| 683 | 0 | 79 |
| 下载次数 | 被引频次 | 阅读次数 |
在自动驾驶领域，交通参与者的轨迹预测是一个重要而具有挑战性的问题，充分捕捉轨迹数据中复杂的时空特征对于准确预测轨迹至关重要。为解决时空特征提取不足和多模态车辆轨迹预测问题，提出一种基于时空特征交互的多模态车辆轨迹预测模型——STGA。采用基于动态图神经网络和基于融合注意力的时空Transformer网络捕获目标区域内车辆的空间交互特征和时间依赖性；设计特征融合的门控单元，实现对时空特征的有效融合，利用解码器生成目标区域未来车辆轨迹的概率分布；在公开数据集上对该模型进行了评估，并与基准模型进行了比较。实验结果表明，所提方法相比其他基准方法具有更好的性能，相较于最先进的基准方法，平均位移误差(Average Displacement Error, ADE)降低了32.03%,最终位移误差(Final Displacement Error, FDE)降低了14%。
Abstract：In the field of autonomous driving, trajectory prediction of traffic participants is a critical yet challenging problem, where accurately capturing the complex spatio-temporal features within trajectory data is essential for precise prediction. To address the issues of insufficient spatio-temporal feature extraction and multimodal vehicle trajectory prediction, a multimodal vehicle trajectory prediction model based on spatio-temporal feature interaction—STGA is proposed. Firstly, a dynamic graph neural network and a fusion-attention-based spatio-temporal Transformer network are utilized to capture both the spatial interaction features and temporal dependencies of vehicles within the target area. Next, a gated unit for feature fusion is designed to effectively merge spatio-temporal features, followed by the decoder generating a probability distribution for future trajectories of vehicles in the target region. Finally, the proposed model is evaluated on public dataset and compared against baseline models. Experimental results demonstrate that the proposed model outperforms other baseline methods, achieving a 32.03% reduction in Average Displacement Error(ADE) and a 14% reduction in Final Displacement Error(FDE) compared to the comparison models.
[1] MOZAFFARI S,AL-JARRAH O Y,DIANATI M,et al.Deep Learning-based Vehicle Behavior Prediction for Autonomous Driving Applications:A Review[J].IEEE Transactions on Intelligent Transportation Systems,2020,23(1):33-47.
[2] 乔少杰，韩楠，朱新文，等.基于卡尔曼滤波的动态轨迹预测算法[J].电子学报，2018,46(2):418-423.
[3] SCHREIER M,WILLERT V,ADAMY J.An Integrated Approach to Maneuver-based Trajectory Prediction and Criticality Assessment in Arbitrary Road Environments[J].IEEE Transactions on Intelligent Transportation Systems,2016,17(10):2751-2766.
[4] HUANG Y J,DU J T,YANG Z T,et al.A Survey on Trajectory-prediction Methods for Autonomous Driving[J].IEEE Transactions on Intelligent Vehicles,2022,7(3):652-674.
[5] RUDENKO A,PALMIERI L,HERMAN M,et al.Human Motion Trajectory Prediction:A Survey[J].The International Journal of Robotics Research,2020,39(8):895-935.
[6] DEO N,TRIVEDI M M.Multi-modal Trajectory Prediction of Surrounding Vehicles with Maneuver Based LSTM[C]//Proceedings of the 2018 IEEE Intelligent Vehicles Symposium.Changshu:IEEE,2018:1179-1184.
[7] LIM Q,JOHARI K,TAN U X.Gaussian Process Auto Regression for Vehicle Center Coordinates Trajectory Prediction[C]//Proceedings of the TENCON 2019-2019 IEEE Region 10 Conference.Kochi:IEEE,2019:25-30.
[8] ALAHI A,GOEL K,RAMANATHAN V,et al.Social LSTM:Human Trajectory Prediction in Crowded Spaces[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE,2016:961-971.
[9] KAWASAKI A,TASAKI T.Trajectory Prediction of Turning Vehicles Based on Intersection Geometry and Observed Velocities[C]//Proceedings of the 2018 IEEE Intelligent Vehicles Symposium.Changshu:IEEE,2018:511-516.
[10] XING Y,LV C,LIU Y H,et al.Hybrid-learning-based Driver Steering Intention Predictionusing Neuromuscular Dynamics[J].IEEE Transactions on Industrial Electronics,2022,69(2):1750-1761.
[11] ZHANG K P,ZHAO L,DONG C X,et al.AI-TP:Attention-based Interaction-aware Trajectory Prediction for Autonomous Driving[J].IEEE Transactions on Intelligent Vehicles,2022,8(1):73-83.
[12] CARRASCO S,LLORCA D F,SOTELO M A.SCOUT:Socially-Consistent and UndersTandable Graph Attention Network for Trajectory Prediction of Vehicles and VRUs[C]//Proceedings of the 2021 IEEE Intelligent Vehicles Symposium.Nagoya:IEEE,2021:1501-1508.
[13] DIEHL F,BRUNNER T,LE M T,et al.Graph Neural Networks for Modelling Traffic Participant Interaction[C]//Proceedings of the 2019 IEEE Intelligent Vehicles Symposium (IV).Paris:IEEE,2019:695-701.
[14] 胡翔，陈胜，何强，等.基于时空循环图神经网络的电网风险诊断框架[J].无线电工程，2024,54(9):2240-2248.
[15] CHUNG J,GULCEHRE C,CHO K H,et al.Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling[EB/OL].(2014-12-11)[2024-06-10].https://arxiv.org/abs/1412.3555.
[16] CAPOBIANCO S,MILLEFIORI L M,FORTI N,et al.Deep Learning Methods for Vessel Trajectory Prediction Based on Recurrent Neural Networks[J].IEEE Transactions on Aerospace and Electronic Systems,2021,57(6):4329-4346.
[17] 桑海峰，陈旺兴，王海峰，等.基于多模式时空交互的行人轨迹预测模型[J].电子学报，2022,50(11):2806-2812.
[18] CAESAR H,BANKITI V,LANG A H,et al.nuScenes:A Multimodal Dataset for Autonomous Driving[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle:IEEE,2020:11618-11628.
[19] LI X,YING X W,CHUAH M C.GRIP++:Enhanced Graph-based Interaction-aware Trajectory Prediction for Autonomous Driving[EB/OL].(2019-07-17)[2024-06-11].https://arxiv.org/abs/1907.07792.
[20] GU J R,SUN C,ZHAO H.Dense TNT:End-to-End Tra-jectory Prediction from Dense Goal Sets[C]//Proceedings of the 2021 IEEE International Conference on Computer Vision.Montreal:IEEE,2021:15283-15292.
[21] YUAN Y,WENG X S,OU Y L,et al.AgentFormer:Agent-aware Transformers for Socio-temporal Multi-agent Forecasting[C]//Proceedings of the 2021 IEEE International Conference on Computer Vision.Montreal:IEEE,2021:9793-9803.
[22] CHAI Y N,SAPP B,BANSAL M,et al.MultiPath:Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Prediction[EB/OL].(2019-10-12)[2024-06-11].https://arxiv.org/abs/1910.05449.
基本信息:
DOI：
中图分类号:U463.6;TP183
引用信息:
[1]李庆,韩楠,李任杰等.基于时空交互图注意力网络的多模态车辆轨迹预测模型[J].无线电工程,2025,55(02):254-263.
基金信息:
国家自然科学基金(62272066); 四川省科技计划(2025ZNSFSC0044,2025YFHZ0194,2024YFFK0413); 成都市技术创新研发项目重点项目(2024-YF08-00029-GX);成都市技术创新研发项目(2024-YF05-01217-SN); 成都市区域科技创新合作项目(2025-YF11-00031-HZ,2025-YF11-00050-HZ,2023-YF11-00020-HZ); 网络空间安全教育部重点实验室及河南省网络空间态势感知重点实验室开放基金课题(KLCS20240106); 大学生创新创业训练计划项目(202410621195,202410621183)~~