基于时空多图卷积网络的交通站点流量预测

doi:10.19678/j.issn.1000-3428.0056316

计算机工程 ›› 2020, Vol. 46 ›› Issue (5): 26-33. doi: 10.19678/j.issn.1000-3428.0056316

基于时空多图卷积网络的交通站点流量预测

荣斌^1a,1b, 武志昊^1a,1b, 刘晓辉², 赵苡积^1a,1b, 林友芳^1a,1b, 景一真^1a,1b

1. 北京交通大学 a. 计算机与信息技术学院;b. 交通数据分析与挖掘北京市重点实验室, 北京 100044;
2. 中国民用航空局民航旅客服务智能化应用技术重点实验室, 北京 100105

收稿日期:2019-10-17 修回日期:2019-11-26 发布日期:2019-12-04
作者简介:荣斌(1995-),男,硕士研究生,主研方向为数据与知识工程、时空数据挖掘;武志昊,副教授、博士生导师;刘晓辉(通信作者),硕士;赵苡积,博士研究生;林友芳,教授、博士生导师;景一真,硕士研究生。
基金资助:
中央高校基本科研业务费专项资金（2019JBM023）。

Flow Prediction of Traffic Stations Based on Spatio-Temporal Multi-Graph Convolutional Network

RONG Bin^1a,1b, WU Zhihao^1a,1b, LIU Xiaohui², ZHAO Yiji^1a,1b, LIN Youfang^1a,1b, JING Yizhen^1a,1b

1a. School of Computer and Information Technology;1b. Beijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing 100044, China;
2. Key Laboratory of Intelligent Passenger Service of Civil Aviation, Civil Aviation Administration of China, Beijing 100105, China

Received:2019-10-17 Revised:2019-11-26 Published:2019-12-04

摘要/Abstract

摘要： 交通流量预测是智能交通系统中的重要组成部分，但由于交通流量受交通状况、地理位置、时间等多种因素影响，使其具有高度非线性与复杂性，实现精准预测的难度较大。针对交通站点的出入流量预测问题，提出一种基于上下文门控的时空多图卷积网络（CG-STMGCN）模型。根据站点间的相邻关系与流通流量关系构造邻居图与流通流量图表示站点流量之间的邻近相关性与流量依赖性，在两图上分别建立基于上下文门控的时空卷积模块捕获站点流量的时空特征，并使用哈达玛乘积融合两图的输出作为最终预测结果。在真实交通站点数据集上的实验结果表明，CG-STMGCN模型的预测准确性优于同类预测方法，且稳定性更强。

关键词: 智能交通, 流量预测, 交通站点, 时空多图卷积, 上下文门控单元

Abstract: Traffic flow prediction is an important part of intelligent transportation systems.However,due to the influence of traffic conditions,geographical location,time and other factors,traffic flow prediction is highly non-linear and complex,which imposes a great challenge to accurate prediction.This paper proposes a novel Contextual Gated Spatio-Temporal Multi-Graph Convolutional Network(CG-STMGCN) model to predict the inflow and outflow of traffic stations.In this model,the neighborhood graph and the flow-wise graph are constructed based on the adjacent relationships and flow-wise relationships between stations to represent the proximity correlations and flow dependencies between station flows.Then a contextual gated spatio-temporal convolutional module is constructed on two graphs to capture the spatio-temporal features of station flows.Finally,Hadamard product is used to fuse the outputs of the two graphs as the final prediction result.Experimental results on the dataset of real traffic stations show that the proposed CG-STMGCN model outperforms other existing prediction models in terms of prediction performance,and has better stability.

Key words: intelligent transportation, flow prediction, traffic station, spatio-temporal multi-graph convolution, contextual gated unit

中图分类号:

TP311

荣斌, 武志昊, 刘晓辉, 赵苡积, 林友芳, 景一真. 基于时空多图卷积网络的交通站点流量预测[J]. 计算机工程, 2020, 46(5): 26-33.

RONG Bin, WU Zhihao, LIU Xiaohui, ZHAO Yiji, LIN Youfang, JING Yizhen. Flow Prediction of Traffic Stations Based on Spatio-Temporal Multi-Graph Convolutional Network[J]. Computer Engineering, 2020, 46(5): 26-33.

http://www.ecice06.com/CN/Y2020/V46/I5/26

图/表 7

20200513200504

20200513200509

20200513200516

20200513200520

20200513200527

20200513200532

20200513200536

参考文献

[1] WILLIAMS B M,HOEL L A.Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process:theoretical basis and empirical results[J].Journal of Transportation Engineering,2003,129(6):664-672.
[2] WEI Yu,CHEN Muchen.Forecasting the short-term metro passenger flow with empirical mode decomposition and neural networks[J].Transportation Research Part C:Emerging Technologies,2012,21(1):148-162.
[3] SUN Yuxing,LENG Biao,GUAN Wei.A novel wavelet-SVM short-time passenger flow prediction in Beijing subway system[J].Neurocomputing,2015,166:109-121.
[4] LIN Peiqun,ZHOU Nannan.Short-term traffic flow forecast of toll station based on multi-feature GBDT model[J].Journal of Guangxi University(Natural Science Edition),2018,43(3):1192-1199.(in Chinese)林培群,周楠楠.基于多特征GBDT模型的收费站短时交通流量预测[J].广西大学学报(自然科学版),2018,43(3):1192-1199.
[5] LECUN Y,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521(7553):436-444.
[6] WANG Xiangxue,XU Lunhui.Short-term traffic flow prediction based on deep learning[J].Journal of Transportation Systems Engineering and Information Technology,2017,18(1):81-88.(in Chinese)王祥雪,许伦辉.基于深度学习的短时交通流预测研究[J].交通运输系统工程与信息,2017,18(1):81-88.
[7] LI Yuelong,TANG Dehua,JIANG Guiyuan,et al.Short term traffic flow forecasting based on dimension weighted residual LSTM[J].Computer Engineering,2019,45(6):1-5.(in Chinese)李月龙,唐德华,姜桂圆,等.基于维度加权的残差LSTM短期交通流量预测[J].计算机工程,2019,45(6):1-5.
[8] QIAO Songlin,SUN Rencheng,LIU Ji.Short-term traffic flow forecast based on deep learning[J].Journal of Qingdao University(Natural Science Edition),2017,30(4):65-69.(in Chinese)乔松林,孙仁诚,刘吉.基于深度学习的短时交通流量预测[J].青岛大学学报(自然科学版),2017,30(4):65-69.
[9] ZHANG Junbo,ZHENG Yu,QI Dekang.Deep spatio-temporal residual networks for citywide crowd flows prediction[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence.Palo Alto,USA:AAAI Press,2017:1-7.
[10] JIN Wenwei,LIN Youfang,WU Zhihao,et al.Spatio-temporal recurrent convolutional networks for citywide short-term crowd flows prediction[C]//Proceedings of the 2nd International Conference on Compute and Data Analysis.New York,USA:ACM Press,2018:28-35.
[11] YU Bing,YIN Haoteng,ZHU Zhanxing.Spatio-temporal graph convolutional networks:a deep learning framework for traffic forecasting[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Washington D.C.,USA:IEEE Press,2018:3634-3640.
[12] DIAO Zulong,WANG Xin,ZHANG Dafang,et al.Dynamic spatial-temporal graph convolutional neural networks for traffic forecasting[C]//Proceedings of the 33rd AAAI Conference on Artificial Intelligence.Palo Alto,USA:AAAI Press,2019:1-11.
[13] GUO Shengnan,LIN Youfang,LI Shijie,et al.Attention based spatial-temporal graph convolutional networks for traffic flow forecasting[C]//Proceedings of AAAI Conference on Artificial Intelligence.Palo Alto,USA:AAAI Press,2019:922-929.
[14] CHAI Di,WANG Leye,YANG Qiang.Bike flow prediction with multi-graph convolutional networks[C]//Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems.New York,USA:ACM Press,2018:397-400.
[15] GENG Xu,LI Yaguang,WANG Leye,et al.Spatiotemporal multi-graph convolution network for ride-hailing demand forecasting[C]//Proceedings of 2019 AAAI Conference on Artificial Intelligence.Palo Alto,USA:AAAI Press,2019:3656-3663.
[16] NIEPERT M,AHMED M,KUTZKOV K.Learning convolutional neural networks for graphs[C]//Proceedings of International Conference on Machine Learning.Palo Alto,USA:AAAI Press,2016:2014-2023.
[17] BRUNA J,ZAREMBA W,SZLAM A,et al.Spectral networks and locally connected networks on graphs[EB/OL].[2019-09-11].https://www.researchgate.net/publication/259441017_Spectral_Networks_and_Locally_Connected_Networks_on_Graphs.
[18] DEFFERRARD M,BRESSON X,VANDERGHEYNST P.Convolutional neural networks on graphs with fast localized spectral filtering[EB/OL].[2019-09-11].https://www.researchgate.net/publication/304641430_Convolutional_Neural_Networks_on_Graphs_with_Fast_Localized_Spectral_Filtering.
[19] CHEN Long,ZHANG Hanwang,XIAO Jun,et al.SCA-CNN:spatial and channel-wise attention in convolutional networks for image captioning[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:5659-5667.
[20] HU Jie,SHEN Li,SUN Gang.Squeeze-and-excitation networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2018:7132-7141.

选择文件类型/文献管理软件名称

选择包含的内容

基于时空多图卷积网络的交通站点流量预测

Flow Prediction of Traffic Stations Based on Spatio-Temporal Multi-Graph Convolutional Network

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	葛宇然, 付强. 基于时空联合学习的城市交通流短时预测模型[J]. 计算机工程, 2023, 49(1): 270-278.
[2]	冯思芸, 施振佺, 曹阳. 基于全局时空特性的城市路网交通速度预测模型[J]. 计算机工程, 2022, 48(5): 112-117.
[3]	彭桐歆, 韩勇, 王程, 张志浩. 面向短时地铁客流量预测的混合深度学习模型[J]. 计算机工程, 2022, 48(5): 297-305.
[4]	包恒彬, 马玉鹏, 杨奉毅, 韩云飞. 基于时空注意力机制的加油站级客流量预测[J]. 计算机工程, 2021, 47(4): 291-297.
[5]	孙秀芳, 李建波, 吕志强, 董传浩. 基于多组件融合与空洞图卷积的车道占用率预测模型[J]. 计算机工程, 2021, 47(11): 62-68,76.
[6]	杨吉云, 姚锐冬, 周洁, 高凌云. 基于切比雪夫混沌映射的车联网高效认证方案[J]. 计算机工程, 2021, 47(10): 34-42,51.
[7]	姚博凡, 邓红平, 蔡铭. 基于随机抽样GMM的城市交通运行状态模式分类[J]. 计算机工程, 2020, 46(12): 36-42.
[8]	张小瑞, 陈旋, 孙伟, 葛楷. 基于深度学习的车辆再识别研究进展[J]. 计算机工程, 2020, 46(11): 1-11.
[9]	刘佳美, 徐巧枝. 基于机器学习的SDN网络流量预测与部署策略[J]. 计算机工程, 2020, 46(10): 223-230.
[10]	李月龙, 唐德华, 姜桂圆, 肖志涛, 耿磊, 张芳, 吴骏. 基于维度加权的残差LSTM短期交通流量预测[J]. 计算机工程, 2019, 45(6): 1-5.
[11]	柳长源,曹园园,罗一鸣. 基于视频图像的车辆实时检测系统[J]. 计算机工程, 2019, 45(2): 265-269,277.
[12]	李莎,孙丽珺. 基于TFPCM与随机模型的交通滞留量预测[J]. 计算机工程, 2019, 45(1): 29-34.
[13]	王路辉,王桂玲. 基于车牌识别流数据的伴随车辆发现算法[J]. 计算机工程, 2017, 43(8): 193-199.
[14]	刘擎超,蔡英凤,江浩斌,何友国,陈龙. 基于强空间集成的交通状态判别方法[J]. 计算机工程, 2017, 43(11): 297-302.
[15]	吴伟民,李泽熊,林志毅,吴汪洋. 改进飞蛾捕焰算法在网络流量预测中的应用[J]. 计算机工程, 2017, 43(10): 153-159,166.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于时空多图卷积网络的交通站点流量预测

Flow Prediction of Traffic Stations Based on Spatio-Temporal Multi-Graph Convolutional Network

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献

相关文章 15

编辑推荐

Metrics

本文评价