OOD Generalization of Vehicle Trajectory Prediction Based on Causal Model of Social Interaction

HE Peng; WAN Zhifei; CUI Yaping; WU Dapeng; WANG Ruyan; PAN Mengquan

doi:10.12263/DZXB.20250827

您当前的位置：

首页 >

文章列表页 >

OOD Generalization of Vehicle Trajectory Prediction Based on Causal Model of Social Interaction

PAPERS | 更新时间：2026-07-30

- OOD Generalization of Vehicle Trajectory Prediction Based on Causal Model of Social Interaction
- ACTA ELECTRONICA SINICA Vol. 54, Issue 4, Pages: 1885-1899(2026)
- 作者机构：
  
  1.重庆邮电大学通信与信息工程学院，重庆 400065
  2.先进网络与智能互联技术重庆市高校重点实验室，重庆 400065
  3.泛在感知与智能互联重庆市重点实验室，重庆 400065
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(U24A20211;62271096);Science and Technology Research Program of Chongqing Municipal Education Commission(KJQN202500603;KJQN202300621);Natural Science Foundation of Chongqing(CSTB2025NSCQ-LZX0144;CSTB2024NSCQ-LZX0124;CSTB2023NSCQ-LZX0134);University Innovation Research Group of Chongqing(CXQT20017);Youth Innovation Group Support Program of ICE Discipline of CQUPT(SCIE-QN-2022-04);Sichuan Science and Technology Program(2024YFHZ0093)
- DOI：10.12263/DZXB.20250827
  CLC： TP391;
- Received：23 September 2025，
  
  Accepted：25 March 2026，
  
  Online First：20 May 2026，
  
  Published：25 April 2026
- 稿件说明：
移动端阅览
何鹏, 万志飞, 崔亚平, 等. 车辆轨迹预测的分布外泛化：一种社会交互因果模型框架[J]. 电子学报, 2026, 54(04): 1885-1899.

HE Peng, WAN Zhifei, CUI Yaping, et al. OOD Generalization of Vehicle Trajectory Prediction Based on Causal Model of Social Interaction[J]. Acta Electronica Sinica, 2026, 54(04): 1885-1899.
何鹏, 万志飞, 崔亚平, 等. 车辆轨迹预测的分布外泛化：一种社会交互因果模型框架[J]. 电子学报, 2026, 54(04): 1885-1899. DOI：10.12263/DZXB.20250827

HE Peng, WAN Zhifei, CUI Yaping, et al. OOD Generalization of Vehicle Trajectory Prediction Based on Causal Model of Social Interaction[J]. Acta Electronica Sinica, 2026, 54(04): 1885-1899. DOI：10.12263/DZXB.20250827

摘要

自动驾驶车辆在真实开放道路的可靠部署，高度依赖其在未知交通场景中的轨迹预测泛化能力，精准的车辆轨迹预测更是连接自动驾驶环境感知与决策规划模块的关键纽带，直接决定了自动驾驶系统的运行安全性与环境适应性。现有基于深度学习的预测方法主要依赖历史与未来轨迹间的统计相关性，其性能严重受限于训练与测试数据独立同分布（Independent and Identically Distributed，IID）的基本假设。然而，实际复杂交通环境中地图几何、道路拓扑结构等的显著差异导致数据分布偏移，IID假设难以成立，造成各类预测模型在分布外（Out-Of-Distribution，OOD）场景下性能显著下降，这成为自动驾驶系统开放道路部署的核心技术瓶颈。为应对上述挑战，本文提出一种即插即用的基于社会交互因果模型的车辆轨迹预测（Vehicle Trajectory Prediction based on Causal Model，CMVTP）框架。该框架可灵活嵌入现有主流轨迹预测模型，无需对原模型进行复杂修改或重训练，具备高度的通用性与适配性。CMVTP通过构建面向交通场景的结构因果模型（Structural Causal Model，SCM），显式刻画域不变因果特征与域可变因果特征间的内在关联，精准揭示轨迹数据中的潜在因果关系，有效提取驾驶风格、社会交互等跨环境稳定的域不变特征；同时设计因果特征学习模块实现域不变与域可变特征的显式分离，并提出双路径的不变特征转移模块，通过隐藏层特征对齐与输出分布对齐强化模型对域不变特征的提取能力，降低模型对地图特征等域可变特征的依赖。实验结果表明，CMVTP能有效提升多种主流预测模型的分布外泛化性能，在各类未见场景中，模型预测精度获得显著提升，最高可达12.75%。本文提出的CMVTP框架突破了传统基于统计相关性的轨迹预测方法在分布外场景的性能局限，为提升自动驾驶系统在开放动态环境中的适应性和安全性提供了重要技术支撑，也为车辆轨迹预测的泛化建模研究提供了新的思路与方法。

Abstract

The reliable deployment of autonomous vehicles on real-world open roads is highly dependent on their trajectory prediction generalization ability in unknown traffic scenarios. Accurate vehicle trajectory prediction serves as a key bridge between the autonomous driving environment perception and decision-making and planning modules

which directly determines the operational safety and environmental adaptability of autonomous driving system. The existing prediction methods based on deep learning mainly rely on the statistical correlation between historical and future trajectories

and their performance is severely limited by the basic assumption that training and test data are independent and identically distributed (IID). However

in complex real-world traffic environments

significant variations in map geometry and road topology lead to distribution shifts

making the IID assumption difficult to hold. Consequently

various prediction models suffer from substantial performance degradation in out-of-distribution (OOD) scenarios

which has become a critical technical bottleneck for the deployment of autonomous driving systems on open roads. To address these challenges

this paper proposes a plug-and-play vehicle trajectory prediction framework based on a causal model (CMVTP). The framework can be flexibly integrated into existing mainstream trajectory prediction models without requiring complex modifications or retraining

demonstrating strong generality and adaptability. By constructing a traffic scene-oriented structural causal model (SCM)

CMVTP explicitly characterizes the intrinsic relationships between domain-invariant and domain-variant causal features

accurately reveals the underlying causal relationships in trajectory data

and effectively extracts domain-invariant features such as driving style and social interactions. Meanwhile

a causal feature learning module is designed to explicitly disentangle domain-invariant and domain-variant features

and a dual-path invariant feature transfer module is proposed. The hidden layer feature alignment and output distribution alignment enhance the model’s ability to extract domain-invariant features and reduce its reliance on domain-variant features such as map characteristics. The experimental results show that CMVTP can effectively improve the out-of-distribution generalization performance of various mainstream prediction models. In various unseen scenarios

prediction accuracy is significantly improved

with gains of up to 12.75%. The CMVTP framework proposed in this paper breaks through the performance limitations of traditional trajectory prediction methods based on statistical correlations in out-of-distribution scenarios

provides important technical support for improving the adaptability and safety of autonomous driving systems in open dynamic environments

and offering new insights and methodologies for generalization modeling in vehicle trajectory prediction.

关键词

Keywords

references

Hakak S , Gadekallu T R , Maddikunta P K R , et al . Autonomous vehicles in 5G and beyond: A survey [J ] . Vehicular Communications , 2023 , 39 : 100551 . DOI: 10.1016/j.vehcom.2022.100551 http://dx.doi.org/10.1016/j.vehcom.2022.100551

陈建 , 苏思教 , 黄立勤 , 等 . 自动驾驶中的3D目标检测研究进展 [J ] . 电子学报 , 2025 , 53 ( 6 ): 2131 - 2156 .

Chen Jian , Su Sijiao , Huang Liqin , et al . Research advances on 3D object detection in autonomous driving [J ] . Acta Electronica Sinica , 2025 , 53 ( 6 ): 2131 - 2156 . (in Chinese)

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 , 2022 , 23 ( 1 ): 33 - 47 . DOI: 10.1109/tits.2020.3012034 http://dx.doi.org/10.1109/tits.2020.3012034

Zhang Yuxiao , Carballo A , Yang Hanting , et al . Perception and sensing for autonomous vehicles under adverse weather conditions: A survey [J ] . ISPRS Journal of Photogrammetry and Remote Sensing , 2023 , 196 : 146 - 177 . DOI: 10.1016/j.isprsjprs.2022.12.021 http://dx.doi.org/10.1016/j.isprsjprs.2022.12.021

Crosato L , Tian Kai , Shum H P H , et al . Social interaction‐aware dynamical models and decision‐making for autonomous vehicles [J ] . Advanced Intelligent Systems , 2024 , 6 ( 3 ): 2300575 . DOI: 10.1002/aisy.202300575 http://dx.doi.org/10.1002/aisy.202300575

Meng Qingyu , Guo Hongyan , Li Jialin , et al . Vehicle trajectory prediction method driven by raw sensing data for intelligent vehicles [J ] . IEEE Transactions on Intelligent Vehicles , 2023 , 8 ( 7 ): 3799 - 3812 . DOI: 10.1109/tiv.2023.3265412 http://dx.doi.org/10.1109/tiv.2023.3265412

Giannaros A , Karras A , Theodorakopoulos L , et al . Autonomous vehicles: Sophisticated attacks, safety issues, challenges, open topics, blockchain, and future directions [J ] . Journal of Cybersecurity and Privacy , 2023 , 3 ( 3 ): 493 - 543 . DOI: 10.3390/jcp3030025 http://dx.doi.org/10.3390/jcp3030025

Zhao Junwu , Qu Ting , Gong Xun , et al . Interaction-aware personalized trajectory prediction for traffic participant based on interactive multiple model [J ] . IEEE Transactions on Intelligent Vehicles , 2023 , 8 ( 3 ): 2184 - 2196 . DOI: 10.1109/tiv.2022.3207275 http://dx.doi.org/10.1109/tiv.2022.3207275

苏越阳 , 姚迪 , 毕经平 . 基于噪声标签重加权的车辆轨迹异常检测方法 [J ] . 电子学报 , 2025 , 53 ( 1 ): 182 - 192 .

Su Yueyang , Yao Di , Bi Jingping . A vehicle trajectory anomaly detection method based on noise label re-weighting [J ] . Acta Electronica Sinica , 2025 , 53 ( 1 ): 182 - 192 . (in Chinese)

Chen Yuxiao , Veer S , Karkus P , et al . Interactive joint planning for autonomous vehicles [J ] . IEEE Robotics and Automation Letters , 2024 , 9 ( 2 ): 987 - 994 . DOI: 10.1109/lra.2023.3332474 http://dx.doi.org/10.1109/lra.2023.3332474

Wang Yijing , Liu Zhengxuan , Zuo Zhiqiang , et al . Trajectory planning and safety assessment of autonomous vehicles based on motion prediction and model predictive control [J ] . IEEE Transactions on Vehicular Technology , 2019 , 68 ( 9 ): 8546 - 8556 . DOI: 10.1109/TVT.2019.2930684 http://dx.doi.org/10.1109/TVT.2019.2930684

裴炤 , 邱文涛 , 王淼 , 等 . 基于Transformer动态场景信息生成对抗网络的行人轨迹预测方法 [J ] . 电子学报 , 2022 , 50 ( 7 ): 1537 - 1547 . DOI: 10.12263/DZXB.20210762 http://dx.doi.org/10.12263/DZXB.20210762

Pei Zhao , Qiu Wentao , Wang Miao , et al . Pedestrian trajectory prediction method using dynamic scene information based transformer generative adversarial network [J ] . Acta Electronica Sinica , 2022 , 50 ( 7 ): 1537 - 1547 . (in Chinese) . DOI: 10.12263/DZXB.20210762 http://dx.doi.org/10.12263/DZXB.20210762

Deo N , Trivedi M M . Convolutional social pooling for vehicle trajectory prediction [C ] // Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) . Piscataway : IEEE , 2018 : 1549 - 15498 . DOI: 10.1109/cvprw.2018.00196 http://dx.doi.org/10.1109/cvprw.2018.00196

Ip A , Irio L , Oliveira R . Vehicle trajectory prediction based on LSTM recurrent neural networks [C ] // Proceedings of 2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring) . Piscataway : IEEE , 2021 : 1 - 5 . DOI: 10.1109/vtc2021-spring51267.2021.9449038 http://dx.doi.org/10.1109/vtc2021-spring51267.2021.9449038

李鑫 , 陆伟 , 马召祎 , 等 . 基于图注意力和改进Transformer的节点分类方法 [J ] . 电子学报 , 2024 , 52 ( 8 ): 2799 - 2810 . DOI: 10.12263/DZXB.20230515 http://dx.doi.org/10.12263/DZXB.20230515

Li Xin , Lu Wei , Ma Zhaoyi , et al . A node classification method based on graph attention and improved transformer [J ] . Acta Electronica Sinica , 2024 , 52 ( 8 ): 2799 - 2810 . (in Chinese) . DOI: 10.12263/DZXB.20230515 http://dx.doi.org/10.12263/DZXB.20230515

Zhuolei Chaochen , Zhang Qichao , Li Ding , et al . Vehicle trajectory prediction based on graph attention network [C ] // Proceedings of the 6th International Conference on Cognitive Systems and Information Processing . Heidelberg : Springer , 2021 : 427 - 438 . DOI: 10.1007/978-981-16-9247-5_33 http://dx.doi.org/10.1007/978-981-16-9247-5_33

Li Zirui , Lu Chao , Yi Yangtian , et al . A hierarchical framework for interactive behaviour prediction of heterogeneous traffic participants based on graph neural network [J ] . IEEE Transactions on Intelligent Transportation Systems , 2022 , 23 ( 7 ): 9102 - 9114 . DOI: 10.1109/tits.2021.3090851 http://dx.doi.org/10.1109/tits.2021.3090851

Zhao Ziyi , Fang Haowen , Jin Zhao , et al . GISNet: Graph-based information sharing network for vehicle trajectory prediction [C ] // Proceedings of 2020 International Joint Conference on Neural Networks (IJCNN) . Piscataway : IEEE , 2020 : 1 - 7 . DOI: 10.1109/IJCNN48605.2020.9206770 http://dx.doi.org/10.1109/IJCNN48605.2020.9206770

Vaswani A , Shazeer N , Parmar N , et al . Attention is all you need [C ] // Proceedings of the 31st International Conference on Neural Information Processing Systems . New York : Curran Associates Inc. , 2017 : 6000 - 6010 .

Hong J , Sapp B , Philbin J . Rules of the road: Predicting driving behavior with a convolutional model of semantic interactions [C ] // Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2019 : 8446 - 8454 . DOI: 10.1109/CVPR.2019.00865 http://dx.doi.org/10.1109/CVPR.2019.00865

Casas S , Gulino C , Suo S , et al . The importance of prior knowledge in precise multimodal prediction [C ] // Proceedings of 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . Piscataway : IEEE , 2020 : 2295 - 2302 . DOI: 10.1109/iros45743.2020.9341199 http://dx.doi.org/10.1109/iros45743.2020.9341199

Liu Yicheng , Zhang Jinghuai , Fang Liangji , et al . Multimodal motion prediction with stacked transformers [C ] // Proceedings of 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2021 : 7573 - 7582 . DOI: 10.48550/arXiv.2103.11624 http://dx.doi.org/10.48550/arXiv.2103.11624

Chen Xiaobo , Zhang Huanjia , Zhao Feng , et al . Intention-aware vehicle trajectory prediction based on spatial-temporal dynamic attention network for internet of vehicles [J ] . IEEE Transactions on Intelligent Transportation Systems , 2022 , 23 ( 10 ): 19471 - 19483 . DOI: 10.1109/TITS.2022.3170551 http://dx.doi.org/10.1109/TITS.2022.3170551

Wang Jindong , Lan Cuiling , Liu Chang , et al . Generalizing to unseen domains: A survey on domain generalization [J ] . IEEE Transactions on Knowledge and Data Engineering , 2023 , 35 ( 8 ): 8052 - 8072 . DOI: 10.1109/TKDE.2022.3178128 http://dx.doi.org/10.1109/TKDE.2022.3178128

Mahajan D , Tople S , Sharma A . Domain generalization using causal matching [C/OL ] // Proceedings of the 38th International Conference on Machine Learning . PMLR , 2021 : 7313 - 7324 .

Sui Yongduo , Wang Xiang , Wu Jiancan , et al . Causal attention for interpretable and generalizable graph classification [C ] // Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . New York : ACM , 2022 : 1696 - 1705 . DOI: 10.1145/3534678.3539366 http://dx.doi.org/10.1145/3534678.3539366

Chen Yongqiang , Zhang Yonggang , Bian Yatao , et al . Learning causally invariant representations for out-of-distribution generalization on graphs [C ] // Proceedings of the 36th International Conference on Neural Information Processing Systems . New York : Curran Associates Inc. , 2022 : 1608 . DOI: 10.52202/068431-1608 http://dx.doi.org/10.52202/068431-1608

Zhou Kaiyang , Liu Ziwei , Qiao Yu , et al . Domain generalization: A survey [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2023 , 45 ( 4 ): 4396 - 4415 .

Dendorfer P , Elflein S , Leal-Taixé L . MG-GAN: A multi-generator model preventing out-of-distribution samples in pedestrian trajectory prediction [C ] // Proceedings of 2021 IEEE/CVF International Conference on Computer Vision (ICCV) . Piscataway : IEEE , 2021 : 13138 - 13147 . DOI: 10.1109/iccv48922.2021.01291 http://dx.doi.org/10.1109/iccv48922.2021.01291

Liu Yuejiang , Cadei R , Schweizer J , et al . Towards robust and adaptive motion forecasting: A causal representation perspective [C ] // Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 17081 - 17092 . DOI: 10.1109/cvpr52688.2022.01657 http://dx.doi.org/10.1109/cvpr52688.2022.01657

Feng Lan , Bahari M , Amor K M B , et al . UniTraj: A unified framework for scalable vehicle trajectory prediction [C ] // Proceedings of the 18th European Conference on Computer Vision . Heidelberg : Springer , 2024 : 106 - 123 . DOI: 10.1007/978-3-031-73254-6_7 http://dx.doi.org/10.1007/978-3-031-73254-6_7

Zhang Linfeng , Bao Chenglong , Ma Kaisheng . Self-distillation: Towards efficient and compact neural networks [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2022 , 44 ( 8 ): 4388 - 4403 .

Lin Yong , Dong Hanze , Wang Hao , et al . Bayesian invariant risk minimization [C ] // Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 16000 - 16009 . DOI: 10.1109/cvpr52688.2022.01555 http://dx.doi.org/10.1109/cvpr52688.2022.01555

Mao Chaojin , Zhao Liang , Min Geyong , et al . Informative causality-based vehicle trajectory prediction architecture for domain generalization [C ] // Proceedings of GLOBECOM 2023 - 2023 IEEE Global Communications Conference . Piscataway : IEEE , 2023 : 1 - 6 . DOI: 10.1109/globecom54140.2023.10437409 http://dx.doi.org/10.1109/globecom54140.2023.10437409

Tang Xiaolong , Kan Meina , Shan Shiguang , et al . HPNet: Dynamic trajectory forecasting with historical prediction attention [C ] // Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2024 : 15261 - 15270 . DOI: 10.1109/CVPR52733.2024.01445 http://dx.doi.org/10.1109/CVPR52733.2024.01445

Zhan Wei , Sun Liting , Wang Di , et al . INTERACTION dataset: An INTERnational, adversarial and cooperative moTION dataset in interactive driving scenarios with semantic maps [PP/OL ] . V1. arXiv ( 2019-09-30 )[ 2025-09-21 ] . https://arxiv.org/abs/1910.03088 https://arxiv.org/abs/1910.03088 . DOI: 10.48550/arXiv.1910.03088 http://dx.doi.org/10.48550/arXiv.1910.03088

Halkias J , Colyar J . Interstate 80 freeway dataset [DS/OL ] . ( 2006-12 ) [ 2025-09-21 ] . https://www.fhwa.dot.gov/publications/research/operations/06137/ https://www.fhwa.dot.gov/publications/research/operations/06137/ .

Colyar J , Halkias J . US highway 101 dataset [DS/OL ] . ( 2007-01 ) [ 2025-09-21 ] . https://www.fhwa.dot.gov/publications/research/operations/07030/ https://www.fhwa.dot.gov/publications/research/operations/07030/ .

Montanino M , Punzo V . Making NGSIM data usable for studies on traffic flow theory: Multistep method for vehicle trajectory reconstruction [J ] . Transportation Research Record: Journal of the Transportation Research Board , 2013 , 2390 ( 1 ): 99 - 111 . DOI: 10.3141/2390-11 http://dx.doi.org/10.3141/2390-11

Li Xin , Ying Xiaowen , Chuah M C . GRIP: Graph-based interaction-aware trajectory prediction [C ] // Proceedings of 2019 IEEE Intelligent Transportation Systems Conference (ITSC) . Piscataway : IEEE , 2019 : 3960 - 3966 . DOI: 10.1109/itsc.2019.8917228 http://dx.doi.org/10.1109/itsc.2019.8917228

Gupta A , Johnson J , Fei-Fei L , et al . Social GAN: Socially acceptable trajectories with generative adversarial networks [C ] // Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2018 : 2255 - 2264 . DOI: 10.1109/cvpr.2018.00240 http://dx.doi.org/10.1109/cvpr.2018.00240

Cui Yaping , Wan Zhifei , Pan Mengquan , et al . Incorporating driving style in spatial-temporal transformer for vehicle trajectory prediction [J ] . Engineering Applications of Artificial Intelligence , 2026 , 171 : 114133 . DOI: 10.1016/j.engappai.2026.114133 http://dx.doi.org/10.1016/j.engappai.2026.114133

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on Mixed Architecture Hamilton-Jacobi-Bellman Proximal Policy Optimization Method for Autonomous Driving

Hierarchical Envelope Adversarial Domain Adaptation with Loose-Tight Coupled Trajectory Prediction Model

Research Advances on 3D Object Detection in Autonomous Driving

Motion Compensation Optimization Method for 3D Multi-Object Tracking

Knowledge-Guided Multi-Modal Modular Satellite Network for Space-Air-Ground Integrated Intelligent Satellite Computing

Related Author

WANG Jinqiang

SONG Lirong

ZHOU Qingguo

LI Yan

YONG Binbin

JIANG Yuanbo

LI Wen-zheng

LI Yong-ming

Related Institution

School of Information Science and Engineering, Lanzhou University

School of Microelectronics and Communication Engineering, Chongqing University

Chongqing Open University

College of Physics and Information Engineering, Fuzhou University

Fujian Key Laboratory for Intelligent Processing and Wireless Transmission of Media Information

⁰