计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (6A): 220200042-7.doi: 10.11896/jsjkx.220200042

• 大数据&数据科学 • 上一篇    下一篇

面向交通流量预测的时空Graph-CoordAttention网络

刘建松, 康雁, 李浩, 王韬, 王海宁   

  1. 云南大学软件学院 昆明 650504
  • 出版日期:2023-06-10 发布日期:2023-06-12
  • 通讯作者: 李浩(lihao707@ynu.edu.cn)
  • 作者简介:(12020219072@mail.ynu.edu.cn)
  • 基金资助:
    国家自然科学基金(61762092);云南省软件工程重点实验室开放基金项目(2020SE303);云南省科技厅重大专项:面向智慧旅游的行业资源共享及业务协同关键技术研究及应用(202002AD080047);材料基因工程-基于Metcloud的集成计算功能模块计算软件开发(2019CLJY06/202002AB080001-6);云南省稀贵金属材料基因工程-稀贵金属材料高通量集成计算和数据分析技术研发及示范应用(2019ZE001-1,202002AB080001)

Spatial-Temporal Graph-CoordAttention Network for Traffic Forecasting

LIU Jiansong, KANG Yan, LI Hao, WANG Tao, WANG Hailing   

  1. School of Software,Yunnan University,Kunming 650504,China
  • Online:2023-06-10 Published:2023-06-12
  • About author:LIU Jiansong,born in 1996,postgra-duate.His main research interests include deep learning and machine lear-ning. LI Hao,born in 1970,Ph.D,professor.His main research interests include distributed computing,grid and cloud computing.
  • Supported by:
    National Natural Science Foundation of China(61762092),Yunnan Provincial Software Engineering Key Laboratory Open Fund Project(2020SE303),Major Special Project of Yunnan Provincial Science and Technology Department:Key Technology Research and Application of Industry Resource Sharing and Business Collaboration for Smart Tourism(202002AD080047),Genetic Engineering of Mate-rials-Metcloud-based Computational Software Development for Integrated Computational Function Modules(2019CLJY06/202002AB080001-6) and Genetic Engineering of Rare Precious Metal Materials in Yunnan Province-Research and Development of High-Throughput Integrated Computing and Data Analysis Technology for Rare Precious Metal Materials and Demonstration Applications(2019ZE001-1,202002AB080001).

PDF (PC)

289

摘要： 交通预测是城市智能交通系统的一个重要研究组成部分,使人们的出行更加效率和安全。由于复杂的时间和空间依赖性,准确预测交通流量仍然是一个巨大的挑战。近年来,图卷积网络(GCN)在交通预测方面表现出巨大的潜力,但基于GCN的模型往往侧重于单独捕捉时间和空间的依赖性,忽视了时间和空间依赖性之间的动态关联性,不能很好地融合它们。此外,以前的方法使用现实世界的静态交通网络来构建空间邻接矩阵,这可能忽略了动态的空间依赖性。为了克服这些局限性,并提高模型的性能,提出了一种新颖的时空Graph-CoordAttention网络(STGCA)。具体来说,提出了时空同步模块,用来建模不同时刻的时空依赖交融关系。然后,提出了一种动态图学习的方案,基于车流量之间数据关联,挖掘出潜在的图信息。在4个公开的数据集上和现有基线模型进行对比实验,STGCA表现了优异的性能。

关键词: 交通流量预测, 时空预测, 图卷积网络, 注意力机制, 时空依赖

Abstract: Traffic prediction is an important research component of urban intelligent transportation systems to make our travel more efficient and safer.Accurately predicting traffic flow remains a huge challenge due to complex temporal and spatial depen-dencies.In recent years,graph convolutional network(GCN) has shown great potential for traffic prediction,but GCN-based mo-dels tend to focus on capturing temporal and spatial dependencies,ignoring the dynamic correlation between temporal and spatial dependencies and failing to integrate them well.In addition,previous approaches use real-world static traffic networks to construct spatial adjacency matrices,which may ignore the dynamic spatial dependencies.To overcome these limitations and improve the performance of the model,a novel spatial-temporal Graph-CoordAttention network(STGCA) is proposed.Specifically,the spatial-temporal synchronization module is proposed to model the spatial-temporal dependence of the crossing relations at different moments.Then,a dynamic graph learning scheme is proposed to mine potential graph information based on data correlation between traffic flows.Compared with the existing baseline models on four publicly available datasets,STGCA exhibits excellent perfor-mance.

Key words: Traffic flow forecast, Spatial-temporal forecasting, Graph convolution network, Attention mechanism, Spatial-Temporal dependence

中图分类号: 

  • TP181
[1]LIU J,WEI G.A Summary of Traffic Flow Forecasting Methods[J].Journal of Highway and Transportation Research and Development,2004,21(3):82-85.
[2]ZHANG J,WANG F Y,WANG K,et al.Data-driven intelligent transportation systems:A survey[J].IEEE Transactions on Intelligent Transportation Systems,2011,12(4):1624-1639.
[3]XIAO L,PENG X,WANG Z,et al.Research on traffic monitoring network and its traffic flow forecast and congestion control model based on wireless sensor networks[C]//2009 International Conference on Measuring Technology and Mechatronics Automation.IEEE,2009:142-147.
[4]LI Y,YU R,SHAHABI C,et al.Diffusion convolutional recurrent neural network:Data-driven traffic forecasting[J].arXiv:1707.01926,2017.
[5]ZHAO L,SONG Y,ZHANG C,et al.T-gcn:A temporal graph convolutional network for traffic prediction[J].IEEE Transactions on Intelligent Transportation Systems,2019,21(9):3848-3858.
[6]BAI J,ZHU J,SONG Y,et al.A3t-gcn:Attention temporalgraph convolutional network for traffic forecasting[J].ISPRS International Journal of Geo-Information,2021,10(7):485.
[7]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[8]CHO K,VAN MERRINBOER B,GULCEHRE C,et al.Learning phrase representations using RNN encoder-decoder for statistical machine translation[J].arXiv:1406.1078,2014.
[9]WU Z,PAN S,LONG G,et al.Graph wavenet for deep spatial-temporal graph modeling[J].arXiv:1906.00121,2019.
[10]LI M,ZHU Z.Spatial-temporal fusion graph neural networksfor traffic flow forecasting[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021,35(5):4189-4196.
[11]KAMARIANAKIS Y,PRASTACOS P.Forecasting traffic flow conditions in an urban network:Comparison of multivariate and univariate approaches[J].Transportation Research Record,2003,1857(1):74-84.
[12]AHMED M S,COOK A R.Analysis of freeway traffic time-series data by using Box-Jenkins techniques[M].Transportation Research Board,1979.
[13]WU C H,HO J M,LEE D T.Travel-time prediction with support vector regression[J].IEEE Transactions on Intelligent Transportation Systems,2004,5(4):276-281.
[14]OKUTANI I,STEPHANEDES Y J.Dynamic prediction of traffic volume through Kalman filtering theory[J].Transportation Research Part B:Methodological,1984,18(1):1-11.
[15]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//Advances in Neural Information Processing Systems.2017:5998-6008.
[16]LIU Y,ZHENG H,FENG X,et al.Short-term traffic flow prediction with Conv-LSTM[C]//2017 9th International Confe-rence on Wireless Communications and Signal Processing(WCSP).IEEE,2017:1-6.
[17]YAO H,WU F,KE J,et al.Deep multi-view spatial-temporal network for taxi demand prediction[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2018:2588-2595.
[18]BRUNA J,ZAREMBA W,SZLAM A,et al.Spectral networks and locally connected networks on graphs[J].arXiv:1312.6203,2013.
[19]DEFFERRARD M,BRESSON X,VANDERGHEYNST P.Convolutional neural networks on graphs with fast localized spectral filtering[J].Advances in Neural Information Processing Systems,2016,29:3844-3852.
[20]KIPF T N,WELLING M.Semi-supervised classification withgraph convolutional networks[J].arXiv:1609.02907,2016.
[21]HAMILTON W,YING Z,LESKOVEC J.Inductive representation learning on large graphs[J].Advances in Neural Information Processing Systems,2017,30:1024-1034.
[22]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph attention networks[J].arXiv:1710.10903,2017.
[23]Spatio-Temporal Graph Convolutional Networks:A Deep Learning Framework for Traffic Forecasting[C]//IJCAI-18.2017:3634-3640.
[24]GUO S,LIN Y,FENG N,et al.Attention based spatial-temporal graph convolutional networks for traffic flow forecasting[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2019:922-929.
[25]WANG X,MA Y,WANG Y,et al.Traffic flow prediction via spatial temporal graph neural network[C]//Proceedings of the Web Conference 2020.2020:1082-1092.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发