郭亚男,王晨腾,张本奎,常颖,刘志哲,曹林,杜康宁

• 1:北京信息科技大学信息与通信工程学院
• 2:信息与通信系统信息产业部重点实验室
• 3:目标认知与应用技术国家级重点实验室
• 4:中国科学院空天信息创新研究院

摘要(Abstract)：

船舶轨迹预测对保障海上安全、优化交通管理和实现智能决策具有重要意义。现有方法往往受限于数据多样性不足及历史信息利用不充分，难以应对复杂动态的海洋环境。为此，提出一种时域检索与频域扰动双域联合增强的船舶轨迹预测方法。该方法通过频率扰动扩充样本多样性，并借鉴相似历史轨迹序列增强模型对长时依赖的表征能力，在保持高精度的同时，显著提升了预测模型的泛化性和稳健性。在多个真实船舶轨迹数据集上的实验结果表明，所提方法在预测精度和稳定性方面均优于LSTM、BiLSTM、DLinear、Transformer、Informer及iTransformer等主流时间序列预测模型，验证了该方法在船舶轨迹预测任务中的有效性与优越性。

关键词(KeyWords)： 轨迹预测;检索增强生成;数据增广;自动识别系统

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(U20A20163,62201066);; 目标认知与应用技术重点实验室开放基金(2023-CXPT-LC-005);; 北京市自然科学基金项目(4264103)

作者(Author): 郭亚男,王晨腾,张本奎,常颖,刘志哲,曹林,杜康宁

DOI: 10.16508/j.cnki.11-5866/n.2026.01.001

参考文献(References)：

• [1]SUO Y F,CHEN W K,CLARAMUNT C,et al. A ship trajectory prediction framework based on a recurrent neural network[J].Sensors,2020,20(18):5133.
• [2]WANG C,REN H X,LI H J. Vessel trajectory prediction based on AIS data and bidirectional GRU[C]//2020 International Conference on Computer Vision,Image and Deep Learning(CVIDL). New York, USA:IEEE,2020:260-264.
• [3]TANG H,YIN Y, SHEN H L. A model for vessel trajectory prediction based on long short-term memory neural network[J].Journal of Marine Engineering&Technology,2022,21(3):136-145.
• [4]PARK J,JEONG J,PARK Y. Ship trajectory prediction based on Bi-LSTM using spectral-clustered AIS data[J]. Journal of Marine Science and Engineering,2021,9(9):1037.
• [5]WU W X,CHEN P F,CHEN L Y,et al. Ship trajectory prediction:an integrated approach using ConvLSTM-based sequence-to-sequence model[J]. Journal of Marine Science and Engineering,2023,11(8):1484.
• [6]NGUYEN D D,LE VAN C,ALI M I. Vessel trajectory prediction using sequence-to-sequence models over spatial grid[C]//Proceedings of the 12th ACM International Conference on Distributed and Event-based Systems. New York, USA:Association for Computing Machinery,2018:258-261.
• [7]古英汉，王峰.基于Transformer的海上船舶轨迹预测方法[J].舰船电子工程，2024,44(6):36-40.GU Y H,WANG F. Marine ship trajectory prediction method based on Transformer[J]. Ship Electronic Engineering,2024,44(6):36-40.(in Chinese)
• [8]韩子堯，陈晓慧，张然，等.基于因果长时序混合模型的船舶航迹长期预测方法[J].信息工程大学学报，2025,26(6):740-747.HAN Z Y,CHEN X H,ZHANG R,et al. Long-term prediction method for ship trajectories based on a causal long-sequence hybrid model[J]. Journal of Information Engineering University,2025,26(6):740-747.(in Chinese)
• [9]ZHAO J S,YAN Z W,CHEN X Q,et al. k-GCN-LSTM:a khop graph convolutional network and long-short-term memory for ship speed prediction[J]. Physica A:Statistical Mechanics and its Applications,2022,606:128107.
• [10]WANG S W,LI Y,XING H,et al. Vessel trajectory prediction based on spatio-temporal graph convolutional network for complex and crowded sea areas[J]. Ocean Engineering,2024,298:117232.
• [11]ZHANG Z Y,NI G X,XU Y G. Ship trajectory prediction based on LSTM neural network[C]//2020 IEEE 5th Information Technology and Mechatronics Engineering Conference(ITOEC). New York, USA:IEEE,2020:1356-1364.
• [12]MOOTHA S,SRIDHAR S,SEETHARAMAN R,et al. Stock price prediction using bi-directional LSTM based sequence to sequence modeling and multitask learning[C]//2020 11th IEEE Annual Ubiquitous Computing, Electronics&Mobile Communication Conference(UEMCON). New York, USA:IEEE,2020:78-86.
• [13]ZENG A L,CHEN M X,ZHANG L,et al. Are Transformers effective for time series forecasting?[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2023,37(9):11121-11128.
• [14]DONANDT K,B??TTGER K,S??FFKER D. Short-term inland vessel trajectory prediction with encoder-decoder models[C]//2022 IEEE 25th International Conference on Intelligent Transportation Systems(ITSC). New York, USA:IEEE,2022:974-979.
• [15]ZHOU H Y,ZHANG S H,PENG J Q,et al. Informer:beyond efficient Transformer for long sequence time-series forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2021,35(12):11106-11115.
• [16]LIU Y,HU T G,ZHANG H R,et al. iTransformer:inverted Transformers are effective for time series forecasting[C]//International Conference on Learning Representations 2024.Vienna:ICLR,2024:11116-11140.