面向交通流量预测的多组件时空图卷积网络
|
摘要
流量预测一直是交通领域研究者和实践者关注的热点问题.流量数据具有高度的非线性和复杂性,对其进行精准预测具有很大的挑战,现有的预测方法大多不能很好地捕获数据的时空相关性.提出一种新颖的基于深度学习的多组件时空图卷积网络(MCSTGCN),以解决交通流量预测问题.MCSTGCN通过3个组件分别建模流量数据的近期、日周期、周周期特性,每个组件同时利用空间维图卷积和时间维卷积有效捕获交通数据的时空相关性.在美国加利福尼亚州高速公路流量公开数据集上进行了实验,结果表明,MCSTGCN模型的预测效果优于现有的预测方法.
Forecasting the traffic flows is a hot issue for researchers and practitioners in the transportation field. It is very challenging to forecast the traffic flows due to the high nonlinearity and complexity of the data, and most of the existing methods cannot effectively capture the spatial-temporal correlations of traffic flow data. In this paper, we propose a novel deep learning based model, multicomponent spatial-temporal graph convolution networks(MCSTGCN), to solve the problem of traffic flow forecasting. MCSTGCN employs three components to respectively model the recent, daily and weekly characteristics of traffic flow data. Each component uses graph convolutions in the spatial dimension and convolutions in the temporal dimension to effectively capture the spatial-temporal correlations of traffic data. Experiments on a public California freeway dataset show that the prediction performance of the MCSTGCN model is better than other existing prevalent methods.
Forecasting the traffic flows is a hot issue for researchers and practitioners in the transportation field. It is very challenging to forecast the traffic flows due to the high nonlinearity and complexity of the data, and most of the existing methods cannot effectively capture the spatial-temporal correlations of traffic flow data. In this paper, we propose a novel deep learning based model, multicomponent spatial-temporal graph convolution networks(MCSTGCN), to solve the problem of traffic flow forecasting. MCSTGCN employs three components to respectively model the recent, daily and weekly characteristics of traffic flow data. Each component uses graph convolutions in the spatial dimension and convolutions in the temporal dimension to effectively capture the spatial-temporal correlations of traffic data. Experiments on a public California freeway dataset show that the prediction performance of the MCSTGCN model is better than other existing prevalent methods.
引文
[1]Zhang JP,Wang FY,Wang KF,Lin WH,Xu X,Chen Ch.Data-driven intelligent transportation systems:A survey.IEEE Trans.on Intelligent Transportation Systems,2011,12(4):1624-1639.
[2]Williams BM,Hoel LA.Modeling and forecasting vehicular traffic flow as a seasonal arima process:Theoretical basis and empirical results.Journal of Transportation Engineering,2003,129(6):664-672.
[3]Van Lint H,Van Hinsbergen C.Short-term traffic and travel time prediction models.Transportation Research E-circular,2012,22(1):22-41.
[4]Jeong YS,Byon YJ,Castro-Neto MM,Easa SM.Supervised weighting-online learning algorithm for short-term traffic flow prediction.IEEE Trans.on Intelligent Transportation Systems,2013,14(4):1700-1707.
[5]Zhang J,Zheng Y,Qi D,Li RY,Yi XW,Li TR.Predicting citywide crowd flows using deep spatio-temporal residual networks.Artificial Intelligence,2018,259:147-166.
[6]Niepert M,Ahmed M,Kutzkov K.Learning convolutional neural networks for graphs.In:Proc.of the ICML.2016.2014-2023.
[7]Li C,Cui Z,Zheng W,Xu C,Yang J.Spatio-temporal graph convolution for skeleton based action recognition.In:Proc.of the AAAI.2018.
[8]Bruna J,Zaremba W,Szlam A,Lecun Y.Spectral networks and locally connected networks on graphs.In:Proc.of the ICLR.2014.
[9]Defferrard M,Bresson X,Vandergheynst P.Convolutional neural networks on graphs with fast localized spectral filtering.In:Proc.of the NIPS.2016.3844-3852.
[10]Seo Y,Defferrard M,Vandergheynst P,et al.Structured sequence modeling with graph convolutional recurrent networks.arXiv preprint arXiv:1612.07659,2016.
[11]Li Y,Yu R,Shahabi C,et al.Diffusion convolutional recurrent neural network:data-driven traffic forecasting.arXiv preprint arXiv:1707.01926v1,2017.
[12]Yu B,Yin H,Zhu Z.Spatio-temporal graph convolutional networks:A deep learning framework for traffic forecasting.In:Proc.of the IJCAI.2018.
[13]Henaff M,Bruna J,LeCun Y.Deep convolutional networks on graph-structured data.arXiv preprint arXiv:1506.05163,2015.
[14]Simonovsky M,Komodakis N.Dynamic edge-conditioned filters in convolutional neural networks on graphs.In:Proc.of the CVPR.2017.29-38.
[15]Kipf TN,Welling M.Semi-supervised classification with graph convolutional networks.In:Proc.of the ICLR.2017.
[16]Chen C,Petty K,Skabardonis A,Varaiya P,Jia ZF.Freeway performance measurement system:Mining loop detector data.Journal of the Transportation Research Board,2001,1748:96-102.
[17]Hochreiter S,Schmidhuber J.Long short-term memory.Neural Computation,1997,9(8):1735-1780.
[18]Chung J,Gulcehre C,Cho K,Bengio Y.Empirical evaluation of gated recurrent neural networks on sequence modeling.arXiv:Neural and Evolutionary Computing,2014.
[2]Williams BM,Hoel LA.Modeling and forecasting vehicular traffic flow as a seasonal arima process:Theoretical basis and empirical results.Journal of Transportation Engineering,2003,129(6):664-672.
[3]Van Lint H,Van Hinsbergen C.Short-term traffic and travel time prediction models.Transportation Research E-circular,2012,22(1):22-41.
[4]Jeong YS,Byon YJ,Castro-Neto MM,Easa SM.Supervised weighting-online learning algorithm for short-term traffic flow prediction.IEEE Trans.on Intelligent Transportation Systems,2013,14(4):1700-1707.
[5]Zhang J,Zheng Y,Qi D,Li RY,Yi XW,Li TR.Predicting citywide crowd flows using deep spatio-temporal residual networks.Artificial Intelligence,2018,259:147-166.
[6]Niepert M,Ahmed M,Kutzkov K.Learning convolutional neural networks for graphs.In:Proc.of the ICML.2016.2014-2023.
[7]Li C,Cui Z,Zheng W,Xu C,Yang J.Spatio-temporal graph convolution for skeleton based action recognition.In:Proc.of the AAAI.2018.
[8]Bruna J,Zaremba W,Szlam A,Lecun Y.Spectral networks and locally connected networks on graphs.In:Proc.of the ICLR.2014.
[9]Defferrard M,Bresson X,Vandergheynst P.Convolutional neural networks on graphs with fast localized spectral filtering.In:Proc.of the NIPS.2016.3844-3852.
[10]Seo Y,Defferrard M,Vandergheynst P,et al.Structured sequence modeling with graph convolutional recurrent networks.arXiv preprint arXiv:1612.07659,2016.
[11]Li Y,Yu R,Shahabi C,et al.Diffusion convolutional recurrent neural network:data-driven traffic forecasting.arXiv preprint arXiv:1707.01926v1,2017.
[12]Yu B,Yin H,Zhu Z.Spatio-temporal graph convolutional networks:A deep learning framework for traffic forecasting.In:Proc.of the IJCAI.2018.
[13]Henaff M,Bruna J,LeCun Y.Deep convolutional networks on graph-structured data.arXiv preprint arXiv:1506.05163,2015.
[14]Simonovsky M,Komodakis N.Dynamic edge-conditioned filters in convolutional neural networks on graphs.In:Proc.of the CVPR.2017.29-38.
[15]Kipf TN,Welling M.Semi-supervised classification with graph convolutional networks.In:Proc.of the ICLR.2017.
[16]Chen C,Petty K,Skabardonis A,Varaiya P,Jia ZF.Freeway performance measurement system:Mining loop detector data.Journal of the Transportation Research Board,2001,1748:96-102.
[17]Hochreiter S,Schmidhuber J.Long short-term memory.Neural Computation,1997,9(8):1735-1780.
[18]Chung J,Gulcehre C,Cho K,Bengio Y.Empirical evaluation of gated recurrent neural networks on sequence modeling.arXiv:Neural and Evolutionary Computing,2014.