摘要
旅行时间预测是城市智能交通系统的重要指标。采用深度学习中的长短期记忆神经网络(Long Short-Term Memory,LSTM)方法预测道路旅行时间,通过调节LSTM隐藏层单元数和训练次数得到最优的时间相关的LSTM模型;而后将改进时间型LSTM模型和传统BP(Back Propagation)神经网络模型、支持向量机模型、k NN模型以及时间序列ARIMA模型进行对比分析。实验结果表明,改进的T-LSTM模型训练效率和预测精度均优于其他四种模型。
Travel time prediction is an significant part of the intelligent transportation system. The paper uses the Long Short-Term Memory( LSTM) framework to predict urban road travel time. The optimal T-LSTM model is obtained by adjusting the number of LSTM hidden layer units and the number of training times,then compared with the traditional BP( Back Propagation) neural network model,SVM model,k NN model and ARIMA model. The results show that both the training efficiency and prediction accuracy of the LSTM model with time correlation are better than those of the other four models.
引文
[1]上半年全国机动车和驾驶人保持高位增长[J].道路交通管理,2018(8):8.
[2]姚志胜,邵春福,熊志华.支持向量机在路段行程时间预测中的应用研究[J].公路交通科技,2007,24(9):96-99.
[3]Daniel Billings,Jiann-Shiou Yang.Application of the ARI-MA Models to Urban Roadway Travel Time Prediction-ACase Study[J].IEEE International Conference on Systems,2006(3):2529-2534
[4]M Jiwon,DK Kim,SY Kho,CH Park.Travel Time Prediction Using k Nearest Neighbor Method with Combined Data from Vehicle Detector System and Automatic Toll Collection System[J].Transportation Research Record Journal of the Transportation Research Board,2011,20(2256):51-59
[5]邱祥.基于神经网络的智能交通控制系统设计[D].扬州:扬州大学,2016
[6]SI Chien,CM Kuchipudi.Dynamic Travel Time Prediction with Real-Time and Historic Data[J].Journal of Transportation Engineering,2003,129(6):608-616
[7]孙志军,薛磊,许阳明,等.深度学习研究综述[J].计算机应用研究,2012(8):2806-2810
[8]C Siripanpornchana,S Panichpapiboon,P Chaovalit.Travel-time prediction with deep learning[C]∥2016IEEE Region 10 Conference(TENCON).USA:IEEE,2016:1859-1962
[9]Zhang J,Chen H,Zhou H,et al.Freeway Travel Time Prediction Research Based on A Deep Learning Approach[C]]∥4th International Conference on Advanced Materials and Information Technology Processing(AMITP2016),Guilin,China,Atlantis:Atlantis Press,2016:487-493
[10]Wang J,Gu Q,Wu J.Traffic Speed Prediction and Congestion Source Exploration:A Deep Learining Method[C]∥2016 IEEE 16th International Conference on Data Mining(ICDM),Barcelona,Spain,2016.USA:IEEE,2016:499-508
[11]Duan Y,Lv Y,Wang F.Travel time prediction with LSTM neural network[C]∥2016 IEEE 19thInternational Conference on Intelligent Transportation Systems(ITSC),Rio de Janeiro,Brazil,2016.USA:IEEE,2016:1053-1058
[12]Highways England[EB/OL].[2018-07-02]https:∥data.gov.uk/dataset/highways-agency-network-journey-time-andtraffic-flow-data.
[13]M Abadi,A Agarwal,P Barham etc.Tensor Flow:LargeScale Machine Learning on Heterogeneous Distributed Systems[M].California:2015.