| 485 | 0 | 355 |
| 下载次数 | 被引频次 | 阅读次数 |
随着自动驾驶、智能导航等领域的快速发展,对时空轨迹预测的准确性和鲁棒性的要求不断提高。传统轨迹预测方法主要依赖运动历史数据,忽略了环境中的语义信息,在复杂场景下往往难以取得理想的预测效果。对轨迹预测领域相关研究进行综述,特别是基于空间语义分析的轨迹预测研究进展。重点探讨了视觉语言模型(Vision Language Model, VLM)和大语言模型(Large Language Model, LLM)在轨迹预测方面的应用,介绍了多种基于空间语义分析的轨迹预测模型。通过实验结果分析发现,VLM和LLM能够显著提升轨迹预测的准确率。基于空间语义分析的轨迹预测方法未来将考虑多模态融合、提升模型架构、提高推理速度等方向,以进一步提升大规模轨迹预测的性能。
Abstract：With the rapid development of autonomous driving, intelligent navigation and other fields, the requirements of accuracy and robustness of spatio-temporal trajectory prediction are continuously increasing. Traditional trajectory prediction methods mainly rely on motion history data, thereby ignoring the semantic information in the context, which often makes it difficult to achieve ideal prediction results in complex scenarios. The relevant research in the field of trajectory prediction is reviewed, especially the research progress of trajectory prediction techniques based on spatial semantic analysis. The application of Visual Language Model( VLM) and Large Language Model( LLM) in trajectory prediction is mainly discussed, and a variety of trajectory prediction models based on spatial semantic analysis are introduced. Experimental results show that the VLM and LLM can significantly improve the accuracy of trajectory prediction. In the future, the trajectory prediction methods based on spatial semantic analysis will take into consideration the directions of multi-modality fusion, model architecture improvement and increasing reasoning speed to further improve the performance of largescale trajectory prediction.
[1]张志远,倪国新,徐艳国.轨迹预测技术的现状及发展综述[J].电子测量技术,2020,43(13):111-116.
[2]章璐璐,李思照.基于深度学习的智能体轨迹预测文献综述[J].无线电工程,2023,53(3):644-656.
[3]李任杰,乔少杰,李庆,等. RHTP-STD:基于时空数据库的实时启发式轨迹预测模型[J/OL].无线电通信技术,1-12[2024-10-14]. http:∥kns. cnki. net/kcms/detail/13. 1099. TN. 20240802. 1320. 008. html.
[4]李庆,乔少杰,陈浩,等.融合时空查询的情景感知多模态车辆轨迹预测模型[J/OL].无线电通信技术,1-14[2024-10-14]. http:∥kns. cnki. net/kcms/detail/13. 1099. TN. 20240806. 1033. 003. html.
[5]谢添丞,乔少杰,张桃,等.基于情景感知与移动数据挖掘的行人轨迹预测方法[J].无线电通信技术,2023,49(4):606-615.
[6]李雪松,张锲石,宋呈群,等.自动驾驶场景下的轨迹预测技术综述[J].计算机工程,2023,49(5):1-11.
[7]乔少杰,吴凌淳,韩楠,等.情景感知驱动的移动对象多模式轨迹预测技术综述[J].软件学报,2023,34(1):312-333.
[8] WU W T,JIN W Z,LIN P Q. Research on Choice of Travel Mode Model Based on Naive Bayesian Method[C]∥2011International Conference on Business Management and Electronic Information. Guangzhou:IEEE,2011:439-444.
[9] BOUBEZOUL A,KOITA A,DAUCHER D. Vehicle Trajectories Classification Using Support Vectors Machines for Failure Trajectory Prediction[C]∥2009 International Conference on Advances in Computational Tools for Engineering Applications. Beirut:IEEE,2009:486-491.
[10]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.
[11] RADFORD A,KIM J W,HALLACY C,et al. Learning Transferable Visual Models from Natural Language Supervision[EB/OL].(2021-02-26)[2024-10-14]. https:∥arxiv.org/pdf/2103. 00020v1.
[12] LU J S,BATRA D,PARIKH D,et al. Vilbert:Pretraining Task-agnostic Visiolinguistic Representations for Visionand-Language Tasks[C]∥Proceedings of the 33rd International Conference on Neural Information Processing Systems. Vancouver:Curran Associates Inc.,2019:12-23.
[13]秦涛,杜尚恒,常元元,等. Chat GPT的工作原理、关键技术及未来发展趋势[J].西安交通大学学报,2024,58(1):1-12.
[14]张显炀,朱晓宇,林浩申,等.基于高斯混合-变分自编码器的轨迹预测算法[J].计算机工程,2020,46(7):50-57.
[15]袁祎,陈光武.基于改进HMM的车辆轨迹匹配方法研究(英文)[J]. Journal of Measurement Science and Instrumentation,2024,15(2):235-243.
[16] JIANG Y D,ZHU B,YANG S,et al. Vehicle Trajectory Prediction Considering Driver Uncertainty and Vehicle Dynamics Based on Dynamic Bayesian Network[J]. IEEE Transactions on Systems,Man,and Cybernetics:Systems,2022,53(2):689-703.
[17] JU C,WANG Z,LONG C,et al. Interaction-aware Kalman Neural Networks for Trajectory Prediction[C]∥2020IEEE Intelligent Vehicles Symposium(IV). Las Vegas:IEEE,2020:1793-1800.
[18] SCHLECHTRIEMEN J,WEDEL A,HILLENBRAND J,et al. A Lane Change Detection Approach Using Feature Ranking with Maximized Predictive Power[C]∥2014IEEE Intelligent Vehicles Symposium Proceedings. Dearborn:IEEE,2014:108-114.
[19] XIE G T,GAO H B,QIAN L J,et al. Vehicle Trajectory Prediction by Integrating Physics-and Maneuver-based Approaches Using Interactive Multiple Models[J]. IEEE Transactions on Industrial Electronics, 2017, 65(7):5999-6008.
[20]乔少杰.融合机器学习的移动数据挖掘技术导读[J].无线电通信技术,2023,49(4):604-605.
[21]乔少杰,薛骐,杨国平,等.基于动态自适应时空图的多元时序预测模型[J].计算机学报,2024,47(12):2925-2937.
[22] FENG Y,YAN X. Support Vector Machine Based Lanechanging Behavior Recognition and Lateral Trajectory Prediction[J]. Computational Intelligence and Neuroscience,2022,2022:3632333.
[23] NIKHIL N,TRAN M B. Convolutional Neural Network for Trajectory Prediction[C]∥Proceedings of the European Conference on Computer Vision(ECCV)Workshops. Munich:Springer,2018:186-196.
[24] ALTCHéF,FORTELLE A D L. An LSTM Network for Highway Trajectory Prediction[C]∥2017 IEEE 20th International Conference on Intelligent Transportation Systems(ITSC). Yokohama:IEEE,2017:353-359.
[25] ZHAO Z L,KARIMZADEH M,PACHECO L,et al. Mobility Management with Transferable Reinforcement Learning Trajectory Prediction[J]. IEEE Transactions on Network and Service Management,2020,17(4):2102-2116.
[26] HE C Z,CHEN L P,XU L M,et al. IRLSOT:Inverse Reinforcement Learning for Scene-oriented Trajectory Prediction[J]. IET Intelligent Transport Systems,2022,16(6):769-781.
[27] NAYAKANTI N, AL-RFOU R, ZHOU A, et al.Wayformer:Motion Forecasting via Simple&Efficient Attention Networks[C]∥2023 IEEE International Conference on Robotics and Automation(ICRA). London:IEEE,2023:2980-2987.
[28]王挺进,赵永威,李弼程.基于显著图加权视觉语言模型的图像分类方法[J].计算机工程,2015,41(3):204-210.
[29] YU W W,LIU Y L,WEI H,et al. Turning a CLIP Model into a Scene Text Detector[C]∥2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Vancouver:IEEE,2023:6978-6988.
[30] KUO Y L,HUANG X,BARBU A,et al. Trajectory Prediction with Linguistic Representations[C]∥2022 International Conference on Robotics and Automation(ICRA).Philadelphia:IEEE,2022:2868-2875.
[31] LONG K K,SHI H T,LIU J X,et al. VLM-MPC:Vision Language Foundation Model(VLM)-guided Model Predictive Controller(MPC)for Autonomous Driving[EB/OL].(2024-08-09)[2024-10-14]. https:∥arxiv. org/abs/2408. 04821v2.
[32] BAE I,LEE J,JEON H G. Can Language Beat Numerical Regression? Language-based Multimodal Trajectory Prediction[C]∥Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Seattle:IEEE,2024:753-766.
[33] JOULIN A,GRAVE E,BOJANOWSKI P,et al. Bag of Tricks for Efficient Text Classification[EB/OL].(2016-07-06)[2024-10-14]. https:∥arxiv. org/abs/1607. 01759.
[34] BAI S J,KOLTER J Z,KOLTUN V. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling[EB/OL].(2018-03-04)[2024-10-14]. https:∥arxiv. org/abs/1803. 01271.
[35] GREFF K, SRIVASTAVA R K, KOUTNíK J, et al.LSTM:A Search Space Odyssey[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016,28(10):2222-2232.
[36] VASWANI A,SHAZEER N,PARMAR N,et al. Attention Is All You Need[EB/OL].(2017-06-12)[2024-10-14]. https:∥arxiv. org/abs/1706. 03762.
[37] DEVLIN J,CHANG M W,LEE K,et al. BERT:Pre-training of Deep Bidirectional Transformers for Language Understanding[EB/OL].(2018-10-11)[2024-10-14].https:∥arxiv. org/abs/1810. 04805.
[38]罗文,王厚峰.大语言模型评测综述[J].中文信息学报,2024,38(1):1-23.
[39] CHEN X D,PENG M X,TIU P H,et al. Gen Follower:Enhancing Car-following Prediction with Large Language Models[EB/OL].(2024-07-08)[2024-10-14].https:∥arxiv. org/abs/2407. 05611.
[40] DUAN Y Q,ZHANG Q,XU R J. Prompting Multi-modal Tokens to Enhance End-to-End Autonomous Driving Imitation Learning with LLMs[EB/OL].(2024-04-07)[2024-10-14]. https:∥arxiv. org/abs/2404. 04869.
[41] PENG M X,GUO X S,CHEN X D,et al. LC-LLM:Explainable Lane-change Intention and Trajectory Predictions with Large Language Models[EB/OL].(2024-03-27)[2024-10-14]. https:∥arxiv. org/abs/2403. 18344.
[42] MOON S,WOO H,PARK H,et al. Vision Trap:VisionAugmented Trajectory Prediction Guided by Textual Descriptions[EB/OL].(2024-07-17)[2024-10-14].https:∥arxiv. org/abs/2407. 12345.
[43] TIAN X Y,GU J R,LI B L,et al. Drive VLM:The Convergence of Autonomous Driving and Large Vision-Language Models[EB/OL].(2024-02-19)[2024-10-14].https:∥arxiv. org/abs/2402. 12289.
[44] WU D M,HAN W C,WANG T C,et al. Language Prompt for Autonomous Driving[EB/OL].(2023-09-08)[2024-10-14]. https:∥arxiv. org/abs/2309. 04379.
[45] KOOPMAN P,WAGNER M D. Autonomous Vehicle Safety:An Interdisciplinary Challenge[J]. IEEE Intelligent Transportation Systems Magazine,2017,9(1):90-96.
[46] COIFMAN B,LI L Z. A Critical Evaluation of the Next Generation Simulation(NGSIM)Vehicle Trajectory Dataset[J]. Transportation Research Part B:Methodological,2017,105(4):362-377.
[47] KRAJEWSKI R,BOCK J,KLOEKER L,et al. The Highd Dataset:A Drone Dataset of Naturalistic Vehicle Trajectories on German Highways for Validation of Highly Automated Driving Systems[C]∥2018 21st International Conference on Intelligent Transportation Systems(ITSC).Maui:IEEE,2018:2118-2125.
[48] ZHU P F,WEN L Y,DU D W,et al. Detection and Tracking Meet Drones Challenge[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 44(11):7380-7399.
[49] CAESAR H,BANKITI V,LANG A H,et al. nuScenes:A Multimodal Dataset for Autonomous Driving[C]∥2020IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Seattle:IEEE,2020:11618-11628.
[50] HOUSTON J, ZUIDHOF G, BERGAMINI L, et al. One Thousand and One Hours:Self-driving Motion Prediction Dataset[EB/OL].(2020-06-25)[2024-10-14].https:∥arxiv. org/abs/2006. 14480v2.
[51] HUANG X Y,CHENG X J,GENG Q C,et al. The Apol-loscape Dataset for Autonomous Driving[C]∥2018 IEEE/CVF Conference on Computer Vision and Pattern Recogni-tion Workshops(CVPRW). Salt Lake City:IEEE,2018:954-960.
[52] CHANG M F,LAMBERT J,SANGKLOY P,et al. Argoverse:3D Tracking and Forecasting with Rich Maps[C]∥2019IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Long Beach:IEEE,2019:8740-8749.
[53] ZHAN W,SUN L T,WANG D,et al. Interaction Dataset:An International,Adversarial and Cooperative Motion Dataset in Interactive Driving Scenarios with Semantic Maps[EB/OL].(2019-09-30)[2024-10-14]. https:∥arxiv. org/abs/1910. 03088.
[54] XUE J,FANG J,LI T,et al. BLVD:Building a Large-scale5D Semantics Benchmark for Autonomous Driving[C]∥2019 International Conference on Robotics and Automa-tion(ICRA). Montreal:IEEE,2019:6685-6691.
[55] BOCK J,KRAJEWSKI R,MOERS T,et al. The in D Data-set:A Drone Dataset of Naturalistic Road User Trajecto-ries at German Intersections[C]∥2020 IEEE Intelligent Vehicles Symposium(IV). Las Vegas:IEEE,2020:1929-1934.
[56] KRAJEWSKI R, MOERS T, BOCK J, et al. The Round Dataset:A Drone Dataset of Road User Trajectories at Roundabouts in Germany[C]∥2020 IEEE 23rd Interna-tional Conference on Intelligent Transportation Systems(ITSC). Rhodes:IEEE,2020:1-6.
[57]王雨露,张龑,彭乾.基于知识学习的行人轨迹预测方法综述[J].湖北大学学报(自然科学版),2024,46(1):33-44.
[58] GUPTA A,JOHNSON J,LI F F,et al. Social GAN:Social-ly Acceptable Trajectories with Generative Adversarial Networks[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE,2018:2255-2264.
基本信息:
DOI：
中图分类号:TP18;TP391.1
引用信息:
[1]杨博渊,张力航,李成等.基于空间语义分析的轨迹预测技术综述[J].无线电工程,2025,55(03):611-620.
基金信息:
国家自然科学基金(62272066); 四川省科技计划(2023YFG0027,2024YFFK0413); 成都市技术创新研发项目(2024-YF05-01217-SN);成都市技术创新研发项目重点项目(2024-YF08-00029-GX); 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); 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); 大学生创新创业训练计划项目(202410621195,202410621183)~~