基于层次图注意的异构多目标轨迹预测方法

doi:10.19562/j.chinasae.qcgc.2023.08.015

摘要/Abstract

摘要：

有效预测周边多目标的未来轨迹是智能汽车决策和运动规划成功的关键。大多数现有研究考虑车辆个体行为之间的成对交互关系，而忽略异构交通参与者之间不同的反应模式和其他场景因素对预测的影响，使得预测轨迹的合理性低，影响运动控制的安全性。鉴于此，本文提出了一种基于层次图注意的异构多目标轨迹预测方法HGATP，首先创新性地构建类别-目标-车道的3层次图，并使用GRU和GCN分别对不同类型的目标进行独立的类别编码，捕捉不同类别的特征；其次，为强化异构目标交互图的边缘表示，构建层次图注意力机制分别获取类别和类别之间的交互以及目标和车道之间的交互，实现异构多目标间高效交互和共享地图；最后，基于目标轨迹信息和区域的车道信息构建预测网络预测多目标的轨迹。为评估模型性能，分别在INTERACTION和nuScenes数据集上进行实验。实验表明，所提模型在nuScenes数据集上单目标轨迹输出的平均误差损失和最终位移损失均减小了20%以上，在INTERACTION数据集上多目标轨迹输出的ADE损失效果较基线方法减小了2 m误差，提升了复杂道路结构下车辆行驶轨迹预测的合理性。

关键词: 智能汽车, 轨迹预测, 异构多目标, 层次图交互

Abstract:

Effectively predicting the future trajectories for surrounding multiple targets is critical to the success of autonomous vehicle decision-making and motion planning. Most existing studies consider pairwise interactions between individual vehicle behaviors， while ignoring the influence of different reaction patterns among heterogeneous traffic participants and other scene factors on prediction， which reduces the rationality of the predicted trajectories and affects the safety of motion control. In view of this， this paper proposes a heterogeneous multi-target trajectory prediction method HGATP based on hierarchical graph attention. Firstly， a category-target-lane three-level graph is innovatively constructed， and different types of targets are independently coded with categories using GRU and GCN respectively to capture the features of different categories. Secondly， to enhance the edge representation of the heterogeneous target interaction graph， the attention mechanism of hierarchical graph is constructed to separately capture the interaction between categories and categories and the interaction between targets and lanes so as to achieve efficient interaction and sharing of maps among heterogeneous multiple targets. Finally， a prediction network is constructed to predict the trajectories of multiple targets based on the target trajectory information and the lane information of the region. To evaluate the performance of the model， experiments are conducted on the INTERACTION and nuScenes datasets respectively. The experiments show that the proposed model reduces the average displacement error and final displacement error of single-target trajectory output on the nuScenes dataset by more than 20%， with the ADE loss effect of multi-target trajectory output on the INTERACTION dataset reduced by 2 m error compared with the baseline method， which improves the reasonableness of vehicle trajectory prediction under complex road structures.

Key words: intelligent vehicles, trajectory prediction, heterogeneous multi-objective, hierarchical graph interaction

胡启慧,蔡英凤,王海,陈龙,董钊志,刘擎超. 基于层次图注意的异构多目标轨迹预测方法[J]. 汽车工程, 2023, 45(8): 1448-1456.

Qihui Hu,Yingfeng Cai,Hai Wang,Long Chen,Zhaozhi Dong,Qingchao Liu. Heterogeneous Multi-object Trajectory Prediction Method Based on Hierarchical Graph Attention[J]. Automotive Engineering, 2023, 45(8): 1448-1456.

图/表 8

图1

图2

图3

表1

表2

图4

图5

图6

参考文献 25

1	LEFÈVRE S， LAUGIER C， IBAÑEZ-GUZMÁN J. Exploiting map information for driver intention estimation at road intersections［C］. 2011 IEEE Intelligent Vehicles Symposium （iv）. IEEE， 2011： 583-588.
2	AOUDE G S， DESARAJU V R， STEPHENS L H， et al. Behavior classification algorithms at intersections and validation using naturalistic data［C］. 2011 IEEE Intelligent Vehicles Symposium （iv）. IEEE， 2011： 601-606.
3	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， 2016： 961-971.
4	GUPTA A， JOHNSON J， FEI-FEI L， et al. Social GAN： socially acceptable trajectories with generative adversarial networks［C］.Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition， 2018： 2255-2264.
5	BI H， FANG Z， MAO T， et al. Joint prediction for kinematic trajectories in vehicle-pedestrian-mixed scenes［C］. Proceedings of the IEEE/CVF International Conference on Computer Vision， 2019： 10383-10392.
6	JAIN A， ZAMIR A R， SAVARESE S， et al. Structural-RNN： deep learning on spatio-temporal graphs［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition， 2016： 5308-5317.
7	LI X， YING X， CHUAH M C. Grip++： enhanced graph-based interaction-aware trajectory prediction for autonomous driving［J］. arXiv preprint arXiv：， 2019.
8	YU C， MA X， REN J， et al. Spatio-temporal graph transformer networks for pedestrian trajectory prediction［C］. Computer Vision-ECCV 2020： 16th European Conference， Glasgow， UK， August 23-28， 2020， Proceedings， Part XII 16. Springer International Publishing， 2020： 507-523.
9	VASWANI A， SHAZEER N， PARMAR N， et al. Attention is all you need［J］. Advances in Neural Information Processing Systems， 2017， 30.
10	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.
11	GU J， SUN C， ZHAO H. Densetnt： end-to-end trajectory prediction from dense goal sets［C］. Proceedings of the IEEE/CVF International Conference on Computer Vision， 2021： 15303-15312.
12	LIANG M， YANG B， HU R， et al. Learning lane graph representations for motion forecasting［C］. Computer Vision-ECCV 2020： 16th European Conference， Glasgow， UK， August 23-28， 2020， Proceedings， Part II 16. Springer International Publishing， 2020： 541-556.
13	CHANDRA R， BHATTACHARYA U， BERA A， et al. Traphic： trajectory prediction in dense and heterogeneous traffic using weighted interactions［C］. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition， 2019： 8483-8492.
14	MA Y， ZHU X， ZHANG S， et al. Trafficpredict： trajectory prediction for heterogeneous traffic-agents［J］. Proceedings of the AAAI Conference on Artificial Intelligence， 2019， 33（1）： 6120-6127.
15	CARRASCO S， LLORCA D F， SOTELO M A. Scout： socially-consistent and understandable graph attention network for trajectory prediction of vehicles and vrus［C］. 2021 IEEE Intelligent Vehicles Symposium （IV）. IEEE， 2021： 1501-1508.
16	ZHENG F， WANG L， ZHOU S， et al. Unlimited neighborhood interaction for heterogeneous trajectory prediction［C］. Proceedings of the IEEE/CVF International Conference on Computer Vision， 2021： 13168-13177.
17	XU K， BA J， KIROS R， et al. Show， attend and tell： neural image caption generation with visual attention［C］. International Conference on Machine Learning. PMLR， 2015： 2048-2057.
18	CAI Y， WANG Z， WANG H， et al. Environment-attention network for vehicle trajectory prediction［J］. IEEE Transactions on Vehicular Technology， 2021， 70（11）： 11216-11227.
19	张雪翔，吴训成，史训昂，等. LSTGHP：基于分层时空图的异构代理轨迹分布预测［J］. 软件， 2020， 41（9）： 36-42.
	ZHANG Xuexiang， WU Xuncheng， SHI Xunang， et al. LSTGHP： heterogeneous agent trajectory distribution prediction based on layered spatio-temporal graph［J］. Computer Engineering & Software，2020，41（9）：36-42.
20	MO X， HUANG Z， XING Y， et al. Multi-agent trajectory prediction with heterogeneous edge-enhanced graph attention network［J］. IEEE Transactions on Intelligent Transportation Systems， 2022， 23（7）： 9554-9567.
21	CAESAR H， BANKITI V， LANG A H， et al. Nuscenes： a multimodal dataset for autonomous driving［C］. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition， 2020： 11621-11631.
22	ZHAN W， SUN L， WANG D， et al. Interaction dataset： an international， adversarial and cooperative motion dataset in interactive driving scenarios with semantic maps［J］. arXiv preprint arXiv：， 2019.
23	SALZMANN T， IVANOVIC B， CHAKRAVARTY P， et al. Trajectron++： dynamically-feasible trajectory forecasting with heterogeneous data［C］. Computer Vision-ECCV 2020： 16th European Conference， Glasgow， UK， August 23-28， 2020， Proceedings， Part XVIII 16. Springer International Publishing， 2020： 683-700.
24	WANG C， WANG Y， XU M， et al. Stepwise goal-driven networks for trajectory prediction［J］. IEEE Robotics and Automation Letters， 2022， 7（2）： 2716-2723.
25	KIM B D， PARK S H， LEE S， et al. Lapred： lane-aware prediction of multi-modal future trajectories of dynamic agents［C］. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition， 2021： 14636-14645.

[1]	郝建平,苏炎召,钟志华,黄晋. 面向智能汽车的SOA架构及服务调度机制研究[J]. 汽车工程, 2023, 45(9): 1563-1572.
[2]	赵高士,陈龙,蔡英凤,廉玉波,王海,刘擎超,滕成龙. 融合复杂网络和记忆增强网络的轨迹预测技术[J]. 汽车工程, 2023, 45(9): 1608-1616.
[3]	连静,李硕贤,刘一荻,杨东方,李琳辉. 基于车道目标引导的车辆轨迹预测[J]. 汽车工程, 2023, 45(8): 1353-1361.
[4]	高镇海, 鲍明喜, 高菲, 唐明弘. 基于LSTM概率多模态预期轨迹预测方法[J]. 汽车工程, 2023, 45(7): 1145-1152.
[5]	韩勇,林旭洁,黄红武,蔡鸿瑜,罗金镕,李燕婷. 典型汽车碰撞事故场景中行人运动轨迹预测方法[J]. 汽车工程, 2023, 45(6): 1022-1030.
[6]	高翔,陈龙,王歆叶,熊晓夏,李祎承,陈月霞. 基于轨迹预测的智能汽车行驶风险评估方法[J]. 汽车工程, 2023, 45(4): 588-597.
[7]	胡杰,朱琪,陈锐鹏,张敏超,张志豪,刘昊岩. 引入必经点约束的智能汽车全局路径规划[J]. 汽车工程, 2023, 45(3): 350-360.
[8]	张紫微,郑玲,李以农,乔旭强,郑浩,王戡. 考虑前车运动不确定性的多目标自适应巡航控制[J]. 汽车工程, 2023, 45(3): 361-371.
[9]	刘正发,吴亚,刘佩根,顾荣琦,陈广. 基于特征和标签联合分布匹配的智能驾驶跨域自适应目标检测[J]. 汽车工程, 2023, 45(11): 2082-2091.
[10]	赵树廉,来飞,李克强,陈涛,孟璋劼,唐逸超,吴思宇,田浩东. 基于数字孪生技术的智能汽车测试方法研究[J]. 汽车工程, 2023, 45(1): 42-51.
[11]	邵文博,李骏,张玉新,王红. 智能汽车预期功能安全保障关键技术[J]. 汽车工程, 2022, 44(9): 1289-1304.
[12]	汪梓豪,蔡英凤,王海,陈龙,熊晓夏. 基于单目视觉运动估计的周边多目标轨迹预测方法[J]. 汽车工程, 2022, 44(9): 1318-1326.
[13]	梁旺,秦兆博,陈亮,边有钢,胡满江. 基于改进BP神经网络的智能车纵向控制方法[J]. 汽车工程, 2022, 44(8): 1162-1172.
[14]	赵健,宋东鉴,朱冰,吴杭哲,韩嘉懿,刘宇翔. 数据机理混合驱动的交通车意图识别方法[J]. 汽车工程, 2022, 44(7): 997-1008.
[15]	余嘉星,Aliasghar Arab,裴晓飞,过学迅. 考虑路径平滑性和避撞稳定性的智能汽车弯道轨迹规划研究[J]. 汽车工程, 2022, 44(5): 656-663.