基于GNN-LSTM-CNN网络的6G车辆轨迹预测算法

doi:10.19665/j.issn1001-2400.2023.03.005

摘要/Abstract

摘要：

6G时代将实现万物互联,建立多层级、全覆盖的无缝连接,车联网作为通信、交通、汽车等多个行业融合交叉的关键领域将借助6G技术发展、部署。针对6G车联网中车辆轨迹预测精度不足的问题,采用深度学习的方法,提出了一种三通道神经网络模型。该模型考虑了车辆之间的交互信息、目标车辆的轨迹信息和车道结构信息对轨迹的影响。模型使用长短期记忆网络(LSTM)提取车辆轨迹信息特征,使用图神经网络(GNN)提取不同车辆之间的交互特征,使用卷积神经网络(CNN)提取车道结构特征。通过计算三通道特征向量的权重得到目标车辆预测的轨迹;通过NGSIM数据集对模型进行训练和测试。测试结果表明:与其他预测模型相比,考虑多维度信息的三通道网络预测方法在预测精度和长时域预测上有优势,预测精度提高了20%以上。降低6G车联网系统的数据传输量,可提升车联网系统的用户隐私安全。

关键词: 自动驾驶, 轨迹预测, 神经网络, 长短期记忆网络

Abstract:

The 6G era will realize the interconnection of all things and establish a multi-layer and full-coverage seamless connection.The Internet of Vehicles will be developed and deployed with the help of the 6G technology as a key area for the integration and intersection of communication,transportation,automobile and other industries.Aiming at the insufficient accuracy of the prediction of vehicle trajectories in the 6G Internet of Vehicles,this paper proposes a three-channel neural network model with the method of deep learning.This model takes the impacts of vehicle interaction information,target vehicle trajectories and lane structure information on trajectories into consideration.The long short-term memory network (LSTM) is used to extract the vehicle track information features,graph neural network (GNN) to extract interaction features between different vehicles,and the convolution neural network (CNN) is used to extract lane structure features.The predicted trajectory of the target vehicle is obtained by calculating the weight of the three-channel feature vector and the model is trained and tested by the NGSIM data set.The test results show that the three-channel network prediction method considering multi-dimension information has advantages in prediction accuracy and long time domain prediction compared with other prediction models,and the prediction accuracy is improved by more than 20%.Reducing the data transmission volume of the 6G Internet of Vehicles system can improve the user’s privacy security of the Internet of Vehicles system.

Key words: autopilot, trajectory prediction, neural network, long and short term memory network

中图分类号:

U461

蔡国庆,刘玲,张冲,周一青. 基于GNN-LSTM-CNN网络的6G车辆轨迹预测算法[J]. 西安电子科技大学学报, 2023, 50(3): 50-60.

CAI Gouqing,LIU Ling,ZHANG Chong,ZHOU Yiqing. Algorithm for prediction of the 6G vehicle trajectory based on the GNN-LSTM-CNN network[J]. Journal of Xidian University, 2023, 50(3): 50-60.

图/表 11

图1

图2

图3

图4

表1

图5

图6

图7

图8

表2

表3

参考文献 33

[1]	LIU L, ZHOU Y, YUAN J, et al. Economically Optimal MS Association for Multimedia Content Delivery in Cache-Enabled Heterogeneous Cloud Radio Access Networks[J]. IEEE Journal on Selected Areas in Communications, 2019, 37(7):1584-1593. doi: 10.1109/JSAC.49
[2]	LIU L, ZHOU Y, ZHUANG W, et al. Tractable Coverage Analysis for Hexagonal Macrocell-Based Heterogeneous UDNs with Adaptive Interference-Aware CoMP[J]. IEEE Transactions on Wireless Communications, 2018, 18(1):503-517. doi: 10.1109/TWC.2018.2882434
[3]	LIU L, ZHOU Y, GARCIA V, et al. Load Aware Joint CoMP Clustering and Inter-Cell Resource Scheduling in Heterogeneous Ultra Dense Cellular Networks[J]. IEEE Transactions on Vehicular Technology, 2017, 67(3):2741-2755. doi: 10.1109/TVT.25
[4]	ZHOU Y, LIU H, PAN Z, et al. Cooperative Multicast with Location Aware Distributed Mobile Relay Selection:Performance Analysis and Optimized Design[J]. IEEE Transactions on Vehicular Technology, 2017, 66(9):8291-8302. doi: 10.1109/TVT.2017.2682903
[5]	ZHOU Y, LIU H, PAN Z, et al. Energy-Efficient Two-Stage Cooperative Multicast:Effect of User Density[J]. IEEE Transactions on Vehicular Technology, 2015, 65(9):7297-7307. doi: 10.1109/TVT.2015.2494066
[6]	ZHOU Y, LIU H, PAN Z, et al. Two-Stage Cooperative Multicast Transmission with Optimized Power Consumption and Guaranteed Coverage[J]. IEEE Journal on Selected Areas in Communications, 2013, 32(2):274-284. doi: 10.1109/JSAC.2014.141208
[7]	ZHOU Y, LIU L, WANG L, et al. Service-Aware 6G:An Intelligent and Open Network Based on the Convergence of Communication,Computing and Caching[J]. Digital Communications and Networks, 2020, 6(3):253-260. doi: 10.1016/j.dcan.2020.05.003
[8]	SU Y, LIU Y, ZHOU Y, et al. Broadband LEO Satellite Communications:Architectures and Key Technologies[J]. IEEE Wireless Communications, 2019, 26(2):55-61.
[9]	ZHOU Y, TIAN L, LIU L, et al. Fog Computing Enabled Future Mobile Communication Networks:A Convergence of Communication and Computing[J]. IEEE Communications Magazine, 2019, 57(5):20-27.
[10]	赛迪智库无线电管理所. 6G概念及愿景白皮书 (2020)[R/OL].[2021-02-17]. http://www.ccidwise.com/plus/view.php?aid=16087&tyid=3.
[11]	MARCANO M, DÍAZ S, PÉREZ J, et al. A Review of Shared Control for Automated Vehicles:Theory and Applications[J]. IEEE Transactions on Human-Machine Systems, 2020, 50(6):475-491. doi: 10.1109/THMS.6221037
[12]	郎平, 田大新. 面向6G的车联网关键技术[J]. 中兴通讯技术, 2021, 27(2):13-16.
	LANG Ping, TIAN Daxin. Key Technologies of Internet of Vehicles System[J]. ZTE Technology Journal, 2021, 27(2):13-16.
[13]	董梦圆, 付宇钏, 牛晓健. 6G车联网中的通感算融合:现状与挑战[J]. 无线电通信技术, 2023, 49(1):37-45.
	DONG Mengyuan, FU Yuchuan, NIU Xiaojian. Integration of Sensing,Communication,Computing for 6G Internet of Vehicles:Current Status and Challenges[J]. Radio Communications Technology, 2023, 49(1):37-45.
[14]	白宝明, 马啸, 陈文, 等. 面向B5G/6G的信息传输与接入技术专题序言[J]. 西安电子科技大学学报, 2020, 47(6):1-4.
	BAI Baoming, MA Xiao, CHEN Wen, et al. Editorial:Introduction to the Special Issue on Information Transmission and Access Technologies for B5G/6G[J]. Journal of Xidian University, 2020, 47(6):1-4.
[15]	VEERARAGHAVAN H, PAPANIKOLOPOULOS N, SCHRATER P. Deterministic Sampling-Based Switching Kalman Filtering for Vehicle Tracking[C]//2006 IEEE Intelligent Transportation Systems Conference. Piscataway:IEEE, 2006:1340-1345.
[16]	BARTH A, FRANKE U. Where will the Oncoming Vehicle be the Next Second?[C]//2008 IEEE Intelligent Vehicles Symposium. Piscataway:IEEE, 2008:1068-1073.
[17]	VERMAAK J, DOUCET A, PATRICK P. Maintaining Multi-Modality through Mixture Tracking[C]//IEEE International Conference on Computer Vision. Piscataway:IEEE, 2003, 3:1110-1110.
[18]	TOLEDO-MOREO R, ZAMORA-IZQUIERDO M A. IMM-Based Lane-Change Prediction in Highways with Low-Cost GPS/INS[J]. IEEE Transactions on Intelligent Transportation Systems, 2009, 10(1):180-185. doi: 10.1109/TITS.2008.2011691
[19]	SCHLECHTRIEMEN J, WEDEL A, HILLENBRAND J, et al. A Lane Change Detection Approach Using Feature Ranking with Maximized Predictive Power[C]//2014 IEEE Intelligent Vehicles Symposium Proceedings. Piscataway:IEEE, 2014:108-114.
[20]	TRAN Q, FIRL J. Online Maneuver Recognition and Multimodal Trajectory Prediction for Intersection Assistance Using Non-Parametric Regression[C]//2014 IEEE Intelligent Vehicles Symposium Proceedings. Piscataway:IEEE, 2014:918-923.
[21]	HOUENOU A, BONNIFAIT P, CHERFAOUI V, et al. Vehicle Trajectory Prediction Based on Motion Model and Maneuver Recognition[C]//2013 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway:IEEE, 2013:4363-4369.
[22]	HOU L, XIN L, LI S E, et al. Interactive Trajectory Prediction of Surrounding Road Users for Autonomous Driving Using Structural-LSTM Network[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(11):4615-4625. doi: 10.1109/TITS.6979
[23]	刘启冉, 连静, 陈实, 等. 考虑交互轨迹预测的自动驾驶运动规划算法[J]. 东北大学学报:自然科学版, 2022, 43(7):930-936.
	LIU Qiran, LIAN Jing, CHEN Shi, et al. Motion Planning Algorithm of Autonomous Driving Considering Interactive Trajectory Prediction[J]. Journal of Northeastern University:Nature Science, 2022, 43(7):930-936.
[24]	KINGMAD,BA J. Adam:AMethod for Stochastic Optimization(2014)[J/OL].[2014-12-22]. https://arxiv.org/abs/1412.6980v1.
[25]	CHEN J, WANG Y, WU R, et al. Spatial-Temporal Graph Neural Network for Interaction-Aware Vehicle Trajectory Prediction[C]// 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE).Piscataway:IEEE, 2021:2119-2125.
[26]	FAN S, LI X, LI F. Intention-Driven Trajectory Prediction for Autonomous Driving[C]//2021 IEEE Intelligent Vehicles Symposium (IV). Piscataway:IEEE, 2021:107-113.
[27]	LI J, YANG F, TOMIZUKA M, et al. Evolvegraph:Multi-Agent Trajectory Prediction with Dynamic Relational Reasoning[J]. Advances in Neural Information Processing Systems, 2020, 33:19783-19794.
[28]	MO X, XING Y, LV C. Interaction-Aware Trajectory Prediction of Connected Vehicles Using CNN-lstm Networks[C]// IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society.Piscataway:IEEE, 2020:5057-5062.
[29]	MO X, XING Y, LV C. Graph and Recurrent Neural Network-based Vehicle Trajectory Prediction for Highway Driving[C]//2021 IEEE International Intelligent Transportation Systems Conference (ITSC). Piscataway:IEEE, 2021:1934-1939.
[30]	GONG L, CHENG Q. Exploiting Edge Features for Graph Neural Networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2019:9211-9219.
[31]	VELICKOVIC P, CUCURULL G, CASANOVA A, et al. Graph Attention Networks (2017)[J/OL].[2017-10-30]. https://arxiv.org/abs/1710.10903.
[32]	MESSAOUD K, YAHIAOUI I, VERROUST-BLONDET A, et al. Attention Based Vehicle Trajectory Prediction[J]. IEEE Transactions on Intelligent Vehicles, 2020, 6(1):175-185. doi: 10.1109/TIV
[33]	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. Piscataway:IEEE, 2018:1468-1476.

预测时间/s	文中方法	参考论文
预测时间/s	三通道法	CNN-LSTM网络^[28]	GNN网络^[29]	MHA-LSTM^[32]	CS-LSTM^[33]
1	0.57	0.64	0.59	0.41	0.62
2	0.77	0.96	0.82	1.01	1.29
3	0.93	1.22	1.21	1.74	2.13
4	1.16	1.53	1.67	2.67	3.20
5	1.60	2.09	1.96	3.83	4.52

预测时间/s	三通道法	双通道不使用GNN	双通道不使用RNN	双通道不使用CNN
1	0.57	0.63	0.59	0.60
2	0.77	0.95	0.82	0.78
3	0.93	1.25	1.03	1.08
4	1.16	1.62	1.27	1.25
5	1.60	2.21	1.92	1.95

模型	参数数量	训练时间(一个周期)/s	测试时间/s
CNN-LSTM网络^[28]	97 821	52	48
MHA-LSTM^[32]	67 342	26	22
三通道法	148 562	71	70