Heterogeneous Agent Trajectory Prediction Method Based on Safety-Enhanced Interaction Modeling
LI Xixi1,2, REN Minglun1,2
1. School of Management and Intelligence Science, Hefei University of Technology, Hefei 230009; 2. Key Laboratory of Process Optimization and Intelligent Decision Making, Ministry of Education, Hefei University of Technology, Hefei 230009
Abstract：In complex scenarios such as urban unsignalized intersections, autonomous vehicles interact intensively with heterogeneous agents. Synchronous trajectory prediction of heterogeneous agents is crucial for driving safety and efficiency. Existing methods neglect dynamic risk factors during interactions, resulting in insufficient prediction safety and accuracy. A heterogeneous agent trajectory prediction method based on safety-enhanced interaction modeling(SEIM) is proposed. Interaction risk, heterogeneous interaction characteristics, road environment, and individual dynamics are jointly considered. In the risk perception layer, interaction risks are characterized via the safety potential field and interaction risk features are extracted. In the heterogeneous interaction layer, two heterogeneous graphs are constructed and a heterogeneous interaction encoder is designed to capture heterogeneous interaction features. Safety potential field-based risk values are introduced as edge attributes of the safety- enhanced directed edge-feature heterogeneous graph to enhance edge features of the graph structure. In the road environment layer, road structure features are extracted from the directed graph. In the individual dynamic layer, individual motion features are captured. The four types of features are fused and fed into trajectory decoders to predict future trajectories of heterogeneous agents. Experiments on the inD dataset demonstrate the effectiveness and accuracy of SEIM in heterogeneous agent trajectory prediction.
[1] 肖尧,杨一箭,苟超.驾驶场景下基于场景图嵌入融合光流特征的风险识别方法[J]. 模式识别与人工智能, 2025, 38(8): 714-726. (Xiao Y, Yang Y J, Gou C.Risk identification in driving scenarios via fusion of scene graph embeddings and optical flow features[J]. Pattern Recognition and Artificial Intelligence, 2025, 38(8): 714-726.) [2] Li Z R, Gong J W, Zhang Z Y, et al. Interactive behavior modeling for vulnerable road users with risk-taking styles in urban scenarios: a heterogeneous graph learning approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(8): 8538-8555. [3] Bharilya V, Kumar N.Machine learning for autonomous vehicle's trajectory prediction: a comprehensive survey, challenges, and future research directions[J/OL]. Vehicular Communications, 2024, 46. https://doi.org/10.1016/j.vehcom.2024.100733. [4] 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), 2023, 53(2): 689-703. [5] Alahi A, Goel K, Ramanathan V, et al. Social LSTM: human tra-jectory prediction in crowded spaces[C]//Proceedings of the 29th IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 961-971. [6] Agrim G, Justin J, Li F F, et al. Social GAN: socially acceptable trajectories with generative adversarial networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 2255-2264. [7] Zhang L T, Wei X L, Li H C, et al. Social-TMAGRU: pedestrian trajectory prediction via temporal social attention in crowded scenes[J]. IEEE Transactions on Computational Social Systems, 2026, 13(2): 1280-1292. [8] Li X, Ying X W, Chuah M C.GRIP++: enhanced graph-based interaction-aware trajectory prediction for autonomous driving[EB/OL]. [2026-05-17]. https://arxiv.org/pdf/1907.07792. [9] 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[J]. 汽车工程, 2023, 45(8): 1448-1456. (Hu Q H, Cai Y F, Wang H, et al. Heterogeneous multi-agent trajectory prediction method based on hierarchical graph attention[J]. Automotive Engineering, 2023, 45(8): 1448-1456.) [10] Mo X Y, Xing Y, Lü C.ReCoG: a deep learning framework with heterogeneous graph for interaction-aware trajectory prediction[EB/OL]. [2026-05-17]. https://arxiv.org/pdf/2012.05032. [11] Mo X Y, Huang Z Y, 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. [12] Ning N W, Tian S H, Li H J, et al. Heterogeneous agents trajectory prediction with dynamic interaction relational reasoning[J/OL]. Neurocomputing Heterogeneous agents trajectory prediction with dynamic interaction relational reasoning[J/OL]. Neurocomputing, 2025, 647. https://doi.org/10.1016/j.neucom. 2025.130543. [13] Zhang S R, Jiang Z M, Sang M, et al. A comprehensive survey on driving compliance assessment methodologies for autonomous vehicles[J]. IEEE Intelligent Transportation Systems Magazine, 2025, 17(4): 56-82. [14] Fang J W, Zhu C, Zhang P, et al. Heterogeneous trajectory forecasting via risk and scene graph learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(11): 12078-12091. [15] Sun S Y, Wang C Y, Xiao B, et al. Heterogeneous multi-agent risk-aware graph encoder with continuous parameterized decoder for autonomous driving trajectory prediction[J/OL]. Electronics Heterogeneous multi-agent risk-aware graph encoder with continuous parameterized decoder for autonomous driving trajectory prediction[J/OL]. Electronics, 2025, 14(1). https://doi.org/10.3390/electronics14010105. [16] 王建强,吴剑,李洋.基于人-车-路协同的行车风险场概念、原理及建模[J]. 中国公路学报, 2016, 29(1): 105-114. (Wang J Q, Wu J, Li Y.Concept, principle and modeling of driving risk field based on driver-vehicle-road interaction[J]. China Journal of Highway and Transport, 2016, 29(1): 105-114.) [17] Wang J Q, Wu J, Zheng X J, et al. Driving safety field theory modeling and its application in pre-collision warning system[J]. Transportation Research Part C(Emerging Technologies), 2016, 72: 306-324. [18] Li L H, Gan J, Ji X K, et al. Dynamic driving risk potential field model under the connected and automated vehicles environment and its application in car-following modeling[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(1): 122-141. [19] Liu X L, Wang Y F, Jiang K, et al. Interactive trajectory prediction using a driving risk map-integrated deep learning method for surrounding vehicles on highways[J]. IEEE Transactions on Inte-lligent Transportation Systems, 2022, 23(10): 19076-19087. [20] Li X X, Ren M L, Xu H M.Multimodal vehicle trajectory prediction at urban uncontrolled intersections considering multiple types of traffic participants and driving risks[J]. Alexandria Engineering Journal, 2026, 135: 194-208. [21] Geng M S, Cai Z E, Zhu Y Z, et al. Multimodal vehicular trajectory prediction with inverse reinforcement learning and risk aversion at urban unsignalized intersections[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(11): 12227-12240. [22] Bock J, Krajewski R, Moers T, et al. The inD dataset: a drone dataset of naturalistic road user trajectories at German intersections[C]//Proceedings of the IEEE Intelligent Vehicles Symposium. Washington, USA: IEEE, 2020: 1929-1934. [23] Wang X, Ji H Y, Shi C, et al. Heterogeneous graph attention network[C]//Proceedings of the World Wide Web Conference. New York, USA: ACM, 2019: 2022-2032. [24] Chai Y N, Sapp B, Bansal M, et al. MultiPath: multiple probabilistic anchor trajectory hypotheses for behavior prediction//Proceedings of the Conference on Robot Learning. San Diego, USA: JMLR, 2020: 86-99. [25] Gu J R, Sun C, Zhao H.DenseTNT: end-to-end trajectory prediction from dense goal sets[EB/OL]. [2026-05-17]. https://arxiv.org/pdf/2108.09640. [26] Jia X S, Wu P H, Chen L, et al. HDGT: heterogeneous driving graph transformer for multi-agent trajectory prediction via scene encoding[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(11): 13860-13875.
2026, Vol. 39 
