计算机科学 ›› 2023, Vol. 50 ›› Issue (9): 318-330.doi: 10.11896/jsjkx.221000064
李可1,2,3, 杨玲1, 赵晏伯4, 陈泳龙4, 罗寿西1,3
LI Ke1,2,3, YANG Ling1, ZHAO Yanbo4, CHEN Yonglong4, LUO Shouxi1,3
摘要: 在大规模动态变化的交通场景下,快速准确地预测车辆驾驶行为是智能交通领域极具挑战的问题之一。车辆驾驶行为的预测不仅要考虑通信的有效性,而且要考虑车辆历史行驶轨迹以及车辆之间的相互影响。文中综合考虑了上述因素,提出了一种新的基于边-增强图卷神经网络的通信有效的分布式机器学习框架EGCN-CeDML(Edge-enhanced Graph Convolutional Neural Network-Communication-efficient Distributed Machine Learning)。相比面向单一设备的集中式预测框架,EGCN- CeDML是通信有效的分布式机器学习框架,该框架无需将所有原始数据发送到云服务器,而是直接将用户数据在本地边缘设备存储、处理和计算。这种在多个边缘设备训练神经网络的方式缓解了集中训练神经网络的压力,降低了传输数据量和通信延迟,提升了数据处理效率,在一定程度上也保护了用户隐私。各个边缘设备部署的复合图卷积网络(EGCN-LSTM)利用边-增强注意力机制和图卷积神经网络的特征传递机制,当周围车辆数量增长至十几辆时仍能快速提取和传递车辆间的交互信息,保证了较准确的预测性能和较低的时间复杂度。不限于车辆驾驶行为预测,各边缘设备可以根据自身的计算能能力和存储能力,在保证神经网络性能的前提下灵活控制神经网络的类型和规模以适用于不同的应用场景。EGCN-CeDML在公开数据集NGSIM上的实验结果表明:无论交通复杂程度如何,EGCN-CeDML的计算时间和预测性能都优于以往模型,精准率可达0.939 1,召回率可达0.955 7,F1分数可达0.947 3;预测时长为1 s时,预测准确率达到了91.21%;即使车辆数目增加,算法也能保持较低的时间复杂度,且稳定在0.1 s以内。
中图分类号:
[1]XING E P,HO Q,DAI W,et al.Petuum:A new platform for distributed machine learning on big data[C]//Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2015:1335-1344. [2]ANDREW G.Bring HPC techniques to deep learning[EB/OL].https://xueshu.baidu.com/bringing-hpc-techniques-deep-learning. [3]MA Y,ZHU X,ZHANG S,et al.Trafficpredict:Trajectory prediction for heterogeneous traffic-agents[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2019:6120-6127. [4]AMDAHL G M.Validity of the single processor approach toachieving large scale computing capabilities[C]//Spring Joint Computer Conference.1967:483-485. [5]GONG L,CHENG Q.Exploiting edge features for graph neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:9211-9219. [6]LEFÈVRE S,VASQUEZ D,LAUGIER C.A survey on motion prediction and risk assessment for intelligent vehicles[J].ROBOMECH Journal,2014,1(1):1-14. [7]LI Z,GONG J,LU C,et al.Interactive Behavior Prediction for Heterogeneous Traffic Participants in the Urban Road:A Graph-Neural-Network-Based Multitask Learning Framework[J].IEEE/ASME Transactions on Mechatronics,2021,26(3):1339-1349. [8]KUMAR P,PERROLLAZ M,LEFEVRE S,et al.Learning-based approach for online lane change intention prediction[C]//2013 IEEE Intelligent Vehicles Symposium(IV).IEEE,2013:797-802. [9]HE G,LI X,LV Y,et al.Probabilistic intention prediction and trajectory generation based on dynamic bayesian networks[C]//2019 Chinese Automation Congress(CAC).IEEE,2019:2646-2651. [10]AOUDE G S,DESARAJU V R,STEPHENS L H,et al.Driver behavior classification at intersections and validation on large naturalistic data set[J].IEEE Transactions on Intelligent Transportation Systems,2012,13(2):724-736. [11]DEO N,TRIVEDI M M.Convolutional social pooling for vehicle trajectory prediction[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops.2018:1468-1476. [12]GAO J,SUN C,ZHAO H,et al.Vectornet:Encoding hd maps and agent dynamics from vectorized representation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:11525-11533. [13]BUCILA C,CARUANA R,NICULESCU-MIZIL A.ModelCompression[C]//ACM SIGKDD International Conference on Knowledge Discovery and Data Mining(KDD'06).2006. [14]HINTON G,VINYALS O,DEAN J.Distilling the knowledge in a neural network[J].arXiv:1503.02531,2015. [15]SCARSELLI F,GORI M,TSOI A C,et al.The graph neural network model[J].IEEE Transactions on Neural Networks,2008,20(1):61-80. [16]WANG X,JI H,SHI C,et al.Heterogeneous graph attention network[C]//The World Wide Web Conference.2019:2022-2032. [17]KRAJEWSKI R,BOCK J,KLOEKER L,et al.The highd dataset:A drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems[C]//2018 21st International Conference on Intelligent Transportation Systems(ITSC).IEEE,2018:2118-2125. [18]GUO J.Research on Precise Hybrid Recommendation MethodBased on Feature Engineering and Efficient Gradient Boosting Tree Algorithm[D].Wuhan:Wuhan University of Science and Technology,2020. [19]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Imagenetclassification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6):84-90. [20]RUMELHART D E,HINTON G E,WILLIAMS R J.Learning Representations by Back Propagating Errors[J].Nature,1986,323(6088):533-536. [21]TEERAPITTAYANON S,MCDANEL B,KUNG H T.Distri-buted Deep Neural Networks Over the Cloud,the Edge and End Devices[C]//2017 IEEE 37th International Conference on Distributed Computing Systems(ICDCS).IEEE,2017:328-339. [22]LECUN Y,BOTTOU L.Gradient-based learning applied to do-cument recognition[C]//Proceedings of the IEEE.1998:2278-2324. |
|