考虑车-路交互作用的车辆轨迹预测算法研究

doi:10.3901/JME.2024.10.076

机械工程学报 ›› 2024, Vol. 60 ›› Issue (10): 76-85.doi: 10.3901/JME.2024.10.076

扫码分享

考虑车-路交互作用的车辆轨迹预测算法研究

李玖法¹, 邹博文^1,2, 任玥³

1. 西南大学人工智能学院重庆 400715;
2. 中国汽车工程研究院股份有限公司重庆 401122;
3. 西南大学工程技术学院重庆 400715

收稿日期:2023-07-20 修回日期:2023-12-25 出版日期:2024-05-20 发布日期:2024-07-24
作者简介:李玖法,男,1998年出生。主要研究方向为车辆轨迹预测和意图识别。
E-mail:ljffuture@163.com
任玥(通信作者),男,1990年出生,博士,讲师。主要研究方向为智能网联汽车决策与规划,路径跟踪控制与底盘集成控制。
E-mail:yueren@swu.edu.cn
基金资助:
重庆市技术创新与应用发展专项重点(CSTB2022TIAD-KPX0038)、重庆市自然科学基金面上(cstc2020jcyj-msxmX0496)和汽车主动安全测试技术重庆市工业和信息化重点实验室开放课题(22AKC03)资助项目。

Research on Vehicle Trajectory Prediction Considering Vehicle-lane Interaction

LI Jiufa¹, ZOU Bowen^1,2, REN Yue³

1. College of Artificial Intelligence, Southwest University, Chongqing 400715;
2. China Automotive Engineering Research Institute Co., Ltd., Chongqing 401122;
3. College of Engineering and Technology, Southwest University, Chongqing 400715

Received:2023-07-20 Revised:2023-12-25 Online:2024-05-20 Published:2024-07-24

摘要/Abstract

摘要： 准确预测动态障碍轨迹是自动驾驶车辆正确决策和精准控制的关键。考虑复杂环境中车辆运动状态受车道信息和周围车辆共同影响，基于编-解码框架，提出一种聚合车辆-车道信息的车辆轨迹预测模型。首先采用有向图表征地图车道节点，然后通过门控循环单元(Gated recurrent unit, GRU)对目标车辆和周围障碍进行融合编码，同时引入人工势场模型，表征车-车相对交互作用。再将车道节点向量和斥力向量融合，并采用注意力机制进一步挖掘编码向量时空耦合机制。最后通过策略网络对车道节点进行评分和聚类，实现障碍物多模态轨迹预测。基于nuScene轨迹预测数据集进行训练和评估，测试结果表明，相较于现有基线模型，提出的预测模型具有更低的预测误差和更好的鲁棒性。另外，将斥力场引入注意力机制使得该模型具有更好的可解释性。

关键词: 轨迹预测, 深度学习, 编码-解码框架, 人工势场, 注意力机制

Abstract: Accurate obstacle trajectory prediction is a key to correct decision making and precise control of autonomous vehicle. Considering the influence of lane information and surrounding obstacles on vehicle motion in complex environment, a vehicle trajectory prediction model aggregated vehicle-lane information is proposed based on encoding-decoding framework. Firstly, the directed graph is adopted to describe the lane nodes. Then the target vehicle and surrounding obstacles are encoded via GRU. Simultaneously, the artificial potential field is introduced to represent the vehicle-vehicle interaction. By concatenating the lane node vector and repulsive force vector, the attention mechanism is utilized to explore the spatial-temporal coupling mechanism. Finally, the multi-modal trajectory prediction of obstacles is achieved by scoring and clustering the lane nodes through the strategy network. The proposed prediction model is trained on the nuScenes motion prediction dataset and evaluated with the state-of-the-art baseline models’ performance. The results demonstrated that the proposed prediction model has lower prediction error and better robustness. On the other hand, it has better interpretability by introducing the potential field into attention mechanism.

Key words: trajectory prediction, deep learning, encoding-decoding framework, potential field, attention mechanism

中图分类号:

TG156

李玖法, 邹博文, 任玥. 考虑车-路交互作用的车辆轨迹预测算法研究[J]. 机械工程学报, 2024, 60(10): 76-85.

LI Jiufa, ZOU Bowen, REN Yue. Research on Vehicle Trajectory Prediction Considering Vehicle-lane Interaction[J]. Journal of Mechanical Engineering, 2024, 60(10): 76-85.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.2024.10.076

http://www.cjmenet.com.cn/CN/Y2024/V60/I10/76

参考文献

[1] 李睿，李春明，苏杰，等. 考虑轨迹预测补偿的履带车辆滑动参数估计方法[J]. 清华大学学报(自然科学版)，2022，62(1):133-140. LI Rui，LI Chunming，SU Jie，et al. Slid parameter estimaters for tracked vehicles with trajectory prediction compensation[J]. Journal of Tsinghua University(Science and Technology)，2022，62(1):133-140.
[2] 张润生，黄小云，刘晶，等. 基于视觉复杂环境下车辆行驶轨迹预测方法[J]. 机械工程学报，2011，47(2):16-24. ZHANG Runsheng，HUANG Xiaoyun，LIU Jing，et al. Image vehicle motion trajectory prediction method under complex environment[J]. Journal of Mechanical Engineering，2011，47(2):16-24.
[3] ALAHI A，GOEL K，RAMANATHAN V. Social-LSTM:Human trajectory prediction in crowded spaces[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR)，Las Vegas，USA，2016:961-971.
[4] SHENG Z，XU Y，XUE S，et al. Graph-based spatial-temporal convolutional network for vehicle trajectory prediction in autonomous driving[J]. IEEE Transactions on Intelligent Transportation System，2022，23(10):17654-17665.
[5] DEO N，TRIVEDI M M. Multi-modal trajectory prediction of surrounding vehicles with maneuver based LSTMs[C]//IEEE Intelligent Vehicles Symposium，Changshu ，China，2018:1179-1184.
[6] XU C，LI M，NI Z，et al. GroupNet:Multiscale hypergraph neural networks for trajectory prediction with relational reasoning[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR)，New Orleans，USA，2022:6488-6497.
[7] KAOUTHER M，ITHERI Y，ANNE V. Attention based vehicle trajectory prediction[J]. IEEE Transactions on Intelligent Vehicles，2021，6(1):175-185.
[8] HUANG Y，BI H，LI Z. STGAT:Modeling spatial-temporal interactions for human trajectory prediction[C]//IEEE International Conference on Computer Vision(ICCV)，Seoul，Republic of Korea. 2019:6271-6280.
[9] LUO C，SUN L，DABIRI D. Probabilistic multi-modal trajectory prediction with lane attention for autonomous vehicles[C]//IEEE International Conference on Intelligent Robots and Systems (IROS)，Las Vegas，USA 2020:2370-2376.
[10] GUO H，MENG Q，CAO D. Vehicle trajectory prediction method coupled with ego vehicle motion trend under dual attention mechanism[J]. IEEE Transactions on Instrumentation and Measurement，2022，71:1-16.
[11] 黄玲，郭亨聪，张荣辉，等. 人机混驾环境下基于LSTM的无人驾驶车辆换道行为模型[J]. 中国公路学报，2020，33(7):156-166. HUANG Ling，GUO Hengcong，ZHANG Ronghui，et al. LSTM-based lane-changing behavior model for unmanned vehicle under environment of heterogeneous human-driven and autonomous vehicles[J]. China Journal of Highway and Transport，2020，33(7):156-166.
[12] GAO J，SUN C，ZHAO H. VectorNet:Encoding HD maps and agent dynamics from vectorized representation[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR)，2020:11522-11530.
[13] LIANG M，YANG B，HU R. Learning lane graph representations for motion forecasting[C]//European Conference on Computer Vision(ECCV)，Glasgow，UK，2020:541-556.
[14] GU J，SUN C，ZHAO H. DenseTNT:End-to-end trajectory prediction from dense goal sets[C]//IEEE International Conference on Computer Vision (ICCV)，Montreal，Canada，2021:15283-15292.
[15] DEO N，WOLFF E，BEIJBOM O. Multimodal trajectory prediction conditioned on lane-graph traversals[C]//Proceedings of the 5th Conference on Robot Learning(PMLR)，London，UK，2022(164):203-212.
[16] 李文礼，肖凯文，石晓辉，等. 基于行人视野注意力场的人车微观交互模型[J]. 汽车工程，2022，44(6):808-820. LI Wenli，XIAO Kaiwen，SHI Xiaohui. Pedestrian-vehicle micro-interaction model based on attention field of pedestrian vision[J]. Automotive Engineering，2022，44(6):808-820.
[17] 陈无畏，胡振国，汪洪波，等. 基于可拓决策和人工势场法的车道偏离辅助系统研究[J]. 机械工程学报，2018，54(16):134-143. CHEN Wuwei，HU Zhenguo，WANG Hongbo，et al. Study on extension decision and artificial potential field based lane departure assistance system[J]. Journal of Mechanical Engineering，2018，54(16):134-143.
[18] HOLGER C，VARUN B，LANG A H，et al. NuScenes:A multimodal dataset for autonomous driving[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR)，2020:11621-11631.
[19] PHAN-MINH T，GRIGORE E C，BOULTON F A. CoverNet:Multimodal behavior prediction using trajectory sets[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR)，2020:14062-14071.
[20] SALZMANN T，IVANOVIC B，CHAKRAVARTY P. Trajectron++:Dynamically-feasible trajectory forecasting with heterogeneous data[C]//European Conference on Computer Vision(ECCV)，Glasgow，UK，2020:683-700.
[21] WANG C，WANG Y，XU M. Stepwise goal-driven networks for trajectory prediction[J]. IEEE Robotics and Automation Letters，2022，7(2):2716-2723.
[22] MESSAOUD K，DEO N，TRIVEDI M M. Trajectory prediction for autonomous driving based on multi-head attention with joint agent-map representation[C]//IEEE Intelligent Vehicles Symposium (IV)，NAGOYA，Japan，2021:165-170.
[23] DEO N，TRIVEDI M M. Trajectory forecasts in unknown environments conditioned on grid-based plans[J]. arXiv:2001.00735，2020.
[24] GILLES T，SABATINI S，TSISHKOU D. GOHOME:Graph-oriented heatmap output for future motion estimation[C]//International Conference on Robotics and Automation (ICRA)，Philadelphia，USA，2022:9107-9114.
[25] GIRGIS R，GOLEMO F，CODEVILLA F. Latent variable sequential set transformers for joint multi-agent motion prediction[C]//International Conference on Learning Representations(ICLR)，2022.

考虑车-路交互作用的车辆轨迹预测算法研究

Research on Vehicle Trajectory Prediction Considering Vehicle-lane Interaction

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王裕祥, 高阳, 蒋恩超, 董娜, 陈旭东, 彭凡, 王立闻. 基于GAF-MACNN的风机叶片健康状态监测方法[J]. 机械工程学报, 2026, 62(7): 139-149.
[2]	雷佳为, 王姝婷, 岳重祥, 林萍, 彭文, 孙杰, 张殿华. 基于深度学习的热轧厚度-活套动态建模[J]. 机械工程学报, 2026, 62(6): 142-153.
[3]	徐扬欢, 王东城, 刘宏民, 党丽莹, 刘云飞, 杨爱民. 基于深度学习的图像板形检测方法研究[J]. 机械工程学报, 2026, 62(6): 185-196.
[4]	贾康康, 都军民, 陈飞宇, 时建纬, 侯国义, 李成. 基于卷积神经网络注意力机制模块的复合材料损伤自动识别研究[J]. 机械工程学报, 2026, 62(6): 247-256.
[5]	石逸涵, 张旭, 庄存波, 刘金山, 王家修, 孙连胜. 集成双重注意力机制CNN-LSTM时空网络的离散车间生产瓶颈预测[J]. 机械工程学报, 2026, 62(5): 49-60.
[6]	唐小林, 张焜埸, 陈止戈, 杨剑英, 杨为. 融合交互预测信息的自动驾驶安全规划方法研究[J]. 机械工程学报, 2026, 62(4): 249-262.
[7]	李雨露, 李俊峰, 万章艺, 魏正英. 基于YOLO算法的激光粉末床熔融成形层形貌分类识别与成形质量预测研究[J]. 机械工程学报, 2026, 62(3): 218-234.
[8]	张志勇, 姜志昊, 胡林, 黄彩霞. 面向连续弯道的改进人工势场建模与动态避障路径规划[J]. 机械工程学报, 2026, 62(2): 271-282.
[9]	王祺, 叶仁传, 马国杰, 马佩珏, 范杰, 杨文龙. 钢材表面缺陷轻量化识别算法研究[J]. 机械工程学报, 2025, 61(8): 18-31.
[10]	秦飞红, 袁懋诞, 祝端阳, 刘晓睿, 李明, 纪轩荣. 基于多层特征融合目标检测网络的超声全聚焦钢板焊缝缺陷自动检测研究[J]. 机械工程学报, 2025, 61(4): 55-66.
[11]	陈其怀, 林添良, 马荣华, 温建和, 缪骋, 任好玲. 基于多模态的履带式挖掘机自动驾驶决策网络[J]. 机械工程学报, 2025, 61(4): 380-388.
[12]	黄思翰, 陈建鹏, 徐哲, 阎艳, 王国新. 基于大语言模型和机器视觉的智能制造系统人机自主协同作业方法[J]. 机械工程学报, 2025, 61(3): 130-141.
[13]	刘翰儒, 陈桂映, 许泽林, 彭世通, 郭嘉楠, 刘伟嵬, 王奉涛. 基于SERepVGG-A2的轻量化卷积神经网络在熔池状态识别中的应用[J]. 机械工程学报, 2025, 61(3): 440-448.
[14]	赵阳, 王中任, 史铁林, 谭孟, 张文, 李文喜. 基于Swin-IMSA的外圆磨削过程中粗糙度在机预测方法[J]. 机械工程学报, 2025, 61(23): 373-380.
[15]	林天骄, 宋浏阳, 崔玲丽, 王华庆. 非理想数据下基于时序对比注意力模型的寿命预测方法[J]. 机械工程学报, 2025, 61(18): 27-37.