Road Network Traffic Accident Risk Prediction Based on Spatio-Temporal Graph Convolution Network

doi:10.3778/j.issn.1002-8331.2203-0361

Abstract

Abstract: Road network traffic accident prediction is one of the most important ways to realize road management and control and route planning. Considering the characteristics of each road section in the road network and the influence of environmental factors, a spatiotemporal gated graph convolution（STGRGCN） model based on graph convolution network（GCN） and gated recurrent unit（GRU） is established to predict the risk of traffic accidents. The spatial correlation between roads is extracted through GCN, the temporal correlation in environmental factors is extracted through GRU, and then the temporal and spatial correlation is extracted through the composite module of GCN and GRU. The relevant data of Los Angeles and Houston in the national traffic accident dataset of the United States are selected to test the model. The root mean square error, average absolute error and recall rate of STGRGCN model are 4.09, 2.14, 0.714 and 5.79, 3.24, 0.683 respectively in the two cities, which are better than the existing statistical model, machine learning model and composite model. The ablation experiment of removing each module is designed to prove that each module of this model is helpful to improve the prediction performance.

Key words: traffic accident risk prediction, graph convolution network（GCN）, gated recurrent unit（GRU）, attention mechanism, deep learning

摘要： 路网交通事故预测是实现道路管控、路线规划的最重要方式之一。考虑到路网中各路段特征与环境因素的影响，建立基于图卷积神经网络（GCN）和门控循环单元（GRU）的时空门控图卷积（STGRGCN）模型预测交通事故风险。通过GCN提取出道路间的空间关联性，通过GRU提取出环境因素中的时间关联性，再通过GCN与GRU的复合模块提取出时空关联性。选取美国全国交通事故数据集中洛杉矶市和休斯顿市相关数据对模型进行检验，STGRGCN模型的均方根误差、平均绝对误差以及召回率在两个城市分别为4.09、2.14、0.714和5.79、3.24、0.683，优于已有统计模型、机器学习模型以及复合模型。设计该除各模块的消融实验，证明该模型各模块皆有助于提升预测性能。

关键词: 交通事故风险预测, 图卷积网络, 门控循环单元, 注意力机制, 深度学习

WANG Qingrong, ZHOU Yutong, ZHU Changfeng, WU Yuyu. Road Network Traffic Accident Risk Prediction Based on Spatio-Temporal Graph Convolution Network[J]. Computer Engineering and Applications, 2023, 59(13): 266-272.

王庆荣, 周禹潼, 朱昌锋, 吴玉玉. 时空图卷积网络下的路网交通事故风险预测[J]. 计算机工程与应用, 2023, 59(13): 266-272.

References

[1] World Health Organization.Global status report on road safety 2018[R].2018.
[2] IHUEZE C C，ONWURAH U O.Road traffic accidents prediction modelling：an analysis of anambra state，nigeria[J].Accident Analysis & Prevention，2018，112：21-29.
[3] CHANG L Y，CHEN W C.Data mining of tree-based models to analyze freeway accident frequency[J].Journal of Safety Research，2005，36（4）：365-375.
[4] HOSSAIN M，MUROMACHI Y.A Bayesian network based framework for real-time crash prediction on the basic freeway segments of urban expressways[J].Accident Analysis & Prevention，2012，45：373-381.
[5] REN H，SONG Y，WANG J，et al.A deep learning approach to the prediction of short-term traffic accident risk[J].arXiv：1710.09543，2017.
[6] NAJJAR A，SHUN’ICHI K，YOSHIKAZU M.Combining satellite imagery and open data to map road safety[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence，San Francisco，Feb 4-9，2017.
[7] CHE N，FAN X，ZHENG C，et al.SDCAE：stack denoising convolutional autoencoder model for accident risk prediction via traffic big data[C]//Proceedings of the 6th International Conference on Advanced Cloud and Big Data（CBD），2018.
[8] YUAN Z N，ZHOU X，YANG T B.Hetero-ConVLSTM：a deep learning approach to traffic accident prediction on heterogeneous spatio-temporal data[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining，London，Aug 19-23，2018：984-992.
[9] ZHU L，TIAN R L，SHENG D D.TA-STAN：a deep spatial-temporal attention learning framework for regional traffic accident risk prediction[C]//Proceedings of 2019 International Joint Conference on Neural Networks（IJCNN），2019.
[10] BRUNA J，ZAREMBA W，SZLAM A，et al.Spectral networks and locally connected networks on graphs[J].arXiv：1312.6203，2013.
[11] BOUKERCHE A，WANG J H.A performance modeling and analysis of a novel vehicular traffic flow prediction system using a hybrid machine learning-based model[J].Ad Hoc Networks，2020，106：102224.
[12] ZHAO L，SONG Y J.T-GCN：a temporal graph convolutional network for traffic prediction[J].IEEE Transactions on Intelligent Transportation Systems，2020，21（9）：3848-3858.
[13] CUI Z Y，HENRICKSON K，KE R，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.
[14] ZHOU J，CUI G，HU S，et al.Graph neural networks：a review of methods and applications[J].AI Open，2020，1：57-81.
[15] WU Z H.A comprehensive survey on graph neural networks[J].IEEE Transactions on Neural Networks and Learning Systems，2020，32（1）：4-24.
[16] CHUNG J，PAN S，CHEN F，et al.Empirical evaluation of gated recurrent neural networks on sequence modeling[J].arXiv：1412.3555，2014.
[17] MNIH V，NICOLAS H，ALEX G.Recurrent models of visual attention[C]//Advances in Neural Information Processing Systems，2014：2204-2212.