基于时序图卷积网络的交通速度预测
Traffic Speed Prediction Based on Temporal Graph Convolution Network
摘要: 准确的交通流预测不仅能够为交通拥堵控制决策提供有效的支持，也为居民出行提供指导。本文结合图卷积网络与时间卷积网络，提出了一种基于时序图卷积网络的交通速度预测模型。根据交通路网拓扑结构和交通流构建动态加权图网络，引入图卷积网络挖掘交通流的空间特征，利用时间卷积网络捕获交通流的短程与长程时间相关性。融合路网空间特征和交通流近邻性、周期性等时间特征，构建基于混合深度学习框架的交通速度预测模型。以真实数据集为基础验证了模型预测效果，实验结构表明提出的模型能够刻画交通流的时空演化特征，和基准模型相比，具有良好的预测性能。
Abstract: Accurate traffic flow prediction can not only provide effective support for traffic congestion decisions, but also provide guidance for residents to travel. Combining graph convolutional network and temporal convolutional network, a traffic speed prediction based on temporal graph convolution network is proposed in this paper. A dynamic weighted graph network is constructed according to the traffic road network topology and traffic flow, a graph convolutional network is used to mine the spatial characteristics of the traffic flow, and a temporal convolutional network is used to capture the short-range and long-range temporal correlation of the traffic flow. The traffic speed prediction based on the hybrid deep learning framework is constructed by integrating the spatial characteristics of the road network with the characteristics of the proximity and periodicity of the traffic flow. Experiments on real datasets show that the proposed model can characterize the spatiotemporal evolution of traffic flow, and has good multi-step prediction performance compared with the base-line model.
文章引用：蒋欣彤, 冯慧芳. 基于时序图卷积网络的交通速度预测[J]. 计算机科学与应用, 2022, 12(6): 1544-1552. https://doi.org/10.12677/CSA.2022.126154

参考文献

[1] Nagy, A.M. and Simon, V. (2018) Survey on Traffic Prediction in Smart Cities. Pervasive and Mobile Computing, 50, 148-163. [Google Scholar] [CrossRef] 
[2] 周晓, 唐宇舟, 刘强. 基于卡尔曼滤波的道路平均速度预测模型研究[J]. 浙江工业大学学报, 2020, 48(4): 392-396+404.
[3] Hong, W.C., Dong, Y.C., Zheng, F.F. and Wei, S.Y. (2011) Hybrid Evolutionary Algorithms in a SVR Traffic Flow Forecasting Model. Applied Mathematics and Computation, 217, 6733-6747. [Google Scholar] [CrossRef] 
[4] 李文勇, 李俊卓, 王涛. 基于Box-Cox指数变换改进的ARIMA模型交通流预测方法[J]. 武汉理工大学学报(交通科学与工程版), 2020, 44(6): 974-977.
[5] 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, 2018, 18(1): 81-88.
[6] He, Z.X., Chow, C.Y. and Zhang, J.D. (2019) STCNN: A Spatio-Temporal Convolutional Neural Network for Long-Term Traffic Predic-tion. 2019 20th IEEE International Conference on Mobile Data Management (MDM), Hong Kong (China), 10-13 June 2019, 226-233. [Google Scholar] [CrossRef] 
[7] 卢生巧, 黄中祥. 基于深度学习的短时交通流预测模型[J]. 交通科学与工程, 2020, 36(3): 74-80.
[8] 陆文琦, 芮一康, 冉斌, 谷远利. 智能网联环境下基于混合深度学习的交通流预测模型[J]. 交通运输系统工程与信息, 2020, 20(3): 47-53.
[9] 冯宁, 郭晟楠, 宋超, 朱琪超, 万怀宇. 面向交通流量预测的多组件时空图卷积网络[J]. 软件学报, 2019, 30(3): 759-769.
[10] Zhao, L., Song, Y.J., Zhang, C., Liu, Y., Wang, P., Lin, T., Deng, M. and Li, H.F. (2020) T-GCN: A Temporal Graph Convolu-tional Network for Traffic Prediction. IEEE Transactions on Intelligent Transportation Systems, 21, 3848-3858. [Google Scholar] [CrossRef] 
[11] Seng, D.W., Lv, F.S., Liang, Z.Y., Shi, X.Y. and Fang, Q.M. (2021) Forecasting Traffic Flows in Irregular Regions with Multi-Graph Convolutional Network and Gated Recurrent Unit. Frontiers of Information Technology & Electronic Engineering, 22, 1179-1193. [Google Scholar] [CrossRef] 
[12] Peng, H., Du, B.W., Liu, M.S., Liu, M., Ji, S., Wang, S., et al. (2021) Dynamic Graph Convolutional Network for Long-Term Traffic Flow Prediction with Reinforcement Learning. Infor-mation Sciences, 578, 401-416. [Google Scholar] [CrossRef] 
[13] Ye, J.X., Zhao, J.J., Ye, K.J. and Xu, C. (2022) How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey. IEEE Transactions on Intelligent Transportation Systems, 23, 3904-3924. [Google Scholar] [CrossRef] 
[14] 周毅, 胡姝婷, 李伟, 承楠, 路宁, 沈学民. 图神经网络驱动的交通预测技术: 探索与挑战[J]. 物联网学报, 2021, 5(4): 1-16.
[15] Yin, X.Y., Wu, G.Z., Wei, J.Z., Shen, Y., Qi, H. and Yin, B. (2021) Deep Learning on Traffic Prediction: Methods, Analysis and Future Directions. IEEE Transac-tions on Intelligent Transportation Systems, 23, 4927-4943. [Google Scholar] [CrossRef] 
[16] Jiang, W.W. and Luo, J.Y. (2021) Graph Neural Network for Traffic Forecasting: A Survey.
[17] Simonovsky, M. and Komodakis, N. (2017) Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, 21-26 July 2017, 3693-3702. [Google Scholar] [CrossRef] 
[18] Yu, F. and Koltun, V. (2016) Multi-Scale Context Aggregation by Dilated Convolutions. International Conference on Learning Representa-tions (ICLR), Caribe Hilton, 2-4 May 2016, 1-4.
[19] Li, Y., Yu, R., Shahabi, C. and Liu, Y. (2018) Diffusion Convo-lutional Recurrent Neural Network: Data-Driven Traffic Forecasting. International Conference on Learning Representa-tions 2018, Vancouver, 30 April 2018-3 May 2018, 1-16.
[20] Fernandez-Manso, A., Quintano, C. and Fernan-dez-Manso, O. (2011) Forecast of NDVI in Coniferous Areas Using Temporal ARIMA Analysis and Climatic Data at a Regional Scale. International Journal of Remote Sensing, 32, 1595-1617. [Google Scholar] [CrossRef] 
[21] Cortes, C. and Vapnik, V.N. (1995) Support-Vector Networks. Machine Learning, 20, 273-297. [Google Scholar] [CrossRef] 
[22] Ilya, S., Oriol, V. and Quoc, V.L. (2014) Sequence to Sequence Learn-ing with Neural Networks. Annual Conference on Neural Information Processing Systems, 3104-3112. [Google Scholar] [CrossRef] 
[23] Oord, A.V.D., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A. and Kavukcuoglu, K. (2016) WaveNet: A Generative Model for Raw Audio. arXiv:1609.03499.
[24] Yu, B., Yin, H.T. and Zhu, Z.X. (2018) Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting. Proceedings of 27th International Joint Conference on Artificial Intel-ligence (IJCAI), Stockholm, 13-19 July 2018, 3634-3640. [Google Scholar] [CrossRef]