-
摘要: 为寻求提升交通预测时空计算任务性能的有效途径，探索图神经网络技术在交通预测中的应用前景和挑战，回顾了交通预测方法的发展，总结了模型驱动方法、统计模型、传统机器学习方法和深度学习方法的优势和局限性；阐述了图网络和交通网络的适配性，归纳了图的构造方法；将应用于交通预测的数据进行了分类，分析了不同交通预测任务的共性与差异；归纳了常用于交通预测任务的图神经网络模型，包括卷积图神经网络、图注意网络、图自编码器和图时空网络，分析了图神经网络模型应用于交通预测时主要考虑的因素和时空模块；对比了多种交通速度预测方法的性能，分析了图神经网络框架中不同组件对其预测性能的影响；从数据多源性、应用多样性、多模式、动态性、模型可解释性、不确定性和小样本学习等多个角度探讨了基于图神经网络的交通预测面临的挑战和机遇，并针对图神经网络发展趋势提出了相关建议。研究结果表明：与只考虑时间相关性的基准模型相比，基于图神经网络的预测方法性能显著提升；多模式时间相关性、时空注意力机制、边特征、外部数据均会显著影响预测性能；图神经网络为建模交通网络复杂动态的时空相关性提供了有力手段，目前针对交通状态预测问题开发了多样化的模型；未来研究可重点从设计高效的动态时空模块集成架构、设计有效整合外部数据的模块、拓展多样化的应用任务、实现多模式交通同步预测、开发高效可靠和易于解释的模型等方面进行突破，实现预测准确性和效率均衡提升，以期发展更高阶段的智慧交通。Abstract: To seek effective ways to improve the performance of spatial-temporal computing tasks in traffic prediction, and explore the prospects and challenges of applying the graph neural network technology in traffic prediction, the development of traffic prediction methods was reviewed. The advantages and limitations of model-driven methods, statistical models, traditional machine learning methods, and deep learning methods were summarized. The compatibility between graph networks and traffic networks was explained. The methods for constructing graphs were summarized. The data used for traffic prediction were classified. The commonalities and differences between different traffic prediction tasks were analyzed. The graph neural network models commonly used for traffic prediction tasks were concluded, including the convolutional graph neural network, graph attention network, graph autoencoder, and graph spatial-temporal network. The main factors and spatial-temporal modules considered when the graph neural network model was applied to traffic prediction were analyzed. The performance of various traffic speed prediction methods was compared. The impacts of different components of the graph neural network framework on the prediction performance were analyzed. The challenges and opportunities faced by traffic prediction based on the graph neural network were discussed from multiple perspectives such as the data multi-sourcing, application diversity, multimodality, dynamicity, model interpretability, uncertainty, and small sample learning. Relevant suggestions for the development trend of graph neural network were proposed. Research results show that compared with the benchmark model that only considers the time correlation, the performance of the prediction method based on the graph neural network improves significantly. The multi-mode time correlation, spatial-temporal attention mechanism, edge features, and external data can all significantly affect the prediction performance. The graph neural network provides a powerful means for modeling the spatial-temporal correlation with complex dynamicity of traffic networks. Currently, diversified models have been developed for the traffic state prediction problems. Future research can focus on developing efficient and dynamic spatial-temporal module integration architectures, designing modules that effectively integrate external data, expanding diversified application tasks, realizing the multi-mode traffic synchronous prediction, and developing efficient, reliable, and easy-to-explain models to achieve a balanced improvement in prediction accuracy and efficiency, so as to develop higher-level intelligent traffic.
-
图 2 基于图的交通预测时空关系
Figure 2. Spatial-temporal relationship of graph-based traffic predictions
图 5 交通流量密度与速度关系及预测结果
Figure 5. Relationship between traffic flow density and speed and prediction results
图 8 各模型在METR-LA数据集下预测准确性对比
Figure 8. Comparison of prediction accuracies of various models under METR-LA dataset
图 9 各模型在PeMS-BAY数据集下预测准确性对比
Figure 9. Comparison of prediction accuracies of various models under PeMS-BAY dataset
图 11 基于拓扑结构和动态出行模式的超图构建
Figure 11. Hypergraph constructions based on topology and dynamic travel modes
表 1 交通预测方法对比
Table 1. Comparison of traffic prediction methods
类型 方法 优势 不足 模型驱动 交通流模型 能够表征单个交通个体的运动特征或者整体交通流的运行规律，可解释性强 模型泛化能力弱，面对不同场景需重新建模和标定参数 元胞自动机模型 在空间和时间上离散，能够描述真实的交通行为 元胞的形状和排列过于规整，且难以整合系统外部因素的影响 数据驱动 统计方法 坚实的数学理论基础 平稳性假设导致准确性降低，非线性关系捕获能力不足 传统机器学习方法 充分捕获非线性关系 受限于浅层架构和人工特征选择方式 时序的深度学习方法 深入挖掘交通数据的时变特征 忽略了交通网络拓扑连接在内的复杂空间关联 卷积神经网络+ 时序深度学习方法 同时捕获时间和空间相关性 交通网络的连接关系不是二维网格，卷积神经网络捕获的空间相关性受限 GNN 与交通数据结构具有高度适配性，能够捕获多种空间相关性 可解释性和训练效率需进一步提升 
下载: 导出CSV
表 2 图神经网络对比
Table 2. Comparison of graph neural networks
方法 类型 优点 缺点 卷积图网络 基于空域 灵活性强，可扩展性强 缺少理论支撑 基于频域 理论基础扎实 结构固定，不适用于有向图 图注意网络 自适应地关注重要邻居 只关注节点的局部结构 图自编码器 重塑图保留动态随机特性 为学习到有效数据分布形式，需要对损失函数进行处理 图时空网络 基于RNN 应用广泛，性能良好 迭代耗时，门控机制复杂 基于CNN 并行计算，稳定梯度 缺乏对长序列的记忆
下载: 导出CSV
表 3 基于图神经网络的交通预测模型总结
Table 3. Summary of traffic prediction models based on graph neural networks
表 4 公开数据集基本信息
Table 4. Basic information about public datasets
类型 数据集 节点数(传感器数) 边数 统计粒度/ min 总时间步 传感器 METR-LA 207 1 515 5 34 272 PeMS-BAY 325 2 369 5 52 116 PeMSD3 358 547 5 26 208 PeMSD4 307(3 848) 340 5 16 992 PeMSD7 883 866 5 28 224 PeMSD7(M) 228 5 12 672 PeMSD8 170(1 979) 295 5 17 856 出租车轨迹 SZ-taxi 156 532 15 2 976 地铁刷卡 SHMETRO 288 958 15 6 072 HZMETRO 80 248 15 1 650 BEIJING-subway 276 630 10/15/ 30 2 700/ 1 800/900
下载: 导出CSV
表 5 TT、STGNN和Dual Graph模型对比
Table 5. Comparison of TT, STGNN and Dual Graph models
模型 时间模块 空间模块 注意力机制 时间片段 TT 变压器 DCNN或GCN 时间维度 连续近期、日周期、周周期 STGNN GRU、变压器 S-GNN 时空维度 连续近期 Dual Graph 编码器-解码器架构 空域图卷积网络 连续近期
下载: 导出CSV
-
[1] 刘静, 关伟. 交通流预测方法综述[J]. 公路交通科技, 2004, 21(3): 82-85. doi: 10.3969/j.issn.1002-0268.2004.03.022LIU Jing, GUAN Wei. A summary of traffic flow forecasting methods[J]. Journal of Highway and Transportation Research and Development, 2004, 21(3): 82-85. (in Chinese) doi: 10.3969/j.issn.1002-0268.2004.03.022 [2] VLAHOGIANNI E I, KARLAFTIS M G, GOLIAS J C. Short-term traffic forecasting: where we are and where we're going[J]. Transportation Research Part C: Emerging Technologies, 2014, 43: 3-19. doi: 10.1016/j.trc.2014.01.005 [3] LANA I, DEL SER J, VELEZ M, et al. Road traffic forecasting: recent advances and new challenges[J]. IEEE Intelligent Transportation Systems Magazine, 2018, 10(2): 93-109. doi: 10.1109/MITS.2018.2806634 [4] YIN Xue-yan, WU Gen-ze, WEI Jin-ze, et al. Deep learning on traffic prediction: methods, analysis and future directions[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(6): 4927-4943. doi: 10.1109/TITS.2021.3054840 [5] LI Fu-xian, FENG Jie, YAN Huan, et al. Dynamic graph convolutional recurrent network for traffic prediction: benchmark and solution[J]. ArXiv Preprint, 2021, DOI: arXiv:2104.14917 .[6] YE Jie-xia, ZHAO Juan-juan, YE Ke-jiang, et al. How to build a graph-based deep learning architecture in traffic domain: a survey[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(5): 3904-3924. doi: 10.1109/TITS.2020.3043250 [7] BUI K H N, CHO J, YI H. Spatial-temporal graph neural network for traffic forecasting: an overview and open research issues[J]. Applied Intelligence, 2022, 52(3): 2763-2774. doi: 10.1007/s10489-021-02587-w [8] JIANG Wei-wei, LUO Jia-yun. Graph neural network for traffic forecasting: a survey[J]. Expert Systems with Applications, 2022, 207: 117921. doi: 10.1016/j.eswa.2022.117921 [9] DHAMANIYA A, CHANDRA S. Speed prediction models for urban arterials under mixed traffic conditions[J]. Procedia—Social and Behavioral Sciences, 2013, 104: 342-351. doi: 10.1016/j.sbspro.2013.11.127 [10] LARTEY J D. Predicting traffic congestion: a queuing perspective[J]. Open Journal of Modelling and Simulation, 2014, 2(2): 57-66. doi: 10.4236/ojmsi.2014.22008 [11] LI Xiao-peng, WANG Xin, OUYANG Yan-feng. Prediction and field validation of traffic oscillation propagation under nonlinear car-following laws[J]. Transportation Research Part B: Methodological, 2012, 46(3): 409-423. doi: 10.1016/j.trb.2011.11.003 [12] KANOH H, FURUKAWA T, TSUKAHARA S, et al. Short-term traffic prediction using fuzzy c-means and cellular automata in a wide-area road network[C]//IEEE. Proceedings of 2005 IEEE Intelligent Transportation Systems. New York: IEEE, 2005: 381-385. [13] SKABARDONIS A, GEROLIMINIS N. Real-time estimation of travel times on signalized arterials[M]//Elsevier. Transportation and Traffic Theory. Amsterdam: Elsevier, 2005: 387-406. [14] OH S, BYON Y J, JANG K, et al. Short-term travel-time prediction on highway: a review on model-based approach[J]. KSCE Journal of Civil Engineering, 2018, 22(1): 298-310. doi: 10.1007/s12205-017-0535-8 [15] TANG Li-yang, ZHAO Yang, CABRERA J, et al. Forecasting short-term passenger flow: an empirical study on Shenzhen metro[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(10): 3613-3622. doi: 10.1109/TITS.2018.2879497 [16] 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. doi: 10.1061/(ASCE)0733-947X(1995)121:3(249) [17] OKUTANI I, STEPHANEDES Y J. Dynamic prediction of traffic volume through Kalman filtering theory[J]. Transportation Research Part B: Methodological, 1984, 18(1): 1-11. doi: 10.1016/0191-2615(84)90002-X [18] MCFADDEN J, YANG W T, DURRANS S R. Application of artificial neural networks to predict speeds on two-lane rural highways[J]. Transportation Research Record, 2001, 1751(1): 9-17. doi: 10.3141/1751-02 [19] LING Xian-yao, FENG Xin-xin, CHEN Zhong-hui, et al. Short-term traffic flow prediction with optimized multi-kernel support vector machine[C]//IEEE. 2017 IEEE Congress on Evolutionary Computation (CEC). New York: IEEE, 2017: 294-300. [20] 张晓利, 贺国光, 陆化普. 基于K-邻域非参数回归短时交通流预测方法[J]. 系统工程学报, 2009, 24(2): 178-183. https://www.cnki.com.cn/Article/CJFDTOTAL-XTGC200902009.htmZHANG Xiao-li, HE Guo-guang, LU Hua-pu. Short-term traffic flow forecasting based on K-nearest neighbors non-parametric regression[J]. Journal of Systems Engineering, 2009, 24(2): 178-183. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XTGC200902009.htm [21] VANAJAKSHI L, RILETT L R. A comparison of the performance of artificial neural networks and support vector machines for the prediction of traffic speed[C]//IEEE. IEEE Intelligent Vehicles Symposium, 2004. New York: IEEE, 2004: 194-199. [22] TAN Hua-chun, XUAN Xuan, WU Yuan-kai, et al. A comparison of traffic flow prediction methods based on DBN[C]// ASCE. 16th COTA International Conference of Transportation Professionals. Reston: ASCE, 2016: 273-283. [23] NGUYEN T, NGUYEN G, NGUYEN B M. EO-CNN: an enhanced CNN model trained by equilibrium optimization for traffic transportation prediction[J]. Procedia Computer Science, 2020, 176: 800-809. doi: 10.1016/j.procs.2020.09.075 [24] MA Xiao-lei, TAO Zhi-min, WANG Yin-hai, et al. Long short-term memory neural network for traffic speed prediction using remote microwave sensor data[J]. Transportation Research Part C: Emerging Technologies, 2015, 54: 187-197. doi: 10.1016/j.trc.2015.03.014 [25] SHAO Hong-xin, SOONG B H. Traffic flow prediction with long short-term memory networks (LSTMs)[C]//IEEE. 2016 IEEE Region 10 Conference (TENCON). New York: IEEE, 2016: 2986-2989. [26] DU Sheng-dong, LI Tian-rui, GONG Xun, et al. Traffic flow forecasting based on hybrid deep learning framework[C]// IEEE. 2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE). New York: IEEE, 2017: 17505846. [27] ZHANG Jun-bo, ZHENG Yu, QI De-kang. Deep spatio-temporal residual networks for citywide crowd flows prediction[C]//ACM. Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. New York: ACM, 2017: 1655-1661. [28] YANG Dan, CHEN Kai-rui, YANG Meng-ning, et al. Urban rail transit passenger flow forecast based on LSTM with enhanced long-term features[J]. IET Intelligent Transport Systems, 2019, 13(10): 1475-1482. doi: 10.1049/iet-its.2018.5511 [29] CAO Miao-miao, LI V O K, CHAN V W S. A CNN-LSTM model for traffic speed prediction[C]//IEEE. 2020 IEEE 91st Vehicular Technology Conference (VTC2020-Spring). New York: IEEE, 2020: 1-5. [30] ZOU Zhe-ne, PENG Hao, LIU Lin, et al. Deep convolutional mesh RNN for urban traffic passenger flows prediction[C]// IEEE. 2018 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI). New York: IEEE, 2018: 1305-1310. [31] SCARSELLI F, GORI M, TSOI A C, et al. The graph neural network model[J]. IEEE Transactions on Neural Networks, 2009, 20(1): 61-80. doi: 10.1109/TNN.2008.2005605 [32] ZHANG Tian-pu, DING Wei-long, CHEN Tao, et al. A graph convolutional method for traffic flow prediction in highway network[J]. Wireless Communications and Mobile Computing, 2021, 2021: 1997212. [33] HAN Yong, PENG Tong-xin, WANG Cheng, et al. A hybrid GLM model for predicting citywide spatio-temporal metro passenger flow[J]. ISPRS International Journal of Geo-Information, 2021, 10(4): 222. doi: 10.3390/ijgi10040222 [34] FENG Si-yuan, KE Jin-tao, YANG Hai, et al. A multi-task matrix factorized graph neural network for co-prediction of zone-based and OD-based ride-hailing demand[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(6): 5704-5716. doi: 10.1109/TITS.2021.3056415 [35] SONG Chao, LIN You-fang, GUO Sheng-nan, et al. Spatial-temporal synchronous graph convolutional networks: a new framework for spatial-temporal network data forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(1): 914-921. doi: 10.1609/aaai.v34i01.5438 [36] MA Xiao-lei, DAI Zhuang, HE Zheng-bing, et al. Learning traffic as images: a deep convolutional neural network for large-scale transportation network speed prediction[J]. Sensors, 2017, 17(4): 818. doi: 10.3390/s17040818 [37] XUE Gang, LIU Shi-feng, REN Long, et al. Forecasting the subway passenger flow under event occurrences with multivariate disturbances[J]. Expert Systems with Applications, 2022, 188: 116057. doi: 10.1016/j.eswa.2021.116057 [38] GUO Kan, HU Yong-li, QIAN Zhen, et al. Optimized graph convolution recurrent neural network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(2): 1138-1149. doi: 10.1109/TITS.2019.2963722 [39] GUO Sheng-nan, LIN You-fang, FENG Ning, 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. doi: 10.1609/aaai.v33i01.3301922 [40] GENG Xu, LI Ya-guang, WANG Le-ye, et al. Spatiotemporal multi-graph convolution network for ride-hailing demand forecasting[C]//ACM. Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence and Thirty-First Innovative Applications of Artificial Intelligence Conference and Ninth AAAI Symposium on Educational Advances in Artificial Intelligence. New York: ACM, 2019: 3656-3663. [41] GAO A, ZHENG Lin-jiang, WANG Zi-xu, et al. Attention based short-term metro passenger flow prediction[C]//Springer. International Conference on Knowledge Science, Engineering and Management. Berlin: Springer, 2021: 598-609. [42] 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 Xiao-liang, ZHENG Chuan-pan, 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. (in Chinese) [43] 曾筠程, 邵敏华, 孙立军, 等. 基于有向图卷积神经网络的交通预测与拥堵管控[J]. 中国公路学报, 2021, 34(12): 239-248. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202112018.htmZENG Yun-cheng, SHAO Min-hua, SUN Li-jun, et al. Traffic prediction and congestion control based on directed graph convolution convolutional neural network[J]. China Journal of Highway and Transport, 2021, 34(12): 239-248. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202112018.htm [44] LEE K, RHEE W. DDP-GCN: multi-graph convolutional network for spatiotemporal traffic forecasting[J]. Transportation Research Part C: Emerging Technologies, 2022, 134: 103466. doi: 10.1016/j.trc.2021.103466 [45] ZHAO Ling, SONG Yu-jiao, ZHANG Chao, et al. T-GCN: a temporal graph convolutional network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(9): 3848-3858. doi: 10.1109/TITS.2019.2935152 [46] YU Bing, YIN Hao-teng, ZHU Zhan-xing. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting[C]//ACM. Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence. New York: ACM, 2018: 3634-3640. [47] LI Ya-guang, YU R, SHAHABI C, et al. Diffusion convolutional recurrent neural network: data-driven traffic forecasting[J]. ArXiv Preprint, 2018, DOI: arXiv:1707.01926 .[48] XU Ming-xing, DAI Wen-rui, LIU Chun-miao, et al. Spatial-temporal transformer networks for traffic flow forecasting[J]. ArXiv Preprint, 2018, DOI: arXiv:2001.02908 .[49] LYU Ming-qi, HONG Zhao-xiong, CHEN Ling, et al. Temporal multi-graph convolutional network for traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(6): 3337-3348. doi: 10.1109/TITS.2020.2983763 [50] XIA Tong, LIN Jun-jie, LI Yong, et al. 3DGCN: 3-dimensional dynamic graph convolutional network for citywide crowd flow prediction[J]. ACM Transactions on Knowledge Discovery from Data, 2021, 15(6): 1-21. [51] BAI Lei, YAO Li-na, KANHERE S S, et al. STG2Seq: spatial-temporal graph to sequence model for multi-step passenger demand forecasting[C]//International Joint Conferences on Artificial Intelligence Organization. Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. Macao: International Joint Conferences on Artificial Intelligence Organization, 2019: 1981-1987. [52] YAO Hua-xiu, WU Fei, KE Jin-tao, et al. Deep multi-view spatial-temporal network for taxi demand prediction[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2018, 32(1): 2588-2595. [53] YU Bing, LI Meng-zhang, ZHANG Ji-yong, et al. 3D graph convolutional networks with temporal graphs: a spatial information free framework for traffic forecasting[J]. ArXiv Preprint, 2019, DOI: arXiv:1903.00919 .[54] BAI Lei, YAO Li-na, LI Can, et al. Adaptive graph convolutional recurrent network for traffic forecasting[J]. Advances in Neural Information Processing Systems, 2020, 33: 17804-17815. [55] HAN Liang-zhe, DU Bo-wen, SUN Lei-lei, et al. Dynamic and multi-faceted spatio-temporal deep learning for traffic speed forecasting[C]//ACM. Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2021: 547-555. [56] ROY A, ROY K K, ALI A A, et al. Unified spatio-temporal modeling for traffic forecasting using graph neural network[C]//IEEE. 2021 International Joint Conference on Neural Networks (IJCNN). New York: IEEE, 2021: 1-8. [57] WANG Xiao-yang, MA Yao, WANG Yi-qi, et al. Traffic flow prediction via spatial temporal graph neural network[C]// ACM. Proceedings of the Web Conference 2020. New York: ACM, 2020: 1082-1092. [58] YU J J Q, GU Jia-tao. Real-time traffic speed estimation with graph convolutional generative autoencoder[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(10): 3940-3951. doi: 10.1109/TITS.2019.2910560 [59] ZHANG Qi, JIN Qi-zhao, CHANG Jian-long, et al. Kernel-weighted graph convolutional network: a deep learning approach for traffic forecasting[C]//IEEE. 2018 24th International Conference on Pattern Recognition (ICPR). New York: IEEE, 2018: 1018-1023. [60] WANG Qiang, XU Chen, ZHANG Wen-qi, et al. GraphTTE: travel time estimation based on attention-spatiotemporal graphs[J]. IEEE Signal Processing Letters, 2021, 28: 239-243. doi: 10.1109/LSP.2020.3048849 [61] ZHU Jia-wei, WANG Qiong-jie, TAO Chao, et al. AST-GCN: attribute-augmented spatiotemporal graph convolutional network for traffic forecasting[J]. IEEE Access, 2021, 9: 35973-35983. doi: 10.1109/ACCESS.2021.3062114 [62] WANG Yuan-dong, YIN Hong-zhi, CHEN Hong-xu, et al. Origin-destination matrix prediction via graph convolution: a new perspective of passenger demand modeling[C]//ACM. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2019: 1227-1235. [63] BAI Lei, YAO Li-na, WANG Xian-zhi, et al. Deep spatial-temporal sequence modeling for multi-step passenger demand prediction[J]. Future Generation Computer Systems, 2021, 121: 25-34. doi: 10.1016/j.future.2021.03.003 [64] WANG Jing-cheng, ZHANG Yong, WEI Yu, et al. Metro passenger flow prediction via dynamic hypergraph convolution networks[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(12): 7891-7903. doi: 10.1109/TITS.2021.3072743 [65] CHEN Wei, LI Zong-ping, LIU Can, et al. A deep learning model with conv-LSTM networks for subway passenger congestion delay prediction[J]. Journal of Advanced Transportation, 2021, 2021: 1-10. [66] LIU Ling-bo, CHEN Jing-wen, WU He-feng, et al. Physical-virtual collaboration modeling for intra- and inter-station metro ridership prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 23(4): 3377-3391. [67] ESSIEN A, PETROUNIAS I, SAMPAIO P, et al. A deep-learning model for urban traffic flow prediction with traffic events mined from twitter[J]. World Wide Web, 2021, 24(4): 1345-1368. doi: 10.1007/s11280-020-00800-3 [68] SHI Liu-shuai, WANG Le, LONG Cheng-jiang, et al. SGCN: sparse graph convolution network for pedestrian trajectory prediction[C]//IEEE. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2021: 8990-8999. [69] ZHANG Jun-bo, ZHENG Yu, SUN Jun-kai, et al. Flow prediction in spatio-temporal networks based on multitask deep learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 32(3): 468-478. doi: 10.1109/TKDE.2019.2891537 [70] 李松江, 祝绍凇, 杨华民, 等. 基于时空相关性多任务神经网络的交通预测[J]. 计算机应用与软件, 2021, 38(9): 286-292. https://www.cnki.com.cn/Article/CJFDTOTAL-JYRJ202109047.htmLI Song-jiang, ZHU Shao-song, YANG Hua-min, et al. Traffic prediction based on spatiotemporal correlation multitask neural network[J]. Computer Applications and Software, 2021, 38(9): 286-292. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JYRJ202109047.htm [71] YIN Xue-yan, WU Gen-ze, WEI Jin-ze, et al. Multi-stage attention spatial-temporal graph networks for traffic prediction[J]. Neurocomputing, 2021, 428: 42-53. doi: 10.1016/j.neucom.2020.11.038 [72] CAI Ling, JANOWICZ K, MAI Geng-chen, et al. Traffic transformer: capturing the continuity and periodicity of time series for traffic forecasting[J]. Transactions in GIS, 2020, 24(3): 736-755. doi: 10.1111/tgis.12644 [73] FANG Shen, ZHANG Qi, MENG Gao-feng, et al. GSTNet: global spatial-temporal network for traffic flow prediction[C]//International Joint Conferences on Artificial Intelligence Organization. Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. Macao: International Joint Conferences on Artificial Intelligence Organization, 2019: 2286-2293. [74] YU B, LEE Y, SOHN K. Forecasting road traffic speeds by considering area-wide spatio-temporal dependencies based on a graph convolutional neural network (GCN)[J]. Transportation Research Part C: Emerging Technologies, 2020, 114: 189-204. doi: 10.1016/j.trc.2020.02.013 [75] WRIGHT M A, EHLERS S F G, HOROWITZ R. Neural-attention-based deep learning architectures for modeling traffic dynamics on lane graphs[C]//IEEE. 2019 IEEE Intelligent Transportation Systems Conference (ITSC). New York: IEEE, 2019: 3898-3905. [76] CHAI Di, WANG Le-ye, YANG Qiang. Bike flow prediction with multi-graph convolutional networks[C]//ACM. Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. New York: ACM, 2018: 397-400. [77] KE Jin-tao, QIN Xiao-ran, YANG Hai, et al. Predicting origin-destination ride-sourcing demand with a spatio-temporal encoder-decoder residual multi-graph convolutional network[J]. Transportation Research Part C: Emerging Technologies, 2021, 122: 102858. doi: 10.1016/j.trc.2020.102858 [78] MOHANTY S, POZDNUKHOV A, CASSIDY M. Region-wide congestion prediction and control using deep learning[J]. Transportation Research Part C: Emerging Technologies, 2020, 116: 102624. doi: 10.1016/j.trc.2020.102624 [79] SUN Ya-sheng, HE Tao, HU Jie, et al. Socially-aware graph convolutional network for human trajectory prediction[C]// IEEE. 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC). New York: IEEE, 2019: 325-333. [80] JEON H, CHOI J, KUM D. SCALE-net: scalable vehicle trajectory prediction network under random number of interacting vehicles via edge-enhanced graph convolutional neural network[C]//IEEE. 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). New York: IEEE, 2020: 2095-2102. [81] WU Zong-han, PAN Shi-rui, CHEN Feng-wen, et al. A comprehensive survey on graph neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(1): 4-24. doi: 10.1109/TNNLS.2020.2978386 [82] ZHOU Jie, CUI Gang-qu, HU Sheng-ding, et al. Graph neural networks: a review of methods and applications[J]. AI Open, 2020, 1: 57-81. doi: 10.1016/j.aiopen.2021.01.001 [83] ZHANG Zi-wei, CUI Peng, ZHU Wen-wu. Deep learning on graphs: a survey[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(1): 249-270. doi: 10.1109/TKDE.2020.2981333 [84] GILMER J, SCHOENHOLZ S S, RILEY P F, et al. Neural message passing for quantum chemistry[J]. ArXiv Preprint, 2017, DOI: arXiv:1704.01212 .[85] WEI Long, YU Zheng-xu, JIN Zhong-ming, et al. Dual graph for traffic forecasting[J]. IEEE Access, 2019(99): 1. [86] HAMILTON W L, YING R, LESKOVEC J. Inductive representation learning on large graphs[C]//ACM. Proceedings of the 31st International Conference on Neural Information Processing Systems. New York: ACM, 2017: 1025-1035. [87] ATWOOD J, TOWSLEY D. Diffusion-convolutional neural networks[C]//NIPS. Proceedings of the 30th International Conference on Neural Information Processing Systems. New York: Curran Associates Inc., 2016: 2001-2009. [88] HAMMOND D K, VANDERGHEYNST P, GRIBONVAL R. Wavelets on graphs via spectral graph theory[J]. Applied and Computational Harmonic Analysis, 2011, 30(2): 129-150. doi: 10.1016/j.acha.2010.04.005 [89] DEFFERRARD M, BRESSON X, VANDERGHEYNST P. Convolutional neural networks on graphs with fast localized spectral riltering[C]//NIPS. Proceedings of the 30th International Conference on Neural Information Processing Systems. New York: Curran Associates Inc., 2016: 3844-3852. [90] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[J]. ArXiv Preprint, 2017, DOI: arXiv:1609.02907 .[91] VELI AČG KOVI AC'G P, CUCURULL G, CASANOVA A, et al. Graph attention networks[J]. ArXiv Preprint, 2017, DOI: arXiv:1710.10903 .[92] ZHANG Jia-ni, SHI Xing-jia, XIE Jun-yuan, et al. GaAN: gated attention networks for learning on large and spatiotemporal graphs[J]. ArXiv Preprint, 2018, DOI: arXiv:1803.07294 .[93] KIPF T N, WELLING M. Variational graph auto-encoders[J]. ArXiv Preprint, 2016, DOI: arXiv:1611.07308 .[94] ZHAO Fei-fei, WANG Wei-ping, SUN Hui-jun, et al. Station-level short-term demand forecast of carsharing system via station-embedding-based hybrid neural network[J]. Transportmetrica B: Transport Dynamics, 2022, 10(1): 1-19. doi: 10.1080/21680566.2021.1951885 [95] ZHOU Fan, YANG Qing, ZHONG Ting, et al. Variational graph neural networks for road traffic prediction in intelligent transportation systems[J]. IEEE Transactions on Industrial Informatics, 2021, 17(4): 2802-2812. doi: 10.1109/TII.2020.3009280 [96] LIPTON Z C, BERKOWITZ J, ELKAN C. A critical review of recurrent neural networks for sequence learning[J]. ArXiv Preprint, 2015, DOI: arXiv:1506.00019 .[97] 倪庆剑, 彭文强, 张志政, 等. 基于信息增强传输的时空图神经网络交通流预测[J]. 计算机研究与发展, 2022, 59(2): 282-293.NI Qing-jian, PENG Wen-qiang, ZHANG Zhi-zheng, et al. Traffic flow prediction of spatiotemporal graph neural network based on information enhancement transmission[J]. Journal of Computer Research and Development, 2022, 59(2): 282-293. (in Chinese) [98] ZHANG Shao-kun, GUO Yao, ZHAO Pei-ze, et al. A graph-based temporal attention framework for multi-sensor traffic flow forecasting[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(7): 7743-7758. doi: 10.1109/TITS.2021.3072118 [99] GUO Kan, HU Yong-li, SUN Yan-fei, et al. Hierarchical graph convolution networks for traffic forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(1): 151-159. doi: 10.1609/aaai.v35i1.16088 [100] LI Guo-peng, KNOOP V L, VAN LINT H. Multistep traffic forecasting by dynamic graph convolution: interpretations of real-time spatial correlations[J]. Transportation Research Part C: Emerging Technologies, 2021, 128: 103185. doi: 10.1016/j.trc.2021.103185 [101] TIAN Chen-yu, CHAN W K. Spatial-temporal attention wavenet: a deep learning framework for traffic prediction considering spatial-temporal dependencies[J]. IET Intelligent Transport Systems, 2021, 15(4): 549-561. doi: 10.1049/itr2.12044 [102] PARK C, LEE C, BAHNG H, et al. ST-GRAT: a novel spatio-temporal graph attention networks for accurately forecasting dynamically changing road speed[C]//ACM. Proceedings of the 29th ACM International Conference on Information and Knowledge Management. New York: ACM, 2020: 1215-1224. [103] GENG Xu, LI Ya-guang, WANG Le-ye, et al. Spatiotemporal multi-graph convolution network for ride-hailing demand forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(1): 3656-3663. doi: 10.1609/aaai.v33i01.33013656 [104] YE Jun-cheng, SUN Lei-lei, DU Bo-wen, et al. Coupled layer-wise graph convolution for transportation demand prediction[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(5): 4617-4625. doi: 10.1609/aaai.v35i5.16591 [105] ZI Wen-jie, XIONG Wei, CHEN Hao, et al. TAGCN: station-level demand prediction for bike-sharing system via a temporal attention graph convolution network[J]. Information Sciences, 2021, 561: 274-285. doi: 10.1016/j.ins.2021.01.065 [106] WANG Xi, CHAI Yi-bo, LI Hui, et al. Graph convolutional network-based model for incident-related congestion prediction: a case study of Shanghai Expressways[J]. ACM Transactions on Management Information Systems, 2021, 12(3): 1-22. [107] ZHANG Jin-lei, CHEN Feng, CUI Zhi-yong, et al. Deep learning architecture for short-term passenger flow forecasting in urban rail transit[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(11): 7004-7014. doi: 10.1109/TITS.2020.3000761 [108] HUA Xin, LIU Wei. Spatial-temporal network data-driven multi-layer traffic knowledge graph reconstruction for dynamic prediction[C]//IEEE. 2022 4th International Conference on Robotics and Computer Vision (ICRCV). New York: IEEE, 2022: 20-24. [109] ZHU Jia-wei, HAN Xing, DENG Han-han, et al. KST-GCN: a knowledge-driven spatial-temporal graph convolutional network for traffic forecasting[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(9): 15055-15065. doi: 10.1109/TITS.2021.3136287 [110] ZHONG Ting, XU Zhe-yang, ZHOU Fan. Probabilistic graph neural networks for traffic signal control[C]//IEEE. ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). New York: IEEE, 2021: 4085-4089. [111] ZHANG Heng-yuan, ZHAO Su-yao, LIU Rui-heng, et al. Automatic traffic anomaly detection on the road network with spatial-temporal graph neural network representation learning[J]. Wireless Communications and Mobile Computing, 2022, 2022: 4222827. [112] JIN Ke-fan, WANG Hong-ye, LIU Chang-xing, et al. Graph neural network based relation learning for abnormal perception information detection in self-driving scenarios[C]//IEEE. 2022 International Conference on Robotics and Automation (ICRA). New York: IEEE, 2022: 8943-8949. [113] LIANG Yue-bing, HUANG Guan, ZHAO Zhan. Joint demand prediction for multimodal systems: a multi-task multi-relational spatiotemporal graph neural network approach[J]. Transportation Research Part C: Emerging Technologies, 2022, 140: 103731. doi: 10.1016/j.trc.2022.103731 [114] TYGESEN M N, PEREIRA F C, RODRIGUES F. Unboxing the graph: towards interpretable graph neural networks for transport prediction through neural relational inference[J]. Transportation Research Part C: Emerging Technologies, 2023, 146: 103946. doi: 10.1016/j.trc.2022.103946 [115] YUAN Hao, YU Hai-yang, GUI Shu-rui, et al. Explainability in graph neural networks: a taxonomic survey[J]. ArXiv Preprint, 2020, DOI: arXiv:2012.15445 .[116] ZHANG Yuan, CHENG Qi-xiu, LIU Yang, et al. Full-scale spatio-temporal traffic flow estimation for city-wide networks: a transfer learning based approach[J]. Transportmetrica B: Transport Dynamics, 2022, 11(1): 869-895. [117] HOQUE J M, ERHARDT G D, SCHMITT D, et al. Estimating the uncertainty of traffic forecasts from their historical accuracy[J]. Transportation Research Part A: Policy and Practice, 2021, 147: 339-349. doi: 10.1016/j.tra.2021.03.015 [118] GAL Y, GHAHRAMANI Z. Dropout as a Bayesian approximation: representing model uncertainty in deep learning[C]//ACM. International Conference on Machine Learning. New York: ACM, 2016: 1050-1059. [119] LI Ming-xi, TANG Yi-hong, MA Wei. Few-sample traffic prediction with graph networks using locale as relational inductive biases[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(2): 1894-1908. [120] QI Yu-xin, WU Jun, BASHIR A K, et al. Privacy-preserving cross-area traffic forecasting in ITS: a transferable spatial-temporal graph neural network approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(9): 1-14. doi: 10.1109/TITS.2022.3201451 -
点击查看大图
图(11) / 表(5)
计量
- 文章访问数: 1420
- HTML全文浏览量: 166
- PDF下载量: 390
- 被引次数: 0
