基于深度学习的短期交通流预测方法综述

崔建勋; 要甲; 赵泊媛

doi:10.19818/j.cnki.1671-1637.2024.02.003

基于深度学习的短期交通流预测方法综述

doi: 10.19818/j.cnki.1671-1637.2024.02.003

1.
哈尔滨工业大学交通科学与工程学院，黑龙江哈尔滨 150090
2.
大连理工大学经济管理学院，辽宁大连 116024

基金项目:

国家自然科学基金项目 72371048

黑龙江省自然科学基金项目 LH2021E074

详细信息

作者简介:
崔建勋(1982-)，男，黑龙江绥化人，哈尔滨工业大学副教授，从事智能交通与自动驾驶研究

中图分类号: U491.1
计量
- 文章访问数: 1272
- HTML全文浏览量: 117
- PDF下载量: 278
- 被引次数: 0
出版历程
- 收稿日期: 2023-10-13
- 网络出版日期: 2024-05-16
- 刊出日期: 2024-04-30

Review on short-term traffic flow prediction methods based on deep learning

1.
School of Transportation Science and Engineering, Harbin Institute of Technology, Harbin 150090, Heilongjiang, China
2.
School of Economics and Management, Dalian University of Technology, Dalian 116024, Liaoning, China

Funds:

National Natural Science Foundation of China 72371048

Natural Science Foundation of Heilongjiang Province LH2021E074

More Information

Author Bio:
CUI Jian-xun(1982-), male, associate professor, PhD, cuijianxun@hit.edu.cn

摘要

摘要: 针对基于深度学习的短期交通流预测问题，揭示了时空相关性建模本质，分析了建模过程中涉及的多尺度时空特性、异质性、动态性、非线性等特点，明确了基于深度学习进行短期交通流预测的核心挑战，阐述了短期交通流预测涉及的外部信息整合、多步预测与单步预测以及单体预测与集成预测等相关问题；按照网格化和拓扑化2种交通流数据组织方式，分别综述了当前最新的基于深度学习的短期交通流预测研究方向。研究结果表明：针对网格化交通流数据，当前研究主要包含了基于2D图像卷积神经网络、基于2D图像卷积神经网络与循环神经网络相结合、基于3D图像卷积神经网络3种预测建模方法；针对拓扑化交通流数据，当前研究主要包含了基于1D因果图像卷积与卷积图神经网络相结合、基于循环神经网络与卷积图神经网络相结合、基于自注意力与卷积图神经网络相结合、基于卷积图神经网络的时空同步学习4种预测建模方法；总体上，基于深度学习方法进行短期交通流预测相较于采用时间序列和经典机器学习方法获得了预测准确性上的极大提升；未来，针对物理理论、知识图谱与深度学习相结合，构建多时空数据挖掘大模型以及轻量化、可解释性、模型结构自动化搜索等维度的相关探索将成为重要研究方向。
- 智能交通 /
- 短期交通流预测 /
- 深度学习 /
- 卷积神经网络 /
- 图神经网络 /
- 循环神经网络
Abstract: For the short-term traffic flow prediction problem based on deep learning, the essence of spatiotemporal correlation modeling was revealed, the multi-scale spatiotemporal characteristics, heterogeneity, dynamics, nonlinearity, and other characteristics involved in the modeling process were analyzed, the core challenges were clarified, and the external information integration, multi-step prediction and single-step prediction, as well as individual prediction and integrated prediction were elaborated. The latest research directions were reviewed with two organization methods of traffic flow data: grid and topology. Research results indicate that for gridded traffic flow data, current research mainly includes three prediction modeling methods: 2D image convolutional neural network, 2D image convolutional neural network combined with recurrent neural network, and 3D image convolutional neural network. For topological traffic flow data, current research mainly includes four prediction modeling methods: 1D causal image convolution combined with convolutional graph neural network, recurrent neural network combined with convolutional graph neural network, self-attention combined with convolutional graph neural network, and spatiotemporal synchronous learning of convolutional graph neural network. Overall, the short-term traffic flow prediction based on deep learning methods significantly improves the prediction accuracy compared to time series method and classical machine learning method. In the future, the combination of physics theory, knowledge graphs, and deep learning, the construction of large-scale models for multi-temporal and spatial data mining, as well as the lightweight, interpretability, and automated model structure search, will become important research directions.
- intelligent transportation /
- short-term traffic flow prediction /
- deep learning /
- convolution neural network /
- graph neural network /
- recurrent neural network

HTML全文

图 1 基于深度学习的短期交通流预测一般框架

Figure 1. General framework of short-term traffic flow prediction based on deep learning

下载: 全尺寸图片幻灯片

图 2 多个历史时段交通流图像拼接集成

Figure 2. Concatenation aggregation of traffic flow images in multiple historical periods

下载: 全尺寸图片幻灯片

图 3 基于2D图像卷积与循环神经网络的时空相关性建模

Figure 3. Spatiotemporal correlation modeling based on 2D image convolution and recurrent neural network

下载: 全尺寸图片幻灯片

图 4 交通观测拓扑化网络示例

Figure 4. Example of traffic observation topological networks

下载: 全尺寸图片幻灯片

图 5 拓扑化道路交通时空观测数据示例

Figure 5. Example of topological road traffic spatiotemporal observation data

下载: 全尺寸图片幻灯片

图 6 基于1D因果图像卷积与卷积图神经网络的时空相关性建模

Figure 6. Spatiotemporal correlation modeling based on 1D causal image convolution and convolution graph neural network

下载: 全尺寸图片幻灯片

图 7 基于循环神经网络与卷积图神经网络的时空相关性建模

Figure 7. Spatiotemporal correlation modeling based on recurrent neural network and convolutional graph neural network

下载: 全尺寸图片幻灯片

图 8 交通流序列中的各种相关性

Figure 8. Various correlations in traffic flow sequence

下载: 全尺寸图片幻灯片

表 1 基于深度学习的短期交通流预测文献概况

Table 1. Overview of literatures on short-term traffic flow prediction based on deep learning

数据组织方式	时空相关性捕获方式	代表性文献	核心理念	优势	劣势
网格化数据	2D图像卷积神经网络	[46]~[48]	将时空网格化的交通流数据转化为多通道图像数据，从而单纯采用卷积神经网络进行时空相关性特征学习	将时空相关性特征学习转化为单纯的空间特征学习，方法相对简单、高效	由于将时间维度的信息转化为图像通道的信息，导致对时间相关性捕获欠佳
	2D图像卷积与循环神经网络相结合	[21]、[25]、[49]~[54]	采用图像卷积网络捕获空间相关性，采用循环神经网络捕获时间相关性	将复杂的时空相关性捕获分解为时间相关性和空间相关性的单独捕获，降低了问题复杂度，模型思路简洁、清晰	时空相关性的分离式建模导致无法捕获时空同步相关性，此外，循环神经网络的训练和推理效率较低
	3D图像卷积神经网络	[18]、[55]~[60]	将历史网格化交通流数据视为“视频流”型数据，采用可以在时空维度上进行滑动的3D卷积核进行时空相关性特征学习	能够一定程度上捕获到时空同步的相关性，且模型仅采用卷积计算，具有并行计算的优势	3D卷积相对于2D卷积的计算量明显上升，模型的训练和推理效率有所下降
拓扑化数据	1D因果图像卷积与卷积图神经网络相结合	[61]~[65]	采用1D因果卷积捕获时间相关性，采用拓扑图卷积捕获空间相关性	采用1D因果卷积捕获时间相关性，相较于循环神经网络在训练和推理效率上明显提升	时空相关性的捕获采用了时空分离的方式，无法捕获到同步的时空相关性
	循环神经网络与卷积图神经网络相结合	[66]~[68]	采用循环神经网络捕获时间相关性，采用卷积图神经网络捕获空间相关性	采用循环神经网络和卷积图神经网络分别进行时间和空间相关性捕获，理论较为成熟	时空相关性的捕获采用了时空分离的方式，无法捕获到同步的时空相关性，此外，循环神经网络训练和推理效率较低
	自注意力与卷积图神经网络	[69]	自注意力机制负责从时间的维度捕获相关性，卷积图神经网络负责从空间维度捕获相关性	注意力机制在时间维度相关性捕获上，由于并行本质，计算高效	无法捕获时空同步相关性，容易忽略时间顺序对未来预测的影响，需要时间位置编码补充
	基于卷积图神经网络的时空同步学习	[70]、[71]	将空间拓扑图拓展到时空拓扑图，从而统一采用拓扑图卷积捕获时空相关性	空间拓扑到时空拓扑变换，导致网络规模急剧增加，模型运算复杂度明显提升	能够捕获到时空同步的相关性

下载: 导出CSV

参考文献(81)

[1]	DU W D, ZHANG Q Y, CHEN Y P, et al. An urban short-term traffic flow prediction model based on wavelet neural network with improved whale optimization algorithm[J]. Sustainable Cities and Society, 2021, 69: 102858. doi: 10.1016/j.scs.2021.102858
[2]	史其信, 郑为中. 道路网短期交通流预测方法比较[J]. 交通运输工程学报, 2004, 4(4): 68-71, 83. http://transport.chd.edu.cn/article/id/200404017 SHI Qi-xin, ZHENG Wei-zhong. Short-term traffic flow prediction methods comparison of road networks[J]. Journal of Traffic and Transportation Engineering, 2004, 4(4): 68-71, 83. (in Chinese) http://transport.chd.edu.cn/article/id/200404017
[3]	王晓全, 邵春福, 尹超英, 等. 基于ARIMA-GARCH-M模型的短时交通流预测方法[J]. 北京交通大学学报, 2018, 42(4): 79-84. https://www.cnki.com.cn/Article/CJFDTOTAL-BFJT201804011.htm WANG Xiao-quan, SHAO Chun-fu, YIN Chao-ying, et al. Short term traffic flow forecasting method based on ARIMA- GARCH-M model[J]. Journal of Beijing Jiaotong University, 2018, 42(4): 79-84. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BFJT201804011.htm
[4]	于泉, 姚宗含. 交通流预测的马尔科夫粒子滤波方法研究[J]. 交通运输系统工程与信息, 2019, 19(2): 209-215. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201902030.htm YU Quan, YAO Zong-han. Markov particle filter traffic flow prediction model[J]. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(2): 209-215. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201902030.htm
[5]	白伟华, 张传斌, 张塽旖, 等. 基于异常值识别卡尔曼滤波器的短期交通流预测[J]. 计算机应用研究, 2021, 38(3): 817-821. https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ202103034.htm BAI Wei-hua, ZHANG Chuan-bin, ZHANG Shuang-yi, et al. O utlier-identified Kalman filter for short-term traffic flow forecasting[J]. Application Research of Computers, 2021, 38(3): 817-821. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ202103034.htm
[6]	杨高飞, 徐睿, 秦鸣, 等. 基于ARMA和卡尔曼滤波的短时交通预测[J]. 郑州大学学报(工学版), 2017, 38(2): 36-40. https://www.cnki.com.cn/Article/CJFDTOTAL-ZZGY201702009.htm YANG Gao-fei, XU Rui, QIN Ming, et al. Short-term traffic volume forecasting based on ARMA and Kalman filter[J]. Journal of Zhengzhou University (Engineering Science), 2017, 38(2): 36-40. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZZGY201702009.htm
[7]	王翔, 陈小鸿, 杨祥妹. 基于K最近邻算法的高速公路短时行程时间预测[J]. 中国公路学报, 2015, 28(1): 102-111. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201501017.htm WANG Xiang, CHEN Xiao-hong, YANG Xiang-mei. Short term prediction of expressway travel time based on K nearest neighbor algorithm[J]. China Journal of Highway and Transport, 2015, 28(1): 102-111. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201501017.htm
[8]	LUO X L, LI D Y, ZHANG S R. Traffic flow prediction during the holidays based on DFT and SVR[J]. Journal of Sensors, 2019, 2019(10): 6461450.
[9]	DONG C J, CUI A. Prediction models of short-term traffic flow based on neural network[J]. Advanced Materials Research, 2013, 671-674: 2908-2911. doi: 10.4028/www.scientific.net/AMR.671-674.2908
[10]	LIU Zhong-bo, YANG Zhao-sheng, GAO Peng. Research on the short-term traffic flow prediction method based on BP neural networks[C]//IEEE. 2012 World Automation Congress. New York: IEEE, 2012: 94214.
[11]	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
[12]	邵春福, 薛松, 董春娇, 等. 考虑时空相关性的网络交通流短期预测[J]. 北京交通大学学报, 2021, 45(4): 37-43. https://www.cnki.com.cn/Article/CJFDTOTAL-BFJT202104006.htm SHAO Chun-fu, XUE Song, DONG Chun-jiao, et al. Short-term traffic flow prediction considering temporal-spatial correlation on road network[J]. Journal of Beijing Jiaotong University, 2021, 45(4): 37-43. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BFJT202104006.htm
[13]	LIU Qing-chao, LIU Tao, CAI Ying-feng, et al. Explanatory prediction of traffic congestion propagation mode: a self-attention based approach[J]. Physica A: Statistical Mechanics and its Applications, 2021, 573: 125940. doi: 10.1016/j.physa.2021.125940
[14]	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. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202004008.htm CHEN Xi-qun, ZHOU Ling-xiao, CAO Zhen. 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. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202004008.htm
[15]	LYU Yi-sheng, DUAN Yan-jie, KANG Wen-wen, et al. Traffic flow prediction with big data: a deep learning approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(2): 865-873.
[16]	GE Liang, LI Si-yu, WANG Ya-qiang, et al. Global Spatial-temporal graph convolutional network for urban traffic speed prediction[J]. Applied Sciences, 2020, 10(4): 1509. doi: 10.3390/app10041509
[17]	LIN Hao-xing, JIA Wei-jia, SUN Yi-ping, et al. Spatial-temporal self-attention network for flow prediction[J]. arXiv, 2019, DOI: 10.48550/arXiv.1912.07663.
[18]	GUO Sheng-nan, LIN You-fang, LI Shi-jie, et al. Deep spatial-temporal 3D convolutional neural networks for traffic data forecasting[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(10): 3913-3926. doi: 10.1109/TITS.2019.2906365
[19]	FU Rui, ZHANG Zuo, LI Li. Using LSTM and GRU neural network methods for traffic flow prediction[C]//IEEE. 31st Youth Academic Annual Conference of Chinese Association of Automation. New York: IEEE, 2016: 324-328.
[20]	SHI Xing-jian, CHEN Zhuo-rong, WANG Hao, et al. Convolutional LSTM Network: a machine learning approach for precipitation nowcasting[C]//NIPS. 29th Annual Conference on Neural Information Processing Systems. San Diego: NIPS, 2015: 802-810.
[21]	WANG Y, WU H, ZHANG J, et al. PredRNN: a recurrent neural network for spatiotemporal predictive learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(2): 2208-2225. doi: 10.1109/TPAMI.2022.3165153
[22]	ZHANG Zheng-chao, LI Meng, LIN Xi, et al. Multistep speed prediction on traffic networks: a deep learning approach considering spatio-temporal dependencies[J]. Transportation Research Part C: Emerging Technologies, 2019, 105: 297-322. doi: 10.1016/j.trc.2019.05.039
[23]	ANDREOLETTI D, TROIA S, MUSUMECI F, et al. Network traffic prediction based on diffusion convolutional recurrent neural networks[C]//IEEE. 2019 INFOCOM IEEE Conference on Computer Communications Workshops. New York: IEEE, 2019: 246-251.
[24]	JIN Cang-hong, TAO Ruan, WU De-xing, et al. HetGAT: a heterogeneous graph attention network for freeway traffic speed prediction[J]. Journal of Ambient Intelligence and Humanized Computing, 2021, DOI: 10.1007/s12652-020-02807-0.
[25]	LIN Zhi-hui, LI Mao-mao, ZHENG Zhuo-bin, et al. Self-attention ConvLSTM for spatiotemporal prediction[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(7): 11531-11538. doi: 10.1609/aaai.v34i07.6819
[26]	CAI L, JANOWICZ K, MAI G C, 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
[27]	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. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200403022.htm LIU 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) https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200403022.htm
[28]	王进, 史其信. 神经网络模型在短期交通流预测领域应用综述[J]. 河南科技大学学报(自然科学版), 2005, 26(2): 22-26. https://www.cnki.com.cn/Article/CJFDTOTAL-LYGX200502008.htm WANG Jin, SHI Qi-xin. Review of application of neural network based models in short-term traffic flow forecasting[J]. Journal of Henan University of Science and Technology (Natural Science), 2005, 26(2): 22-26. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-LYGX200502008.htm
[29]	李振龙, 张利国, 钱海峰. 基于非参数回归的短时交通流预测研究综述[J]. 交通运输工程与信息学报, 2008, 6(4): 34-39. https://www.cnki.com.cn/Article/CJFDTOTAL-JTGC200804009.htm LI Zhen-long, ZHANG Li-guo, QIAN Hai-feng. Review of the short-term traffic flow forecasting based on the non-parametric regression[J]. Journal of Transportation Engineering and Information, 2008, 6(4): 34-39. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JTGC200804009.htm
[30]	郭敏, 肖翔, 蓝金辉. 道路交通流短时预测方法综述[J]. 自动化技术与应用, 2009, 28(6): 8-9, 16. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDHJ200906005.htm GUO Min, XIAO Xiang, LAN Jin-hui. A summary of the short-time traffic flow forecasting methods[J]. Techniques of Automation and Applications, 2009, 28(6): 8-9, 16. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZDHJ200906005.htm
[31]	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
[32]	CUI Zhi-yong, KE Rui-min, PU Zi-yuan, et al. Stacked bidirectional and unidirectional LSTM recurrent neural network for forecasting network-wide traffic state with missing values[J]. Transportation Research Part C: Emerging Technologies, 2020, 118: 102674. doi: 10.1016/j.trc.2020.102674
[33]	PAN Zhe-yi, WANG Zhao-yuan, WANG Wei-feng, et al. Matrix factorization for spatio-temporal neural networks with applications to urban flow prediction[C]//ACM. Proceedings of the 28th ACM International Conference on Information and Knowledge Management. New York: ACM, 2019: 2683-2691.
[34]	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, 2022, 37(3): 645-653. https://www.cnki.com.cn/Article/CJFDTOTAL-KZYC202312008.htm GU Zhen-yu, CHEN Cong, ZHENG Jia-jia, et al. Traffic flow prediction based on STG-CRNN[J]. Control and Decision, 2022, 37(3): 645-653. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KZYC202312008.htm
[35]	CUI Z Y, HENRICKSON K, KE R M, et al. Traffic graph convolutional recurrent neural network: a deep learning framework for network-scale traffic learning and forecasting[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(11): 4883-4894. doi: 10.1109/TITS.2019.2950416
[36]	HUANG Wen-hao, SONG Guo-jie, HONG Hai-kun, et al. Deep architecture for traffic flow prediction: deep belief networks with multitask learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(5): 2191-2201. doi: 10.1109/TITS.2014.2311123
[37]	HUANG Fei-hu, YI Pei-yu, WANG Jin-ce, et al. A dynamical spatial-temporal graph neural network for traffic demand prediction[J]. Information Sciences, 2022, 594: 286-304. doi: 10.1016/j.ins.2022.02.031
[38]	LI Wei, WANG Xin, ZHANG Yi-wen, et al. Traffic flow prediction over muti-sensor data correlation with graph convolution network[J]. Neurocomputing, 2021, 427: 50-63. doi: 10.1016/j.neucom.2020.11.032
[39]	CHEN Li, ZHENG Lin-jiang, YANG Jie, et al. Short-term traffic flow prediction: from the perspective of traffic flow decomposition[J]. Neurocomputing, 2020, 413, 444-456. doi: 10.1016/j.neucom.2020.07.009
[40]	陈喜群, 曹震, 沈楼涛, 等. 融合路段传输模型和深度学习的城市路网短时交通流状态预测[J]. 中国公路学报, 2021, 34(12): 203-216. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202112015.htm CHEN Xi-qun, CAO Zhen, SHEN Lou-tao, et al. Short-term traffic-state prediction of urban road networks based on the fusion of a link-transmission model and deep learning[J]. China Journal of Highway and Transport, 2021, 34(12): 203-216. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202112015.htm
[41]	高华兵, 舒文迪, 刘志. 基于深度学习的城市快速路交通流预测方法[J]. 浙江工业大学学报, 2022, 50(4): 406-412, 463. https://www.cnki.com.cn/Article/CJFDTOTAL-ZJGD202204007.htm GAO Hua-bing, SHU Wen-di, LIU Zhi. Urban expressway traffic flow prediction method based on deep learning[J]. Journal of Zhejiang University of Technology, 2022, 50(4): 406-412, 463. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZJGD202204007.htm
[42]	GUO Sheng-nan, LIN You-fang, WAN Huai-yu, et al. Learning dynamics and heterogeneity of spatial-temporal graph data for traffic forecasting[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(11): 5415-5428. doi: 10.1109/TKDE.2021.3056502
[43]	YE Ji-hua, XUE Sheng-jun, JIANG Ai-wen. Attention-based spatio-temporal graph convolutional network considering external factors for multi-step traffic flow prediction[J]. Digital Communications and Networks, 2022, 8: 343-350. doi: 10.1016/j.dcan.2021.09.007
[44]	XU Yao-bin, LIU Wei-tang, JIANG Zhong-yi, et al. MAF-GNN: multi-adaptive spatiotemporal-flow graph neural network for traffic speed forecasting[J]. arXiv, 2021, DOI: 10.48550/arXiv.2108.03594.
[45]	POLSON N G, SOKOLOV V O. Deep learning for short-term traffic flow prediction[J]. Transportation Research Part C: Emerging Technologies, 2017, 79: 1-17. doi: 10.1016/j.trc.2017.02.024
[46]	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
[47]	ZHANG Jun-bo, YU Zheng, QI De-kang, et al. Predicting citywide crowd flows using deep spatio-temporal residual networks[J]. Artificial Intelligence, 2018, 259: 147-166. doi: 10.1016/j.artint.2018.03.002
[48]	LIANG Y, OUYANG K, WANG Y, et al. Revisiting convolutional neural networks for citywide crowd flow analytics[C]//Springer. European Conference on Machine Learning and Knowledge Discovery in Databases. Berlin: Springer, 2020: 578-594.
[49]	李磊, 张青苗, 赵军辉, 等. 基于改进CNN-LSTM组合模型的分时段短时交通流预测[J]. 应用科学学报, 2021, 39(2): 185-198. https://www.cnki.com.cn/Article/CJFDTOTAL-YYKX202102001.htm LI Lei, ZHANG Qing-miao, ZHAO Jun-hui, et al. Short-term traffic flow prediction method of different periods based on improved CNN-LSTM[J]. Journal of Applied Sciences, 2021, 39(2): 185-198. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YYKX202102001.htm
[50]	KE Rui-min, LI Wan, CUI Zhi-yong, et al. Two-stream multi-channel convolutional neural network for multi-lane traffic speed prediction considering traffic volume impact[J]. Transportation Research Record, 2020, 2674(4): 459-470. doi: 10.1177/0361198120911052
[51]	ZOU Z, HAO P, LIN L, et al. Deep convolutional mesh RNN for urban traffic passenger flows prediction[C]//IEEE. 2018 SmartWorld/UIC/ATC/ScalCom/CBDCom/IoP/SCI. New York: IEEE, 2018: 1305-1310.
[52]	WANG Yun-bo, ZHANG Jian-jin, ZHU Hong-yu, et al. Memory in memory: a predictive neural network for learning higher-order non-stationarity from spatiotemporal dynamics[C]//IEEE. 32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2019: 9146-9154.
[53]	WANG Yun-bo, GAO Zhi-feng, LONG Ming-sheng, et al. PredRNN + + : towards a resolution of the deep-in-time dilemma in spatiotemporal predictive learning[C]//IMLS. 35th International Conference on Machine Learning. San Diego: IMLS, 2018: 8122-8131.
[54]	ZHENG Hai-feng, LIN Feng, FENG Xin-xin, et al. A hybrid deep learning model with attention-based Conv-LSTM networks for short-term traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(11): 6910-6920. doi: 10.1109/TITS.2020.2997352
[55]	彭博, 唐聚, 蔡晓禹, 等. 基于3DCNN-DNN的高空视频交通状态预测[J]. 交通运输系统工程与信息, 2020, 20(3): 39-46. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202003007.htm PENG Bo, TANG Ju, CAI Xiao-yu, et al. 3DCNN-DNN based traffic status prediction from aerial videos[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(3): 39-46. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202003007.htm
[56]	YU Feng, WEI Dan, ZHANG Shu-ting, et al. 3D CNN-based accurate prediction for large-scale traffic flow[C]//IEEE. The 4th International Conference on Intelligent Transportation Engineering (ICITE). New York: IEEE, 2019: 99-103.
[57]	CHEN Yi-bi, ZOU Xiao-feng, LI Ken-li, et al. Multiple local 3D CNNs for region-based prediction in smart cities[J]. Information Sciences, 2021, 542: 476-491. doi: 10.1016/j.ins.2020.06.026
[58]	ZHANG Shuai-chao, ZHOU Ling-xiao, CHEN Xi-qun, et al. Network-wide traffic speed forecasting: 3D convolutional neural network with ensemble empirical mode decomposition[J]. Computer-Aided Civil and Infrastructure Engineering, 2020, 35(10): 1132-1147. doi: 10.1111/mice.12575
[59]	UL ABIDEEN Z, SUN H L, YANG Z, et al. The deep 3D convolutional multi-branching spatial-temporal-based unit predicting citywide traffic flow[J]. Applied Sciences, 2020, 10(21): 7778. doi: 10.3390/app10217778
[60]	CEN C, LI K L, TEO S G, et al. Exploiting spatio-temporal correlations with multiple 3D convolutional neural networks for citywide vehicle flow prediction[C]//IEEE. 2018 IEEE International Conference on Data Mining. New York: IEEE, 2018: 893-898.
[61]	GUO Sheng-nan, LIN You-fang, FENG Ning, et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting[C]//AAAI. The Thirty-Third AAAI Conference on Artificial Intelligence. Washington DC: AAAI, 2019: 922-929.
[62]	CHEN Xu, ZHANG Yuan-xing, DU Lun, et al. TSSRGCN: temporal spectral spatial retrieval graph convolutional network for traffic flow forecasting[C]//IEEE. 2020 IEEE International Conference on Data Mining. New York: IEEE, 2020: 954-959.
[63]	WU Zong-han, PAN Shi-rui, LONG Guo-dong, et al. Graph WaveNet for deep spatial-temporal graph modeling[C]//AAAI. Proceedings of Twenty-Eighth International Joint Conference on Artificial Intelligence IJCAI-19. Washington DC: AAAI, 2019: 1907-1913.
[64]	BAI Jian-dong, ZHU Jia-wei, SONG Yu-jiao, et al. A3T-GCN: attention temporal graph convolutional network for traffic forecasting[J]. ISPRS International Journal of Geo-Information, 2021, 10(7): 485-496. doi: 10.3390/ijgi10070485
[65]	TIAN Yan-ling, ZHANG Qie-shi, REN Zi-liang, et al. Multi-scale dilated convolution network based depth estimation in intelligent transportation systems[J]. IEEE Access, 2019, 7(99): 185179-185188.
[66]	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
[67]	ZHANG Jian-ni, SHI Xing-jian, XIE Jun-juan, et al. GaAN: gated attention networks for learning on large and spatiotemporal graphs[J]. arXiv, 2018, DOI: 10.48550/arXiv.1803.07294.
[68]	CUI Zhi-yong, LIN Long-fei, PU Zi-yuan, et al. Graph Markov network for traffic forecasting with missing data[J]. Transportation Research Part C: Emerging Technologies, 2020, 117: 102671. doi: 10.1016/j.trc.2020.102671
[69]	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. 9th ACM International Conference on Information and Knowledge Management. New York: ACM, 2020: 1215-1224.
[70]	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[C]//AAAI. The Thirty-Fourth AAAI Conference on Artificial Intelligence. Washington DC: AAAI, 2020: 914-921.
[71]	SUN Jun-kai, ZHANG Jun-bo, LI Qiao-fei, et al. Predicting citywide crowd flows in irregular regions using multi-view graph convolutional networks[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(5): 2348-2359. doi: 10.1109/TKDE.2020.3008774
[72]	TANG Shan-shan, LI Bo, YU Hai-jun. ChebNet: efficient and stable constructions of deep neural networks with rectified power units using Chebyshev approximations[J]. arXiv, 2019, DOI: 10.48550/arXiv.1911.05467.
[73]	KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[C]//ICLR. 5th International Conference on Learning Representations. Washington DC: ICLR, 2017: 1-14.
[74]	LI Y G, YU R, SHAHABI C, et al. Diffusion convolutional recurrent neural network: data-driven traffic forecasting[C]//ICLR. 6th International Conference on Learning Representations. Washington DC: ICLR, 2018: 1-16.
[75]	VELIKOVI P, CUCURULL G, CASANOVA A, et al. Graph attention networks[C]//ICLR. 6th International Conference on Learning Representations. Washington DC: ICLR, 2018: 1-12.
[76]	HAMILTON W L, YING R, LESKOVEC J. Inductive representation learning on large graphs[C]//NIPS. 31st Conference on Neural Information Processing Systems. San Diego: NIPS, 2017: 1-11.
[77]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. arXiv, 2017, DOI: 10.48550/arXiv.1706.03762.
[78]	BROWN T B, MANN B, RYDER N, et al. Language models are few-shot learners[C]//NIPS. 34th Conference on Neural Information Processing Systems. San Diego: NIPS, 2020: 1-25.
[79]	ARNAB A, DEHGHANI M, HEIGOLD G, et al. ViViT: a video vision transformer[C]//IEEE. 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE, 2021: 6816-6826.
[80]	LI Y S, BACIU G. SAPCGA N: Self-Attention based generative adversarial network for point clouds[C]//IEEE. 19th IEEE International Conference on Cognitive Informatics and Cognitive Computing. New York: IEEE, 2020: 52-59.
[81]	ZHANG Kun-peng, JIA Ning, ZHENG Liang, et al. A novel generative adversarial network for estimation of trip travel time distribution with trajectory data[J]. Transportation Research Part C: Emerging Technologies, 2019, 108: 223-244. doi: 10.1016/j.trc.2019.09.019