基于时空信息融合学习的路段行程车速短时预测

doi:10.19678/j.issn.1000-3428.0060164

摘要/Abstract

摘要： 路段行程车速的变化受时间和空间维度信息的综合影响，多数神经网络模型仅从时间维度上预测路段行程车速的变化规律，未能全面考虑路网结构和上下游交通状态对路段行程车速的影响。结合图卷积网络和门控循环单元构建深度学习模型，挖掘路段行程车速的时空特性。通过在线地图平台获取路段实时行程车速，使用等维递补方法更新历史序列数据，提高预测实时性。在深圳市部分区域路网上的实验结果表明，该模型的多步预测精度均在90%以上，相比自回归积分滑动平均模型、支持向量机回归模型和门控循环单元模型最高提升了6.9%、1.3%和0.4%，具有更优的路段行程车速预测效果。

关键词: 短时预测, 图卷积网络, 门控循环单元, 时空相关性, 在线地图, 等维递补

Abstract: The change of road travel speed in a certain road is affected by both temporal and spatial factors.Most of the existing neural network models predict the change of road travel speed by considering only the temporal factors, and ignore the impact of road network structure and upstream and downstream traffic.To address the problem, the spatio-temporal features of road travel speed are analyzed, and on this basis a deep learning model is constructed by combining the Graph Convolutional Network(GCN) and the Gated Recurrent Unit(GRU).The real-time road travel speed is obtained through the online map platform, and the historical sequence data is continuously updated by using the method of equal-dimensional recursive compensation, so the real-time performance of prediction is improved.The model is tested with the data of road networks in some regions of Shenzhen.The experimental results show that the proposed model exhibits excellent performance in predicting travel speed in a certain road section, and provides an accuracy of over 90% for multi-step prediction.Its prediction accuracy is up to 6.9% higher than that of the Auto-Regressive Integral Moving Average (ARIMA) model, 1.3% higher than that of the Support Vector Regression(SVR) model, and 0.4% higher than that of the GRU model.

Key words: short-term prediction, Graph Convolutional Network(GCN), Gated Recurrent Unit(GRU), spatio-temporal correlation, online map, equal-dimensional recursive compensation

中图分类号:

TP391.4

杨顶, 邓明君, 徐丽萍. 基于时空信息融合学习的路段行程车速短时预测[J]. 计算机工程, 2021, 47(12): 78-86.

YANG Ding, DENG Mingjun, XU Liping. Short-Term Prediction of Road Travel Speed Based on Spatio-Temporal Information Fusion Learning[J]. Computer Engineering, 2021, 47(12): 78-86.

http://www.ecice06.com/CN/Y2021/V47/I12/78

图/表 13

20211213181731

20211213181735

20211213181738

20211213181741

20211213181744

20211213181747

20211213181751

20211213181754

20211213181757

20211213181800

20211213181803

20211213181806

20211213181809

参考文献

[1] LIEBIG T, PIATKOWSKI N, BOCKERMANN C, et al.Dynamic route planning with real-time traffic predictions[J].Information Systems, 2017, 64:258-265.
[2] 蔡延光, 乐冰, 蔡颢, 等.暴雨天气下高速公路短时交通流预测[J].计算机工程, 2020, 46(6):34-39. CAI Y G, LE B, CAI H, et al.Short-term traffic flow forecast of expressway under heavy rain[J].Computer Engineering, 2020, 46(6):34-39.(in Chinese)
[3] SIVAK M, SCHOETTLE B.Eco-driving:strategic, tactical, and operational decisions of the driver that influence vehicle fuel economy[J].Transport Policy, 2012, 22:96-99.
[4] 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. LIU J, GUAN W.A summary of traffic flow forecasting methods[J].Journal of Highway and Transportation Research and Development, 2004, 21(3):82-85.(in Chinese)
[5] MAKRIDAKIS S, HIBON M.ARMA models and the Box-Jenkins methodology[J].Journal of Forecasting, 1997, 16(3):147-163.
[6] GUO J H, HUANG W, WILLIAMS B M.Adaptive Kalman filter approach for stochastic short-term traffic flow rate prediction and uncertainty quantification[J].Transportation Research Part C:Emerging Technologies, 2014, 43:50-64.
[7] WU C H, HO J M, LEE D T.Travel-time prediction with support vector regression[J].IEEE Transactions on Intelligent Transportation Systems, 2004, 5(4):276-281.
[8] SUN S L, ZHANG C S, YU G Q.A Bayesian network approach to traffic flow forecasting[J].IEEE Transactions on Intelligent Transportation Systems, 2006, 7(1):124-132.
[9] HUANG W H, SONG G J, HONG H K, 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.
[10] 代亮, 梅洋, 钱超, 等.基于深度学习的短时交通量预测研究综述[J].计算机科学, 2019, 46(3):39-47. DAI L, MEI Y, QIAO C, et al.Survey on short-term traffic flow forecasting based on deep learning[J].Computer Science, 2019, 46(3):39-47.(in Chinese)
[11] MA X L, DAI Z, HE Z B, et al.Learning traffic as images:a deep convolutional neural network for large-scale transportation network speed prediction[J].Sensors, 2017, 17(4):818.
[12] YU H Y, WU Z H, WANG S Q, et al.Spatiotemporal recurrent convolutional networks for traffic prediction in transportation networks[J].Sensors, 2017, 17(7):1501.
[13] YU B, YIN H T, ZHU Z X.Spatio-temporal graph convolutional networks:a deep learning framework for traffic forecasting[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm, Sweden:[s.n.], 2018:1-10.
[14] ZHAO L, SONG Y J, ZHANG C, et al.T-GCN:a temporal graph convolutional network for traffic prediction[J].IEEE Transactions on Intelligent Transportation Systems, 2020, 21(9):3848-3858.
[15] 闫旭, 范晓亮, 郑传潘, 等.基于图卷积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 X, FAN X L, ZHENG C P, 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)
[16] 鲁武当.基于交通态势数据的城市道路平均行程速度时空分布特征及预测研究[D].西安:长安大学, 2019. LU W D.Spatial-temporal distribution characteristics and prediction of urban road average travel speed based on traffic status data[D].Xi'an:Chang'an University, 2019.(in Chinese)
[17] 陈华伟, 邵毅明, 敖谷昌, 等.基于在线地图交通状态的关键道路动态识别方法[J].交通运输系统工程与信息, 2019, 19(5):50-58. CHEN H W, SHAO Y M, AO G C, et al.Dynamic identification method of critical roads based on traffic state of online map[J].Journal of Transportation Systems Engineering and Information Technology, 2019, 19(5):50-58.(in Chinese)
[18] ZHANG Z C, LI M, LIN X, 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.
[19] 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. WANG X X, XU L H.Short-term traffic flow prediction based on deep learning[J].Journal of Transportation Systems Engineering and Information Technology, 2018, 18(1):81-88.(in Chinese)
[20] 陈卓, 吕娜, 陈坤, 等.基于时空图卷积网络的无人机网络入侵检测方法[J].北京航空航天大学学报, 2021, 47(5):1068-1076. CHEN Z, LÜ N, CHEN K, et al.UAV network intrusion detection method based on spatio-temporal graph convolutional network[J].Journal of Beijing University of Aeronautics and Astronautics, 2021, 47(5):1068-1076.(in Chinese)
[21] 刘明宇, 吴建平, 王钰博, 等.基于深度学习的交通流量预测[J].系统仿真学报, 2018, 30(11):4100-4105, 4114. LIU M Y, WU J P, WANG Y B, et al.Traffic flow prediction based on deep learning[J].Journal of System Simulation, 2018, 30(11):4100-4105, 4114.(in Chinese)
[22] 赵桂生, 张海文, 刘爱军, 等.基于等维递补灰色GM(1, 1)模型的我国农业灌溉用水预测分析[J].数学的实践与认识, 2018, 48(4):299-304. ZHAO G S, ZHANG H W, LIU A J, et al.Prediction and analysis of agricultural irrigation water in China based on grey GM(1, 1) model with equal dimension[J].Mathematics in Practice and Theory, 2018, 48(4):299-304.(in Chinese)

选择文件类型/文献管理软件名称

选择包含的内容