一种融合纵横时空特征的交通流预测方法

doi:10.19665/j.issn1001-2400.20221101

摘要/Abstract

摘要：

针对现有城市道路交通流预测研究中,上下游交通流时滞特性与空间流动特性挖掘不足、车道级交通流时空特性考虑不充分的问题,提出一种融合纵横时空特征的交通流预测方法。首先,通过计算延迟时间量化并消除上下游交通流断面间的空间时滞影响,增强上下游交通流序列的时空相关性。其次,将消除空间时滞的交通流通过向量拆分数据输入方式传入双向长短时记忆网络,用以捕捉上下游交通流纵向的传递与回溯双向时空关系,同时利用多尺度卷积群挖掘待预测断面内部各车道交通流间多时间步横向时空关系。最后,采用注意力机制动态融合纵横时空特征得到预测输出值。实验结果表明,相较于常规时间序列预测模型,所提方法在单步预测实验中,平均绝对误差、均方根误差分别下降了约15.26%、13.83%,决定系数提升了约1.25%。在中长时多步预测实验中,进一步证明了所提方法可有效挖掘纵横向交通流的细粒化时空特征,并具有一定的稳定性和普适性。

关键词: 城市交通, 交通流预测, 纵横时空相关性, 深度学习, 特征融合

Abstract:

Aiming at the problems of insufficient mining of time delay characteristics and spatial flow characteristics of upstream and downstream traffic flow as well as insufficient consideration of spatiotemporal characteristics of lane-level traffic flow in existing urban road traffic flow prediction research,a traffic flow prediction method for integrating longitudinal and horizontal spatiotemporal characteristics is proposed.First,the method quantifies and eliminates the effect of spatial time lag between upstream and downstream traffic flow by calculating the delay time to enhance the spatiotemporal correlation of upstream and downstream traffic flow sequences.Then,the traffic flow with the elimination of spatial time lag is passed into the bidirectional long short-term memory network through the vector split data input method to capture the longitudinal transmission and backtracking bidirectional spatiotemporal relationship of upstream and downstream traffic flow.At the same time,the multiscale convolution group is used to mine the multi-time step horizontal spatiotemporal relationship between the traffic flows of each lane in the section to be predicted.Finally,the attention mechanism is used to dynamically fuse the longitudinal and horizontal spatiotemporal characteristics to obtain the predicted value.Experimental results show that by applying the proposed method in the single-step prediction experiment,the MAE and RMSE decrease by 15.26% and 13.83% respectively,and increase by 1.25% compared with conventional time series prediction model.In the medium and long-term multi-step prediction experiment,it is further proved that the proposed method can effectively mine the fine-grained spatiotemporal characteristics of longitudinal and horizontal traffic flow,and has a certain stability and universality.

Key words: urban transportation, traffic flow prediction, longitudinal and horizontal spatiotemporal correlation, deep learning, feature fusion

中图分类号:

U491

侯越,郑鑫,韩成艳. 一种融合纵横时空特征的交通流预测方法[J]. 西安电子科技大学学报, 2023, 50(5): 65-74.

HOU Yue,ZHENG Xin,HAN Chengyan. Traffic flow prediction method for integrating longitudinal and horizontal spatiotemporal characteristics[J]. Journal of Xidian University, 2023, 50(5): 65-74.

图/表 11

图1

图2

表1

图3

图4

图5

图6

图7

表2

图8

图9

参考文献 17

[1]	DIMITRAKOPOULOS G, DEMESTICHAS P. Intelligent Transportation Systems[J]. IEEE Vehicular Technology Magazine, 2010, 5(1):77-84. doi: 10.1109/MVT.2009.935537
[2]	VAN DER VOORT M, DOUGHERTY M, WATSON S. Combining Kohonen Maps with ARIMA Time Series Models to Forecast Traffic Flow[J]. Transportation Research Part C:Emerging Technologies, 1996, 4(5):307-318. doi: 10.1016/S0968-090X(97)82903-8
[3]	KUMAR S V, VANAJAKSHI L. Short-Term Traffic Flow Prediction Using Seasonal ARIMA Model with Limited Input Data[J]. European Transport Research Review, 2015, 7(3):1-9. doi: 10.1007/s12544-014-0149-x
[4]	SU H, ZHANG L, YU S. Short-Term Traffic Flow Prediction Based on Incremental Support Vector Regression[C]//Third International Conference on Natural Computation (ICNC 2007).Piscataway:IEEE, 2007:640-645.
[5]	CSIKÓS A, VIHAROS Z J, KIS K B, et al. Traffic Speed Prediction Method for Urban Networks—an ANN Approach[C]//2015 International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS).Piscataway:IEEE, 2015:102-108.
[6]	罗文慧, 董宝田, 王泽胜. 基于CNN-SVR混合深度学习模型的短时交通流预测[J]. 交通运输系统工程与信息, 2017, 17(5):68-74.
	LUO Wenhui, DONG Baotian, WANG Zesheng. Short-Term Traffic Flow Prediction Based onCNN-SVR Hybrid Deep Learning Model[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(5):68-74.
[7]	DAI X, FU R, ZHAO E, et al. Deep Trend 2.0:ALight-Weighted Multi-Scale Traffic Prediction Model Using Detrending[J]. Transportation Research Part C:Emerging Technologies, 2019, 103:142-157. doi: 10.1016/j.trc.2019.03.022
[8]	TIAN Y, PAN L. Predicting Short-Term Traffic Flow by Long Short-Term Memory Recurrent Neural Network[C]//2015 IEEE International Conference on Smart City/SocialCom/SustainCom (SmartCity).Piscataway:IEEE, 2015:153-158.
[9]	陆文琦, 芮一康, 冉斌, 等. 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.
	LU Wenqi, RUI Yikang, RAN Bin, et al. Traffic Flow Prediction Based on Hybrid Deep Learning under Connected and Automated Vehicle Environment[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(3):47-53.
[10]	MA X, TAO Z, WANG Y, 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
[11]	ZHENG H, LIN F, FENG X, 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, 2020, 22(11):6910-6920. doi: 10.1109/TITS.2020.2997352
[12]	刘小明, 田玉林, 唐少虎, 等. 基于时延特性建模的多断面短时交通流预测[J]. 交通运输系统工程与信息, 2020, 20(3):54-60.
	LIU Xiaoming, TIAN Yulin, TANG Shaohu, et al. Short-Term Traffic Flow Prediction of Multi-Sections Based on Time-Delay Modeling[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(3):54-60.
[13]	谷远利, 陆文琦, 李萌, 等. 基于组合深度学习的快速路车道级速度预测研究[J]. 交通运输系统工程与信息, 2019, 19(4):79-86.
	GU Yuanli, LU Wenqi, LI Meng, et al. Lane-Level Traffic Speed Prediction for Expressways Based on a Combined Deep Learning Model[J]. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(4):79-86.
[14]	李桃迎, 王婷, 张羽琪. 考虑多特征的高速公路交通流预测模型[J]. 交通运输系统工程与信息, 2021, 21(3):101-111.
	LI Taoying, WANG Ting, ZHANG Yuqi. Highway Traffic Flow Prediction Model with Multi-Features[J]. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(3):101-111.
[15]	HOCHREITER S, SCHMIDHUBER J. Long Short-Term Memory[J]. Neural Computation, 1997, 9(8):1735-1780. doi: 10.1162/neco.1997.9.8.1735 pmid: 9377276
[16]	戚艳军, 孔月萍, 王佳婧, 等. 一种LSTM与CNN相结合的步态识别方法[J]. 西安电子科技大学学报, 2021, 48(5):78-85.
	QI Yanjun, KONG Yueping, WANG Jiajing, et al. Gait Recognition Method Combining LSTM and CNN[J]. Journal of Xidian University, 2021, 48(5):78-85.
[17]	杨晓莉, 蔺素珍. 一种注意力机制的多波段图像特征级融合方法[J]. 西安电子科技大学学报, 2020, 47(1):120-127.
	YANG Xiaoli, LIN Suzhen. Method for Multi-Band Image Feature-Level Fusion Based on the Attention Mechanism[J]. Journal of Xidian University, 2020, 47(1):120-127.

车道	车道1	车道2	车道3	车道4	车道5
车道1	1.000	0.679	0.642	0.658	0.447
车道2	0.679	1.000	0.815	0.811	0.532
车道3	0.642	0.815	1.000	0.926	0.552
车道4	0.658	0.811	0.928	1.000	0.598
车道5	0.447	0.532	0.552	0.598	1.000

模型	R_MAE	R_RMSE	R²
LSTM	4.633	6.458	0.953 60
Bi-LSTM	4.549	6.333	0.955 38
Bi-LSTM+纵向时空特征	4.344	6.195	0.957 30
LSTM+纵横时空特征	4.318	6.101	0.958 59
Bi-LSTM+纵横时空特征	4.190	5.940	0.960 75
Bi-LSTM-MConv	3.973	5.658	0.964 39
Bi-LSTM-MConv-Attention	3.926	5.565	0.965 55