面向数据稀缺场景的智能交通流量预测
作者:
中图分类号:

TP311

基金项目:

国家自然科学基金(62301099, 62071077); 中国博士后科学基金(2023MD734137); 重庆市自然科学基金创新发展联合基金(2022NSCQ-LZX0191); 重庆市教委科学技术研究项目(KJQN202300638)


Intelligent Traffic Flow Prediction for Data Scarcity Scenarios
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [32]
  • | | | |
  • 文章评论
    摘要:

    交通流预测是智能交通系统(intelligent transportation system, ITS)中交通管理的重要基础和热门研究方向. 传统的交通流预测方法通常需要借助大量高质量历史观测数据进行预测, 而针对更为普遍的数据稀缺的交通路网场景预测精度则急剧下降. 针对这一问题, 提出一种基于时空图卷积网络的迁移学习模型(transfer learning based on spatial-temporal graph convolutional network, TL-STGCN), 结合数据充足的源路网的交通流特征, 辅助预测数据稀缺的目标路网未来交通流. 首先, 采用基于时间注意力的时空图卷积网络学习源路网和目标路网交通流数据的时空特征表示; 其次, 结合迁移学习方法, 提取两个路网特征表示的域不变时空特征; 最后, 利用这些域不变时空特征对目标路网未来交通流做出预测. 为了验证模型的有效性, 在真实世界数据集上进行实验. 结果表明, 与现有方法对比, TL-STGCN在平均绝对误差、均方根误差以及平均绝对百分比误差指标中均取得最高精度, 证明对于数据稀缺的交通路网预测任务, TL-STGCN具有更好的预测性能.

    Abstract:

    Traffic flow prediction is an important foundation and a hot research direction for traffic management in intelligent transportation systems (ITS). Traditional methods for traffic flow prediction typically rely on a large amount of high-quality historical observation data to achieve accurate predictions, but the prediction accuracy significantly decreases in more common scenarios with data scarcity in traffic networks. To address this problem, a transfer learning model is proposed based on spatial-temporal graph convolutional networks (TL-STGCN), which leverages traffic flow features from a source network with abundant data to assist in predicting future traffic flow in a target network with data scarcity. Firstly, a spatial-temporal graph convolutional network based on time attention is employed to learn the spatial and temporal features of the traffic flow data in both the source and target networks. Secondly, domain-invariant spatial-temporal features are extracted from the representations of the two networks using transfer learning techniques. Lastly, these domain-invariant features are utilized to predict the future traffic flow in the target network. To validate the effectiveness of the proposed model, experiments are conducted on real-world datasets. The results demonstrate that TL-STGCN outperforms existing methods by achieving the highest accuracy in mean absolute error, root mean square error, and mean absolute percentage error, which proves that TL-STGCN provides more accurate traffic flow predictions for scenarios with data scarcity in traffic networks.

    参考文献
    [1] Yuan HT, Li GL. A survey of traffic prediction: From spatio-temporal data to intelligent transportation. Data Science and Engineering, 2021, 6(1): 63–85.
    [2] Saleem TJ, Chishti MA. Deep learning for the Internet of Things: Potential benefits and use-cases. Digital Communications and Networks, 2021, 7(4): 526–542.
    [3] Cheng XY, Zhang RQ, Zhou J, Xu W. DeepTransport: Learning spatial-temporal dependency for traffic condition forecasting. In: Proc. of the 2018 Int’l Joint Conf. on Neural Networks (IJCNN). Rio de Janeiro: IEEE, 2018. 1–8. [doi: 10.1109/IJCNN.2018.8489600]
    [4] Zhuang WQ, Cao YB. Short-term traffic ?ow prediction based on CNN-BiLSTM with multicomponent information. Applied Sciences, 2022, 12(17): 8714.
    [5] Guo K, Hu YL, Qian Z, Hao L, Zhang K, Sun YF, Gao JB, Yin BC. Optimized graph convolution recurrent neural network for traffic prediction. IEEE Trans. on Intelligent Transportation Systems, 2021, 22(2): 1138–1149.
    [6] Cui ZY, Henrickson K, Ke RM, Wang YH. Traffic graph convolutional recurrent neural network: A deep learning framework for network-scale traffic learning and forecasting. IEEE Trans. on Intelligent Transportation Systems, 2020, 21(11): 4883–4894.
    [7] 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. 软件学报, 2019, 30(3): 759–769. http://www.jos.org.cn/1000-9825/5697.htm
    Feng N, Guo SN, Song C, Zhu QC, Wan HY. Multi-component spatial-temporal graph convolution networks for traffic flow forecasting. Ruan Jian Xue Bao/Journal of Software, 2019, 30(3): 759–769 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/5697.htm
    [8] Yu B, Yin HT, Zhu ZX. Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. In: Proc. of the 27th Int’l Joint Conf. on Artificial Intelligence. Stockholm: AAAI Press, 2018. 2634–3640.
    [9] Wang SZ, Miao H, Li JY, Cao JN. Spatio-temporal knowledge transfer for urban crowd flow prediction via deep attentive adaptation networks. IEEE Trans. on Intelligent Transportation Systems, 2022, 23(5): 4695–4705.
    [10] Li JY, Guo FC, Sivakumar A, Dong YJ, Krishnan R. Transferability improvement in short-term traffic prediction using stacked LSTM network. Transportation Research Part C: Emerging Technologies, 2021, 124: 102977.
    [11] Wan XB, Liu H, Xu H, Zhang XC. Network traffic prediction based on LSTM and transfer learning. IEEE Access, 2022, 10: 86181–86190.
    [12] Wang LY, Geng X, Ma XJ, Liu F, Yang Q. Cross-city transfer learning for deep spatio-temporal prediction. arXiv:1802.00386, 2018.
    [13] Tang YH, Qu A, Chow AHF, Lam WHK, Wong SC, Ma W. Domain adversarial spatial-temporal network: A transferable framework for short-term traffic forecasting across cities. In: Proc. of the 31st ACM Int’l Conf. on Information & Knowledge Management. Atlanta: ACM, 2022. 1905–1915. [doi: 10.1145/3511808.3557294]
    [14] Mou LT, Zhao PF, Xie HT, Chen YY. T-LSTM: A long short-term memory neural network enhanced by temporal information for traffic ?ow prediction. IEEE Access, 2019, 7: 98053–98060.
    [15] Poonia P, Jain VK. Short-term traffic flow prediction: Using LSTM. In: Proc. of the 2020 Int’l Conf. on Emerging Trends in Communication, Control and Computing (ICONC3). Lakshmangarh: IEEE, 2020. 1–4. [doi: 10.1109/ICONC345789.2020.9117329]
    [16] Zhang D, Kabuka MR. Combining weather condition data to predict traffic flow: A GRU-based deep learning approach. IET Intelligent Transport Systems, 2018, 12(7): 578–585.
    [17] Ma XL, Dai Z, He ZB, Ma JH, Wang Y, Wang YP. Learning traffic as images: A deep convolutional neural network for large-scale transportation network speed prediction. Sensors, 2017, 17(4): 818.
    [18] Ye JX, Zhao JJ, Ye KJ, Xu CZ. How to build a graph-based deep learning architecture in traffic domain: A survey. IEEE Trans. on Intelligent Transportation Systems, 2022, 23(5): 3904–3924.
    [19] 吴博, 梁循, 张树森, 徐睿. 图神经网络前沿进展与应用. 计算机学报, 2022, 45(1): 35–68.
    Wu B, Liang X, Zhang SS, Xu R. Advances and applications in graph neural network. Chinese Journal of Computers, 2022, 45(1): 35–68 (in Chinese with English abstract).
    [20] Zhao L, Song YJ, Zhang C, Liu Y, Wang P, Lin T, Deng M, Li HF. T-GCN: A temporal graph convolutional network for traffic prediction. IEEE Trans. on Intelligent Transportation Systems, 2020, 21(9): 3848–3858.
    [21] Zhang ZC, Li M, Lin X, Wang YH, He F. Multistep speed prediction on traffic networks: A deep learning approach considering spatio-temporal dependencies. Transportation Research Part C: Emerging Technologies, 2019, 105: 297–322.
    [22] Guo SN, Lin YF, Feng N, Song C, Wan HY. Attention based spatial-temporal graph convolutional networks for traffic ?ow forecasting. In: Proc. of the 33rd AAAI Conf. on Arti?cial Intelligence. Honolulu: AAAI, 2019. 922–929. [doi: 10.1609/aaai.v33i01.3301922]
    [23] 赵文竹, 袁冠, 张艳梅, 乔少杰, 王森章, 张雷. 多视角融合的时空动态GCN城市交通流量预测. 软件学报, 2024, 35(4): 1751–1773. http://www.jos.org.cn/1000-9825/7018.htm
    Zhao WZ, Yuan G, Zhang YM, Qiao SJ, Wang SZ, Zhang L. Multi-view fused spatial-temporal dynamic GCN for urban traffic flow prediction. Ruan Jian Xue Bao/Journal of Software, 2024, 35(4): 1751–1773 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/7018.htm
    [24] Zhuang FZ, Qi ZY, Duan KY, Xi DB, Zhu YC, Zhu HS, Xiong H, He Q. A comprehensive survey on transfer learning. Proc. of the IEEE, 2021, 109(1): 43–76.
    [25] Ren Y, Chen X, Wan S, Xie KQ, Bian KG. Passenger flow prediction in traffic system based on deep neural networks and transfer learning method. In: Proc. of the 4th Int’l Conf. on Intelligent Transportation Engineering (ICITE). Singapore: IEEE, 2019. 115–120. [doi: 10.1109/ICITE.2019.8880220]
    [26] Shen J, Qu YR, Zhang WN, Yu Y. Wasserstein distance guided representation learning for domain adaptation. In: Proc. of the 32nd AAAI Conf. on Arti?cial Intelligence. New Orleans: AAAI, 2018. 4058–4065. [doi: 10.1609/aaai.v32i1.11784]
    [27] Hussain B, Afzal MK, Ahmad S, Mostafa AM. Intelligent traffic flow prediction using optimized GRU model. IEEE Access, 2021, 9: 100736–100746.
    [28] Wu F, Zhang TY, de Souza AH Jr, Fifty C, Yu T, Weinberger KQ. Simplifying graph convolutional networks. In: Proc. of the 36th Int’l Conf. on Machine Learning. Long Beach: ICML, 2019. 6861–6871.
    [29] Mo JQ, Gong ZG. Cross-city multi-granular adaptive transfer learning for traffic flow prediction. IEEE Trans. on Knowledge and Data Engineering, 2023, 35(11): 11246–11258.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

李云,高雅,姚枝秀,夏士超,吴广富.面向数据稀缺场景的智能交通流量预测.软件学报,2025,36(8):3787-3801

复制
分享
文章指标
  • 点击次数:373
  • 下载次数: 696
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2023-07-12
  • 最后修改日期:2024-03-15
  • 在线发布日期: 2024-11-18
文章二维码
您是第19984686位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京市海淀区中关村南四街4号,邮政编码:100190
电话:010-62562563 传真:010-62562533 Email:jos@iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号