基于ARMA过滤器的时空图卷积网络短时交通流预测

doi:10.3778/j.issn.1002-8331.2310-0041

摘要/Abstract

摘要： 针对大多数现有的时空融合图卷积网络模型在分析交通流数据所采用的过滤器提取空间特征时，可能会导致网络节点特征过于平滑从而丢失原始信息、计算量大等问题，将一种能有效逼近任何所需响应的基于自回归移动平均（ARMA）过滤器的图卷积网络与门控循环单元（GRU）相融合，提出一种基于ARMA过滤器的时空图卷积网络短时交通流预测模型。在该模型中，采用GRU提取交通流数据的时间特征；利用基于ARMA过滤器的图卷积网络提取空间特征；通过平均池化层得到预测结果。在美国加州高速公路公开数据集PeMS04、PeMS07和PeMS08上进行了实验。结果表明，相比于切比雪夫多项式等传统的过滤器，ARMA过滤器能够更好地捕获交通流数据中的空间特征且能有效提升计算效率，在三组数据集上相比于GRU、DCRNN、T-GCN、ASTGCN和MSTGCN等基准模型，新模型的平均绝对误差指标平均降低了26.28%、28.25%、39.59%、26.24%和19.07%，训练时间平均降低了26.54%、95.00%、78.87%、93.58%和89.03%。

关键词: 交通流预测, 图卷积网络, ARMA过滤器, 门控循环单元, 时空特征提取

Abstract: In view of the problems that the filter used by most existing spatial-temporal fusion graph convolutional network models in the analysis of traffic flow data may lead to the loss of original information when extracting spatial features due to too smooth network node features and large amount of computation, an auto-regressive moving average (ARMA) filter based graph convolutional network which can effectively approximate any required response is integrated with the gated recurrent unit (GRU). An ARMA filter based spatio-temporal graph convolutional network short-term traffic flow prediction model is proposed. In this model, the GRU is used to extract the temporal features of traffic flow data. Then, a graph convolutional network based on the ARMA filter is used to extract the spatial features. Finally, the prediction results are obtained by means of the average pooling layer. Experiments are conducted on California Highway open datasets PeMS04, PeMS07 and PeMS08 in the United States. The results show that compared with Chebyshev polynomial and other traditional filters, ARMA filter can better capture the spatial correlation in traffic flow data and effectively improve the computational efficiency. Compared with GRU, DCRNN, T-GCN, ASTGCN and MSTGCN, the proposed model performs much better. The average absolute errors are decreased by 26.28%, 28.25%, 39.59%, 26.24% and 19.07% on average, and the training times are decreased by 26.54%, 95.00%, 78.87%, 93.58% and 89.03% on average.

Key words: traffic flow prediction, graph convolutional network, ARMA filter, gated recurrent unit, spatio-temporal feature extraction

肖培成, 曹阳, 沈琴琴, 施佺. 基于ARMA过滤器的时空图卷积网络短时交通流预测[J]. 计算机工程与应用, 2024, 60(21): 308-314.

XIAO Peicheng, CAO Yang, SHEN Qinqin, SHI Quan. Short-Term Traffic Flow Prediction Based on Spatial-Temporal Graph Convolutional Network with ARMA Filter[J]. Computer Engineering and Applications, 2024, 60(21): 308-314.

参考文献

[1] MEDINA-SALGADO B, SáNCHEZ-DELACRUZ E, POZOS-PARRA P, et al. Urban traffic flow prediction techniques: a review[J]. Sustainable Computing: Informatics and Systems, 2022, 35: 100739.
[2] KAFFASH S, NGUYEN A T, ZHU J. Big data algorithms and applications in intelligent transportation system: a review and bibliometric analysis[J]. International Journal of Production Economics, 2021, 231: 107868.
[3] 姚俊峰, 何瑞, 史童童, 等. 基于机器学习的交通流预测方法综述[J]. 交通运输工程学报, 2023, 23(3): 44-67.
YAO J F, HE R, SHI T T, et al. Review on machine learning-based traffic flow prediction methods[J]. Journal of Traffic and Transportation Engineering, 2023, 23(3): 44-67.
[4] YUAN H, LI G. A survey of traffic prediction: from spatio-temporal data to intelligent transportation[J]. Data Science and Engineering, 2021, 6(1): 63-85.
[5] YANG H, LI X, QIANG W, et al. A network traffic forecasting method based on SA optimized ARIMA-BP neural network[J]. Computer Networks, 2021, 193: 108102.
[6] CAI L, ZHANG Z, YANG J, et al. A noise-immune Kalman filter for short-term traffic flow forecasting[J]. Physica A: Statistical Mechanics and Its Applications, 2019, 536: 122601.
[7] SHEN Q Q, CAO Y, ZENG B, et al. Stable computation of least squares problems of the OGM(1,N) model and short-term traffic flow prediction[J]. East Asian Journal on Applied Mathematics, 2022, 12(2): 264-284.
[8] LIN G, LIN A, GU D. Using support vector regression and K-nearest neighbors for short-term traffic flow prediction based on maximal information coefficient[J]. Information Sciences, 2022, 608: 517-531.
[9] RASKAR C, NEMA S. Metaheuristic enabled modified hidden Markov model for traffic flow prediction[J]. Computer Networks, 2022, 206: 108780.
[10] YIN X, WU G, WEI J, et al. Deep learning on traffic prediction: methods, analysis, and future directions[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(6): 4927-4943.
[11] TEDJOPURNOMO D A, BAO Z, ZHENG B, et al. A survey on modern deep neural network for traffic prediction: trends, methods and challenges[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(4): 1544-1561.
[12] 张壮壮, 屈立成, 李翔, 等. 基于时空卷积神经网络的数据缺失交通流预测[J]. 计算机工程与应用, 2022, 58(7): 259-265.
ZHANG Z Z, QU L C, LI X, et al. Traffic flow prediction with missing data based on spatial-temporal convolutional neural networks[J]. Computer Engineering and Applications, 2022, 58(7): 259-265.
[13] 吴博, 梁循, 张树森, 等. 图神经网络前沿进展与应用[J]. 计算机学报, 2022, 45(1): 35-68.
WU B, LIANG X, ZHANG S S, et al. Advances and applications in graph neural network[J]. Chinese Journal of Computers, 2022, 45(1): 35- 68.
[14] LI Y, YU R, SHAHABI C, et al. Diffusion convolutional recurrent neural network: data-driven traffic forecasting[C]//Proceedings of the International Conference on Learning Representations, Vancouver, 2018.
[15] 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, 2020, 21(9): 3848-3858.
[16] 陈丹蕾, 陈红, 任安虎. 考虑时空影响下的图卷积网络短时交通流预测[J]. 计算机工程与应用, 2021, 57(13): 269-275.
CHEN D L, CHEN H, REN A H. Short-time traffic flow prediction of graph convolutional network considering influence of space and time[J]. Computer Engineering and Applications, 2021, 57(13): 269-275.
[17] 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, Hawaii, 2019: 922-929.
[18] 陈喜群, 周凌霄, 曹震. 基于图卷积网络的路网短时交通流预测研究[J]. 交通运输系统工程与信息, 2020, 20(4): 49-55.
CHEN X Q, ZHOU L X, CAO Z. Short-term network-wide traffic prediction based on graph convolutional network[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(4): 49-55.
[19] TREMBLAY N, GON?ALVES P, BORGNAT P. Design of graph filters and filterbanks[M]//Cooperative and graph signal processing.[S.l.]: Academic Press, 2018: 299-324.
[20] BRUNA J, ZAREMBA W, SZLAM A, et al. Spectral networks and deep locally connected networks on graphs[C]// Proceedings of the International Conference on Learning Representations, Banff, 2014.
[21] DEFFERRARD M, BRESSON X, VANDERGHEYNST P. Convolutional neural networks on graphs with fast localized spectral filtering[C]//Advances in Neural Information Processing Systems, 2016: 3844-3852.
[22] WELLING M, KIPF T N. Semi-supervised classification with graph convolutional networks[C]//Proceedings of the International Conference on Learning Representations, Toulon, 2017.
[23] 戴俊明, 曹阳, 沈琴琴, 等. 基于多时空图卷积网络的交通流预测[J]. 计算机应用研究, 2022, 39(3): 780-784.
DAI J M, CAO Y, SHEN Q Q, et al. Traffic flow prediction based on multi-spatial-temporal graph convolutional network[J]. Application Research of Computers, 2022, 39(3): 780-784.
[24] XU B, SHEN H, CAO Q, et al. Graph convolutional networks using heat kernel for semi-supervised learning[C]// Proceedings of the 28th International Joint Conference on Artificial Intelligence, Macao, 2019: 1928-1934.
[25] BIANCHI F M, GRATTAROLA D, LIVI L, et al. Graph neural networks with convolutional ARMA filters[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(7): 3496-3507.
[26] CHUNG F R K. Spectral graph theory[M]. [S.l.]:American Mathematical Society, 1997: 6-7.