机动意图可知的结构化道路车辆轨迹预测方法研究

doi:10.19678/j.issn.1000-3428.0069627

摘要/Abstract

摘要：

准确地预测周围车辆的未来轨迹对于自动驾驶汽车(ADV)理解复杂动态环境至关重要。然而, 现有的池化策略仅依赖于欧氏坐标系表征下的历史位置特征编码, 难以有效地捕捉车辆的机动意图等隐变量特征。为此, 提出一种机动意图可知的车辆轨迹预测方法。首先, 构建基于极坐标系特征表征和高阶特征编码的池化机制, 以捕捉车辆间的相互依赖关系; 其次, 设计基于高斯概率分布的位置和加速度机动类型判别策略, 以准确地模拟结构化道路场景下车辆的预期机动; 然后, 设计基于随机采样的规划轨迹与历史轨迹耦合编码模块, 在增强模型交互特征捕获性能的同时避免冗余编码; 最后, 基于编码器-解码器框架构建轨迹预测模型StructNet, 并基于真实道路数据集NGSIM验证算法的有效性。多组对比实验和消融实验结果表明, 所提出的车辆轨迹预测模型在5 s时的均方根误差指标低于3.5 m, 相较于基准模型提升15.3%, 预测准确率得到显著提高。

关键词: 自动驾驶汽车, 极坐标系, 特征池化, 长短期记忆, 轨迹预测

Abstract:

Accurately predicting the future trajectories of surrounding vehicles is crucial for Autonomous Driving Vehicle (ADV) to understand complex dynamic environments. However, existing pooling strategies rely solely on historical position feature encoding in the Euclidean coordinate system, which fails to capture latent variables, such as vehicle maneuver intentions, effectively. To address this issue, this study proposes a vehicle trajectory prediction method that considers maneuver intentions. First, this study constructs a pooling mechanism based on polar coordinate feature representation and high-order feature encoding to capture intervehicle dependencies. Next, a position and acceleration maneuver type discrimination strategy is designed based on Gaussian probability distribution to simulate the expected maneuvers in structured road scenarios accurately. Furthermore, the study develops a trajectory planning and historical trajectory coupling encoding module based on random sampling, which enhances the interaction feature capture capabilities of the model while avoiding redundant encoding. Finally, the study builds a trajectory prediction model, StructNet, based on an encoder-decoder framework and validates the effectiveness of the algorithm on real-world road datasets from NGSIM. Multiple comparative and ablation experiments demonstrate that the proposed vehicle trajectory prediction model achieves a root mean square error of less than 3.5 m at 5 s, representing a 15.3% improvement over the baseline model, thereby significantly enhancing prediction accuracy.

Key words: Autonomous Driving Vehicle (ADV), polar coordinate system, feature pooling, Long Short-Term Memory (LSTM), trajectory prediction

颜伏伍, 王占彬, 胡杰, 徐文才, 唐郁轩, 陈楠. 机动意图可知的结构化道路车辆轨迹预测方法研究[J]. 计算机工程, 2025, 51(11): 355-365.

YAN Fuwu, WANG Zhanbin, HU Jie, XU Wencai, TANG Yuxuan, CHEN Nan. Research on Structured Road Vehicle Trajectory Prediction Method with Recognizable Maneuver Intentions[J]. Computer Engineering, 2025, 51(11): 355-365.

https://www.ecice06.com/CN/Y2025/V51/I11/355

图/表 17

图1 欧氏坐标系表征和极坐标系表征

Fig.1 Euclidean coordinate system representation and polar coordinate system representation

图2 StructNet轨迹预测框架

Fig.2 StructNet trajectory prediction framework

图3 CEM特征编码示意图

Fig.3 Schematic diagram of CEM feature encoding

图4 池化机制结构示意图

Fig.4 Schematic diagram of pooling mechanism structure

图5 多模态预期机动示意图

Fig.5 Multimodal expectation maneuvering schematic diagram

图6 不同模型对比实验效果

Fig.6 Comparison experimental effects of different models

图7 结构化道路典型机动类型对比效果

Fig.7 Comparison effects of typical motorized types on structured roads

图8 模型训练平台和部署平台示意图

Fig.8 Schematic diagram of model training platform and deployment platform

图9 在线轨迹预测框架

Fig.9 Online trajectory prediction framework

图10 RVIZ可视化结果示意图

Fig.10 Schematic diagram of RVIZ visualization result

参考文献 28

1	HUANG Y J , DU J T , YANG Z R , et al. A survey on trajectory-prediction methods for autonomous driving. IEEE Transactions on Intelligent Vehicles, 2022, 7 (3): 652- 674. doi: 10.1109/TIV.2022.3167103
2	HUANG Z Y, MO X Y, LÜ C. Multi-modal motion prediction with transformer-based neural network for autonomous driving[C]//Proceedings of International Conference on Robotics and Automation. Washington D. C., USA: IEEE Press, 2022: 2605-2611.
3	LUO C X, SUN L, DABIRI D, et al. Probabilistic multi-modal trajectory prediction with lane attention for autonomous vehicles[C]//Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. Washington D. C., USA: IEEE Press, 2020: 2370-2376.
4	赵高士, 陈龙, 蔡英凤, 等. 融合复杂网络和记忆增强网络的轨迹预测技术. 汽车工程, 2023, 45 (9): 1608-1616, 1636.
	ZHAO G S , CHEN L , CAI Y F , et al. Trajectory prediction technology integrating complex network and memory-augmented network. Automotive Engineering, 2023, 45 (9): 1608-1616, 1636.
5	陈龙, 杨晨, 蔡英凤, 等. 基于多模态特征融合的行人穿越意图预测方法. 汽车工程, 2023, 45 (10): 1779- 1790.
	CHEN L , YANG C , CAI Y F , et al. Pedestrian crossing intention prediction method based on multimodal feature fusion. Automotive Engineering, 2023, 45 (10): 1779- 1790.
6	胡启慧, 蔡英凤, 王海, 等. 基于层次图注意的异构多目标轨迹预测方法. 汽车工程, 2023, 45 (8): 1448- 1456.
	HU Q H , CAI Y F , WANG H , et al. Heterogeneous multi-object trajectory prediction method based on hierarchical graph attention. Automotive Engineering, 2023, 45 (8): 1448- 1456.
7	ALAHI A, GOEL K, RAMANATHAN V, et al. Social LSTM: human trajectory prediction in crowded spaces[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2016: 961-971.
8	DEO N, TRIVEDI M M. Convolutional social pooling for vehicle trajectory prediction[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Washington D. C., USA: IEEE Press, 2018: 1468-1476.
9	李文礼, 韩迪, 石晓辉, 等. 基于时-空注意力机制的车辆轨迹预测. 中国公路学报, 2023, 36 (1): 226- 239.
	LI W L , HAN D , SHI X H , et al. Vehicle trajectory prediction based on spatial-temporal attention mechanism. China Journal of Highway and Transport, 2023, 36 (1): 226- 239.
10	CHAI Y N, SAPP B, BANSAL M, et al. MultiPath: multiple probabilistic anchor trajectory hypotheses for behavior prediction[EB/OL]. [2024-02-10]. https://arxiv.org/abs/1910.05449v1.
11	GUPTA A, JOHNSON J, LI F F, et al. Social GAN: socially acceptable trajectories with generative adversarial networks[C]// Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 2255-2264.
12	HU Y P, ZHAN W, SUN L T, et al. Multi-modal probabilistic prediction of interactive behavior via an interpretable model[C]//Proceedings of IEEE Intelligent Vehicles Symposium. Washington D. C., USA: IEEE Press, 2019: 557-563.
13	刘占文, 李文倩, 林杉, 等. 基于稀疏权重共享的多模态轨迹预测. 中国公路学报, 2023, 36 (9): 244- 256.
	LIU Z W , LI W Q , LIN S , et al. Multimodal trajectory prediction based on sparse weight sharing. China Journal of Highway and Transport, 2023, 36 (9): 244- 256.
14	RHINEHART N, MCALLISTER R, KITANI K, et al. PRECOG: PREdiction conditioned on goals in visual multi-agent settings[C]//Proceedings of IEEE/CVF International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2019: 2821-2830.
15	SONG H R , DING W C , CHEN Y X , et al. PiP: planning-informed trajectory prediction for autonomous driving. Berlin, Germany: Springer, 2020: 598- 614.
16	LI D, ZHANG Q C, XIA Z P, et al. Planning-inspired hierarchical trajectory prediction for autonomous driving[EB/OL]. [2024-02-10]. https://arxiv.org/abs/2304.11295v1.
17	CHEN Y X, IVANOVIC B, PAVONE M. ScePT: scene-consistent, policy-based trajectory predictions for planning[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2022: 17103-17112.
18	方华珍, 刘立, 肖小凤, 等. 混合示教长短时记忆网络的车辆轨迹预测研究. 交通运输系统工程与信息, 2023, 23 (4): 80- 87.
	FANG H Z , LIU L , XIAO X F , et al. Vehicle trajectory prediction based on mixed teaching force long short-term memory. Journal of Transportation Systems Engineering and Information Technology, 2023, 23 (4): 80- 87.
19	连静, 李硕贤, 刘一荻, 等. 基于车道目标引导的车辆轨迹预测. 汽车工程, 2023, 45 (8): 1353- 1361.
	LIAN J , LI S X , LIU Y D , et al. Goal supervised attention network for vehicle trajectory prediction. Automotive Engineering, 2023, 45 (8): 1353- 1361.
20	HASAN M, SOLERNOU A, PASCHALIDIS E, et al. Maneuver-aware pooling for vehicle trajectory prediction[EB/OL]. [2024-02-10]. https://arxiv.org/abs/2104.14079v1.
21	高镇海, 鲍明喜, 高菲, 等. 基于LSTM概率多模态预期轨迹预测方法. 汽车工程, 2023, 45 (7): 1145-1152, 1162.
	GAO Z H , BAO M X , GAO F , et al. The method of probabilistic multi-modal expected trajectory prediction based on LSTM. Automotive Engineering, 2023, 45 (7): 1145-1152, 1162.
22	COLTAR J, HALKIAS J. Us highway 101 dataset[R]. [S. 1.]: Federal Highway Administration, 2007.
23	COLTAR J. Us highway I-80 dataset[R]. [S. 1.]: Federal Highway Administration, 2007.
24	DAI S Z , LI L , LI Z H . Modeling vehicle interactions via modified LSTM models for trajectory prediction. IEEE Access, 2019, 7, 38287- 38296. doi: 10.1109/ACCESS.2019.2907000
25	MESSAOUD K , YAHIAOUI I , VERROUST-BLONDET A , et al. Attention based vehicle trajectory prediction. IEEE Transactions on Intelligent Vehicles, 2020, 6 (1): 175- 185.
26	XUE Q F, LI S Y, LI X P, et al. Hierarchical motion encoder-decoder network for trajectory forecasting[EB/OL]. [2024-02-10]. https://arxiv.org/abs/2111.13324v1.
27	GAO K , LI X H , CHEN B , et al. Dual transformer based prediction for lane change intentions and trajectories in mixed traffic environment. IEEE Transactions on Intelligent Transportation Systems, 2023, 24 (6): 6203- 6216. doi: 10.1109/TITS.2023.3248842
28	LEE G. ApolloAuto[EB/OL]. [2024-02-10]. https://github.com/ApolloAuto/apollo.

[1]	闫建红, 刘芝妍, 王震. 融合时空注意力机制的多尺度卷积车辆轨迹预测[J]. 计算机工程, 2025, 51(8): 406-414.
[2]	武东辉, 王金凤, 仇森, 刘国志. 基于EWBiLSTM-ATT的数据手套手语识别[J]. 计算机工程, 2025, 51(8): 107-119.
[3]	武东辉, 王金凤, 仇森, 刘国志. 基于EWBiLSTM-ATT的数据手套手语识别[J]. 计算机工程, 2025, 51(8): 107-119.
[4]	柳大格, 游进国, 耿齐祁. 融合全局与局部语义的跨领域方面词抽取[J]. 计算机工程, 2025, 51(6): 116-126.
[5]	赖小玲, 贺嫚嫚, 胡伟, 张艺, 杜璞良, 刘蕊, 宋晓彤, 郑婷婷. 基于改进变分模态分解与深度学习的多因素电力负荷预测[J]. 计算机工程, 2025, 51(2): 375-386.
[6]	耿贞伟, 李申章, 于凤荣. 考虑空间关联的超短期、短期多区域电力负荷单步和多步预测[J]. 计算机工程, 2025, 51(10): 392-400.
[7]	陈浩, 陈珺, 刘飞. 基于自主探索的移动机器人路径规划研究[J]. 计算机工程, 2025, 51(1): 60-70.
[8]	党小超, 刘涧, 董晓辉, 祝忠彦, 李芬芳. 面向不平衡数据的机械设备故障命名实体识别[J]. 计算机工程, 2024, 50(9): 104-112.
[9]	屈潇雅, 李兵, 温立强. 面向行政执法案件文本的事件抽取研究[J]. 计算机工程, 2024, 50(9): 63-71.
[10]	张亚洲, 和玉, 戎璐, 王祥凯. 基于上下文知识增强型Transformer网络的抑郁检测[J]. 计算机工程, 2024, 50(8): 75-85.
[11]	王炼红, 林飞鹏, 李潇瑶, 谌桂枝, 周莉. 融入课程知识图谱的KMAKT预测[J]. 计算机工程, 2024, 50(7): 23-31.
[12]	刘建敏, 林晖, 汪晓丁. 基于图注意力机制的无地图场景轨迹预测方法[J]. 计算机工程, 2024, 50(7): 144-153.
[13]	周昭辰, 方清茂, 吴晓红, 胡平, 何小海. 基于MacBERT与对抗训练的机器阅读理解模型[J]. 计算机工程, 2024, 50(5): 41-50.
[14]	陈琳, 范元凯, 何震瀛, 刘晓清, 杨阳, 汤路民. SQL-to-text模型的组合泛化能力评估方法[J]. 计算机工程, 2024, 50(3): 326-335.
[15]	丁国辉, 刘宇琪, 王言开, 耿施展, 姜天昊. 基于翻转网络的低相关性序列数据预测研究[J]. 计算机工程, 2024, 50(2): 78-90.

选择文件类型/文献管理软件名称

选择包含的内容