Traffic Flow Prediction Based on Spatio-Temporal Multi-head Graph Attention Network

LIANG Xiu-xia; XIA Man-man; HE Yue-yang; LIANG Tao

doi:10.12263/DZXB.20230474

您当前的位置：

首页 >

文章列表页 >

Traffic Flow Prediction Based on Spatio-Temporal Multi-head Graph Attention Network

PAPERS | 更新时间：2025-12-11

- Traffic Flow Prediction Based on Spatio-Temporal Multi-head Graph Attention Network
- ACTA ELECTRONICA SINICA Vol. 52, Issue 2, Pages: 500-509(2024)
- 作者机构：
  
  河北工业大学人工智能与数据科学学院，天津 300132
- 作者简介：
- 基金信息：
  
  Natural Science Foundation of Hebei Province(F2021202022)
- DOI：10.12263/DZXB.20230474
  CLC： TP273;U491.1
- Received：30 May 2023，
  
  Revised：2023-11-29，
  
  Published：25 February 2024
- 稿件说明：
移动端阅览
梁秀霞,夏曼曼,何月阳,等.基于时空多头图注意力网络的交通流预测[J].电子学报,2024,52(02):500-509.

LIANG Xiu-xia, XIA Man-man, HE Yue-yang, et al.Traffic Flow Prediction Based on Spatio-Temporal Multi-head Graph Attention Network[J].Acta Electronica Sinica, 2024, 52(02): 500-509.
梁秀霞,夏曼曼,何月阳,等.基于时空多头图注意力网络的交通流预测[J].电子学报,2024,52(02):500-509. DOI：10.12263/DZXB.20230474

LIANG Xiu-xia, XIA Man-man, HE Yue-yang, et al.Traffic Flow Prediction Based on Spatio-Temporal Multi-head Graph Attention Network[J].Acta Electronica Sinica, 2024, 52(02): 500-509. DOI：10.12263/DZXB.20230474

摘要

针对当前路网交通流量预测方法中存在的挖掘复杂的动态时空特性和长距离的空间依赖性能力不足等问题，基于多头自注意力网络构建一种新型的交通流预测模型.考虑到交通流在不同周期尺度下呈现出的高度相似性，以及静态时空特征，引入日和周这2种周期尺度下的数据张量作为模型输入，来表达交通流数据的时间相似性，并通过输入数据的时空位置编码获取其静态时空特征.考虑到交通流的动态时空特性和长距离的空间依赖性，主体模型基于多头自注意力机制分别设计时间多头注意力模块和空间多头注意力模块.时间多头注意力模块利用一个图掩码矩阵获得局部注意力，并将其融合到一个多头自注意力中，以提取交通流的动态时间特征.空间多头注意力模块利用两个图掩码矩阵获得局部注意力和全局注意力，并将其融合到一个多头自注意力中，以提取路网节点的动态空间特征和长距离的空间依赖性.最后，设计一个门控融合模块自适应地融合交通流数据的时空相关性特征.在三个真实交通流基准数据集PEMS04，PEMS07和PEMS08上进行模型的有效性验证，结果表明，所建模型在3个数据集上的3个预测精度指标与其他精度最高模型相比，平均提高了4.437%，2.930%，4.275%.

Abstract

To overcome the drawbacks of current traffic network flow prediction methods

such as the low capability of capturing highly dynamic spatio-temporal correlation and long-term spatial dependence

this paper constructs a novel traffic flow prediction model based on multi-head self-attention network. The model takes the data tensor at daily period and weekly period scales as model inputs to express the temporal similarity of traffic flow data

and obtains its static spatio-temporal characteristics by encoding the spatio-temporal position embedding of the input data. The main model designs temporal multi-head attention module and spatial multi-head attention module respectively based on multi-head self-attention mechanism for considering the dynamic spatio-temporal characteristics of traffic flow and the long-range spatial dependences. The temporal multi-head attention module obtains the local attention using a graph masking matrix and fuses it into a multi-head self-attention to extract the dynamic temporal characteristics of traffic flow. The spatial multi-head attention module obtains the local attention and global attention using two graph masking matrices and fuses them into a multi-head self-attention to extract the dynamic spatial characteristics and long-range spatial dependences of road network nodes. Finally

a gated fusion module is designed to adaptively fuse the spatio-temporal correlation characteristics of traffic flow data. The effectiveness of the proposed model is verified on three real traffic flow benchmark datasets PEMS04

PEMS07 and PEMS08

and the results show that the three prediction accuracy metrics of the proposed model on the three data sets improved by 4.437%

2.930%

and 4.275% on average compared with the other models with the highest accuracy.

关键词

Keywords

references

李长乐 , 王硕 , 岳文伟 , 等 . 面向空地一体化交通的虚拟车道: 发展阶段与关键技术 [J]. 电子学报 , 2022 , 50 ( 5 ): 1255 - 1265 .

LI C L , WANG S , YUE W W , et al . Virtual lanes for air-ground integrated transportation systems: Evolution and key techniques [J]. Acta Electronica Sinica , 2022 , 50 ( 5 ): 1255 - 1265 . (in Chinese)

LV Y S , DUAN Y J , KANG W W , et al . Traffic flow prediction with big data: A deep learning approach [J]. IEEE Transactions on Intelligent Transportation Systems , 2015 , 16 ( 2 ): 865 - 873 .

GORI M , MONFARDINI G , SCARSELLI F . A new model for learning in graph domains [C]// Proceedings of 2005 IEEE International Joint Conference on Neural Networks . Piscataway : IEEE , 2005 : 729 - 734 .

JIANG W W , LUO J Y , HE M , et al . Graph neural network for traffic forecasting: The research progress [J]. ISPRS International Journal of Geo-Information , 2023 , 12 ( 3 ): 100 .

杨建喜 , 郁超顺 , 李韧 , 等 . 基于多周期组件时空神经网络的路网通行速度预测 [J]. 交通运输系统工程与信息 , 2021 , 21 ( 3 ): 112 - 119, 139 .

YANG J X , YU C S , LI R , et al . Traffic network speed prediction via multi-periodic-component spatial-temporal neural network [J]. Journal of Transportation Systems Engineering and Information Technology , 2021 , 21 ( 3 ): 112 - 119, 139 . (in Chinese)

GUO S N , LIN Y F , WAN H Y , et al . Learning dynamics and heterogeneity of spatial-temporal graph data for traffic forecasting [J]. IEEE Transactions on Knowledge and Data Engineering , 2022 , 34 ( 11 ): 5415 - 5428 .

VASWANI A , SHAZEER N , PARMAR N , et al . Attention is all you need [C]// Proceedings of the 31st International Conference on Neural Information Processing Systems . New York : ACM , 2017 : 6000 - 6010 .

LIN Z Q , FENG J , LU Z Y , et al . DeepSTN+: Context-aware spatial-temporal neural network for crowd flow prediction in metropolis [J]. Proceedings of the AAAI Conference on Artificial Intelligence , 2019 , 33 ( 1 ): 1020 - 1027 .

ZONOOZI A , KIM J-J , LI X L , et al . Periodic-CRN: A convolutional recurrent model for crowd density prediction with recurring periodic patterns [C]// Proceedings of the 27th International Joint Conference on Artificial Intelligence . Stockholm : AAAI Press , 2018 : 3732 - 3738 .

HOANG V T , JEON H J , YOU E S , et al . Graph representation learning and its applications: A survey [J]. Sensors , 2023 , 23 ( 8 ): 4168 .

JEONG Y S , JEONG M K , OMITAOMU O A . Weighted dynamic time warping for time series classification [J]. Pattern Recognition , 2011 , 44 ( 9 ): 2231 - 2240 .

DRUCKER H , BURGES C J C , KAUFMAN L , et al . Support vector regression machines [C]// Proceedings of the 9th International Conference on Neural Information Processing Systems . New York : ACM , 1996 : 155 - 161 .

HOCHREITER S , SCHMIDHUBER J . Long short-term memory [J]. Neural Computation , 1997 , 9 ( 8 ): 1735 - 1780 .

LI Y G , YU R , SHAHABI C , et al . Diffusion convolutional recurrent neural network: Data-driven traffic forecasting [EB/OL]. ( 2017 )[2023]. http://arxiv.org/abs/1707.01926.pdf http://arxiv.org/abs/1707.01926.pdf .

YU B , YIN H T , ZHU Z X . Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting [C]// Proceedings of the 27th International Joint Conference on Artificial Intelligence . New York : ACM , 2018 : 3634 - 3640 .

WU Z H , PAN S R , LONG G D , et al . Graph WaveNet for deep spatial-temporal graph modeling [EB/OL]. ( 2019 )[2023]. https://arxiv.org/abs/1906.00121 https://arxiv.org/abs/1906.00121 .

GUO S N , LIN Y F , FENG N , et al . Attention based spatial-temporal graph convolutional networks for traffic flow forecasting [J]. Proceedings of the AAAI Conference on Artificial Intelligence , 2019 , 33 ( 1 ): 922 - 929 .

JIANG J W , HAN C K , WANG J Y . BUAA_BIGSCity: Spatial-temporal graph neural network for wind power forecasting in Baidu KDD CUP 2022 [EB/OL]. ( 2022 )[2023]. https://arxiv.org/abs/2302.11159 https://arxiv.org/abs/2302.11159 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Variable Horizon Multi-Directional Scanning Method for Time Series Anomaly Detection

Low-Light Small Target Detection Method Combining Feature Fusion Enhancement and Detail Features

Object Detection Based on EIMYOLO for High-Resolution Remote Sensing Images

Research on Improved Localization Based on Real-Time Semantic Chain List Systems

Related Author

HE Yue-yang

LIANG Tao

HUANG Yu-zhe

GUAN Yong-yuan

WEI Song-jie

JIANG Ze-tao

CHENG Liu-ming

YANG Jian-chen

Related Institution

School of Computer Science and Engineering, School of Cyber Science and Engineering, Nanjing University of Science and Technology

Guangxi Key Laboratory of Image and Graphics Processing and Intelligent Processing, Guilin University of Electronic Technology

School of Information and Technology, Shanxi University

School of Computer Science and Technology, Soochow University

School of Electronics and Information Engineering, Suzhou University of Science and Technology

⁰