23 Revised:2018-06-29 Published:2019-01-04

摘要: 交通流预测是城市智能交通系统的重要组成部分。随着人工智能和机器学习的不断发展,深度学习在交通工程领域得到了广泛的应用。选取门控循环单元(Gated Recurrent Unit, GRU)神经网络作为研究对象,利用交叉验证法探究GRU模型的最佳门控循环单元个数,并与支持向量机回归等三种预测模型通过不同指标进行综合评价和对比。结果表明,与其它3种模型相比,GRU模型具有良好的预测性能。

关键词: 交通工程, GRU, 交通流预测, 深度学习

Abstract: Traffic flow prediction is an important component of urban intelligent transportation system. With the development of machine learning and artificial intelligence, deep learning has been applied in traffic engineering area. Gated recurrent unit (GRU) neural network is selected to predict urban traffic flow. Cross-validation method is used to explore the optimal number of gated recurrent units. The GRU model is compared with other three predictors such as support vector regression and evaluated in different performance measurements. The results show that GRU model has better performance in traffic flow prediction than the other three models.

Key words: traffic engineering, GRU, traffic flow prediction, deep learning

中图分类号: