动态交通流量预测的时空注意力图卷积网络

doi:10.6040/j.issn.1672-3961.0.2023.253

摘要/Abstract

摘要： 针对现有交通流预测方法大多忽略时空耦合相关性、时空变化性以及外部特征对预测结果准确性的影响,提出一种动态交通流量预测的时空注意力图卷积网络(attention-based spatio-temporal graph convolutional network, ATST-GCN)模型。提出基于注意力的双向门控循环单元结构,从动态空间序列中提取时间相关性;构建带残差链接的多层图注意网络(graph attention network, GAT)卷积模块,深入挖掘动态空间相关性;融合时变特征与时常特征,充分利用外部静动态特征的共同作用。采用PeMS数据集对交通流量预测的准确度进行验证,试验结果表明:本研究方法能够有效提高交通流量预测精度,优于现有的多数先进方法。在PeMS08和PeMS03 数据集上,本研究方法相对 STSGCN 模型分别提高 13.44%和 10.96%,相对T-GCN 模型分别提高21.41%和 21.32%,相对 STGCN模型分别提高 8.04%和 6.55%,相对 DMSTGCN 模型分别提高 3.23%和 2.80%,相对 Trendformer 模型分别提高 2.29%和 2.00%。

关键词: 智能交通系统, 交通流量预测, 注意力机制, 时空相关性, 图卷积网络

中图分类号:

U121

邹正标,刘毅志,廖祝华,赵肄江. 动态交通流量预测的时空注意力图卷积网络[J]. 山东大学学报 (工学版), 2024, 54(5): 50-61.

ZOU Zhengbiao, LIU Yizhi, LIAO Zhuhua, ZHAO Yijiang. Attention-based spatio-temporal graph convolutional network for dynamic traffic flow prediction[J]. Journal of Shandong University(Engineering Science), 2024, 54(5): 50-61.

参考文献

[1] LIU Jing, GUAN Wei. A summary of traffic flow forecasting methods[J].Journal of Highway Transportation Research Development, 2004, 21(3): 82-85.
[2] 孙德山.支持向量机分类与回归方法研究[D]. 长沙: 中南大学, 2004. SUN Deshan. The researches on support vector machine classification and regression methods[D]. Changsha: Central South University, 2004.
[3] 初连禹,杨兆升. 基于神经网络的实时交通流量模糊预测[J]. 中国公路学报,1998(增刊1):78-82. CHU Lianyu, YANG Zhaosheng. Real-time traffic volume prediction based on fuzzy neural network[J]. China Journal of Highway and Transport, 1998(Suppl.1): 78-82.
[4] 杨兆升, 王媛, 管青. 基于支持向量机方法的短时交通流量预测方法[J]. 吉林大学学报(工学版), 2006, 36(6): 881-884. YANG Zhaosheng, WANG Yuan, GUAN Qing. Short-term traffic flow prediction method based on SVM[J]. Journal of Jilin University(Engineering and Technology Edition), 2006, 36(6): 881-884.
[5] 薛红军,陈广交,李鑫民, 等. 基于决策树理论的交通流参数短时预测[J].交通信息与安全,2016,34(3): 64-71. XUE Hongjun, CHEN Guangjiao, LI Xinmin, et al. A short-term forecasting of traffic flow parameters based on decision tree theory[J]. Journal of Transport Information and Safety, 2016, 34(3): 64-71.
[6] 赵天信. 基于卷积神经网络的交通流预测方法研究[D].兰州: 兰州理工大学,2023. ZHAO Tianxin. Research on traffic flow prediction method based on convolution neural network[D]. Lanzhou: Lanzhou University of Technology, 2023.
[7] 王媛源. 基于循环神经网络的城市区域性交通流预测研究与系统实现[D].石河子: 石河子大学,2023. WANG Yuanyuan. Research and system implementation of urban regional traffic flow prediction based on cyclic neural network[D]. Shihezi: Shihezi University, 2023.
[8] 李朝阳, 李琳, 陶晓辉.面向动态交通流预测的双流图卷积网络[J].计算机科学与探索, 2022, 16(2): 384-394. LI Chaoyang, LI Lin, TAO Xiaohui. Two-stream graph convolutional neural network for dynamic traffic flow forecasting[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(2): 384-394.
[9] DIAO Z L, WANG X, ZHANG D F, et al.Dynamic spatial-temporal graph convolutional neural networks for traffic forecasting[C] //Proceedings of the AAAI Conference, Honolulu, USA: AAAI, 2019: 890-897.
[10] 谷振宇, 陈聪, 郑家佳, 等.基于时空图卷积循环神经网络的交通流预测[J].控制与决策, 2021, 37(3): 645-653. GU Zhenyu, CHEN Cong, ZHENG Jiajia, et al. Traffic flow prediction based on STG-CRNN[J]. Control and Decision, 2021, 37(3): 645-653.
[11] 杜圣东, 李天瑞, 杨燕, 等. 一种基于序列到序列时空注意力学习的交通流预测模型[J]. 计算机研究与发展, 2020, 57(8): 1715-1728. DU Shengdong, LI Tianrui, YANG Yan, et al. A sequence-to-sequence spatial-temporal attention learning model for urban traffic flow prediction[J]. Journal of Computer Research and Development,2020, 57(8): 1715-1728.
[12] TANNER J C. Effect of weather on traffic flow[J].Nature, 1952, 169(4290): 107.
[13] KOCKELMAN K M. Changes in flow-density relationship due to environmental,vehicle,and driver characteristics[J].Transportation Research Record, 1998, 1644(1): 47-56.
[14] HRANAC R, STERZIN E, KRECHMER D, et al. Empirical studies on traffic flow in inclement weather[R]. USA: Federal Highway Administration, 2006.
[15] GUO S N, LIN Y F, FENG N, et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting[C] //Proceedings of the AAAI Conference on Artificial Intelligence. Honolulu, USA: AAAI, 2019, 33(1): 922-929.
[16] YAO H X, TANG X F, WEI H, et al. Revisiting spatial-temporal similarity: a deep learning framework for traffic prediction[C] //Proceedings of the AAAI Conference on Artificial Intelligence. Honolulu, USA: AAAI, 2019, 33(1): 5668-5675.
[17] 闫旭, 范晓亮, 郑传潘, 等. 基于图卷积神经网络的城市交通态势预测算法[J]. 浙江大学学报(工学版), 2020, 54(6): 1147-1155. YAN Xu, FAN Xiaoliang, ZHENG Chuanpan, et al. Urban traffic flow prediction algorithm based on graph convolutional neural networks[J]. Journal of Zhejiang University(Engineering Science), 2020, 54(6): 1147-1155.
[18] HAMED M M, AL-MASAEID H R, SAID Z M B. Short-term prediction of traffic volume in urban arterials[J]. Journal of Transportation Engineering, 1995, 121(3): 249-254.
[19] SHAO H, SOONG B H.Traffic flow prediction with long short-term memory networks(LSTMS)[C] //In Proceedings of 2016 IEEE Region 10 Conference(TENCON). Singapore: IEEE, 2016: 2986-2989.
[20] QU Z, LI H, LI Z, et al.Short-term traffic flow forecasting method with M-B-LSTM hybrid network[J].IEEE Transactions on Intelligent Transportation Systems, 2020, 23(1): 225-235.
[21] SHI X, CHEN Z, WANG H, et al. Convolutional LSTM network: a machine learning approach for precipitation nowcasting[J]. Advances in Neural Information Processing Systems, 2015, 28: 802-810.
[22] 刘铭, 鱼昕. 基于改进LSTM算法的短时交通流量预测[J]. 桂林理工大学学报, 2021, 41(2): 409-414. LIU Ming, YU Xin. Short-term traffic flow prediction based on MVF-LSTM algorithm[J]. Journal of Guilin University of Technology, 2021, 41(2): 409-414.
[23] DEY R, SALEM F M. Gate-variants of gated recurrent unit(gru)neural networks[C] //Proceedings of 2017 IEEE 60th International Midwest Symposium on Circuits and Systems(MWSCAS). Boston, USA: IEEE, 2017: 1597-1600.
[24] MOGHAR A, HAMICHE M. Stock market prediction using LSTM recurrent neural network[J]. Procedia Computer Science, 2020, 170: 1168-1173.
[25] FU R, ZHANG Z, LI L. Using LSTM and GRU neural network methods for traffic flow prediction[C] //2016 31st Youth Academic Annual Conference of Chinese Association of Automation(YAC). Wuhan, China: IEEE, 2016: 324-328.
[26] SCHUSTER M, PALIWAL K K.Bidirectional recurrent neural networks[J].IEEE Transactions on Signal Processing, 1997, 45(11): 2673-2681.
[27] SU Y, ZHAO Y J, NIU C H, et al.Robust anomaly detection for multivariate time series through stochastic recurrent neural network[C] // Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Halifax, Canada: ACM, 2019: 2828-2837.
[28] ZHUANG W Q, CAO Y B. Short-term traffic flow prediction based on a k-nearest neighbor and bidirectional long short-term memory model[J]. Applied Sciences, 2023, 13(4): 2681.
[29] ZHUANG W Q, CAO Y B. Short-term traffic flow prediction based on CNN-BILSTM with multicomponent information[J]. Applied Sciences, 2022, 12(17): 8714.
[30] WANG S Y, SHAO C F, ZHANG J, et al. Traffic flow prediction using bi-directional gated recurrent unit method[J]. Urban Informatics, 2022, 1(1): 16.
[31] 徐丽, 符祥远, 李浩然. 基于门控卷积的时空交通流预测模型[J]. 计算机应用, 2023, 43(9): 2760-2765. XU Li, FU Xiangyuan, LI Haoran. Spatial-temporal traffic flow prediction model based on gated convolution[J]. Journal of Computer Applications, 2023, 43(9): 2760-2765.
[32] 范晓亮, 彭朝鹏, 郑传潘, 等. 面向大规模交通网络的时空关联挖掘方法[J]. 清华大学学报(自然科学版), 2023, 63(9): 1317-1325. FAN Xiaoliang, PENG Zhaopeng, ZHENG Chuanpan, et al. Spatio-temporal correlation mining method for large-scale traffic networks[J]. Journal of Tsinghua University(Science and Technology), 2023, 63(9): 1317-1325.
[33] BHAUMIK K K, NILOY F F, MAHMUD S, et al. STLGRU: spatio-temporal lightweight graph GRU for traffic flow prediction[EB/OL].(2022-12-08)[2023-12-20]. https: //arxiv.org/abs/2212.04548.
[34] SONG C, LIN Y F, GUO S N, et al. Spatial-temporal synchronous graph convolutional networks: a new framework for spatial-temporal network data forecasting[C] //Proceedings of the AAAI Conference on Artificial Intelligence. New York, USA: AAAI, 2020, 34(1): 914-921.
[35] ZHAO L, SONG Y J, 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.
[36] YU B, YIN H T, ZHU Z X. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting[EB/OL].(2017-09-14)[2023-12-20]. https: //arxiv.org/abs/1709.04875.
[37] LI M Z, ZHU Z X.Spatial-temporal fusion graph neural networks for traffic flow forecasting[C] // Proceedings of the AAAI Conference on Artificial Intelligence. Vancouver, Canada: AAAI, 2021, 35(5): 4189-4196.
[38] ZHENG C P, FAN X L, WANG C, et al.Gman: a graph multi-attention network for traffic prediction[C] // Proceedings of the AAAI Conference on Artificial Intelligence. New York, USA: AAAI, 2020, 34(1): 1234-1241.
[39] HAN L Z, DU B W, SUN L L, et al. Dynamic and multi-faceted spatio-temporal deep learning for traffic speed forecasting[C] //Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. Singapore: [s.n.] , 2021: 547-555.
[40] 刘志, 陈洋, 周涵林, 等.基于时空动态循环图卷积网络的交通流预测[J].浙江工业大学学报, 2023, 51(3): 282-288. LIU Zhi, CHEN Yang, ZHOU Hanlin, et al. Traffic flow prediction based on spatiotemporal dynamic recurrent graph convolutional framework[J]. Journal of Zhejiang University of Technology, 2023, 51(3): 282-288.
[41] 江辉,张阳,杨书敏, 等. 基于动态时空卷积网络的车道级交通流预测[J/OL]. 武汉理工大学学报(交通科学与工程版).(2023-07-04)[2023-12-20]. http: //kns.cnki.net/kcms/detail/42.1824.U.20230704.1417.062.html. JIANG Hui, ZHANG Yang, YANG Shumin, et al. Lane-level traffic flow prediction based on dynamic spatio-temporal convolution network[J/OL]. Journal of Wuhan University of Technology(Transportation Science & Engineering).(2023-07-04)[2023-12-20]. http://kns.cnki.net/kcms/detail/42.1824.U.20230704.1417.062.html.
[42] 谷振宇, 陈聪, 郑家佳, 等. 考虑时空相似性的动态图卷积神经网络交通流预测[J]. 控制与决策, 2023,38(12): 3399-3408. GU Zhenyu, CHEN Cong, ZHENG Jiajia, et al. Traffic flow prediction based on STS-DGCN[J]. Control and Decision, 2023, 38(12): 3399-3408.
[43] BURCHI M, VIELZEUF V. Efficient conformer: progressive downsampling and grouped attention for automatic speech recognition[C] //2021 IEEE Automatic Speech Recognition and Understanding Workshop(ASRU). Cartagena, Colombia: IEEE, 2021: 8-15.
[44] 李志鹃. 时空注意图卷积网络的交通流量预测研究[J]. 智能计算机与应用, 2022, 12(11): 253-257. LI Zhijuan. Research on traffic flow prediction of spatio-temporal attention graph convolutional networks[J]. Intelligent Computer and Applications, 2022, 12(11): 253-257.
[45] 李小妍. 基于图神经网络的交通流量预测[D]. 成都: 电子科技大学, 2020. LI Xiaoyan. Traffic forecasting based on graph neural network[D]. Chengdu: University of Electronic Science and Technology of China, 2020.
[46] YANG Z J, WANG C. Short-term traffic flow prediction based on AST-MTL-CNN-GRU[J]. IET Intelligent Transport Systems, 2023, 17(11): 2205-2220.
[47] CAI B H, WANG Y H, HUANG C, et al. GLSNN network: amulti-scale spatiotemporal prediction model for urban traffic flow[J]. Sensors, 2022, 22(22): 8880.
[48] CHEN X. Research on short-term traffic flow forecasting model based on LSTM[J]. World Scientific Research Journal, 2020, 6(5): 191-200.
[49] 刘建松,康雁,李浩, 等.面向交通流量预测的时空Graph-CoordAttention网络[J].计算机科学,2023,50(增刊1): 568-574. LIU Jiansong, KANG Yan, LI Hao, et al. Spatial-temporal Graph-CoordAttention network for traffic forecasting[J]. Computer Science, 2023, 50(Suppl.1): 568-574.
[50] DO L N N, VU H L, VO B Q, et al. An effective spatial-temporal attention based neural network for traffic flow prediction[J]. Transportation Research Part C: Emerging Technologies, 2019, 108: 12-28.
[51] CHEN C, PETTY K, SKABARDONIS A, et al. Freeway performance measurement system: mining loop detector data[J]. Transportation Research Record, 2001, 1748(1): 96-102.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed