Modern Defense Technology ›› 2025, Vol. 53 ›› Issue (6): 134-148.DOI: 10.3969/j.issn.1009-086x.2025.06.014

• TARGET CHARACTERISTIC, DETECTION AND TRACKING TECHNOLOGY • Previous Articles    

Improved UKF Universal Target Tracking Algorithm Based on GRU Prediction

Quanyin HUANG, Yichao CAI, Hao LI, Weihua WU, Chenyang WANG   

  1. Air Force Early Warning Academy,Wuhan 430000,China
  • Received:2024-09-15 Revised:2024-11-04 Online:2025-12-28 Published:2025-12-31

基于GRU预测的改进UKF普适目标跟踪算法

黄权印, 蔡益朝, 李浩, 吴卫华, 王辰洋   

  1. 空军预警学院,湖北 武汉 430000
  • 作者简介:黄权印(1999-),男,陕西咸阳人。硕士生,研究方向为机动目标跟踪。
  • 基金资助:
    国家自然科学基金青年项目(61502522);国家社科基金重点项目(2022-SKJJ-B-056);湖北省自科基金面上项目(2019CFC897)

Abstract:

To address the difficulties and poor universality of traditional algorithms for accurate modeling of high maneuvering targets, this paper proposes an improved unscented Kalman filter (UKF) universal target tracking algorithm based on gated recurrent unit (GRU) prediction. The self-attention mechanism is introduced into the traditional GRU model, so that the model can better mine the kinematic equation of the target from the historical observation data and solve the problem that it is difficult to accurately model the maneuvering target manually. The learned model is applied to the UKF algorithm to achieve target tracking. Simulation experiments show that compared with traditional GRU models and classical algorithms such as UKF and interacting multiple model (IMM), the proposed algorithm has better tracking accuracy and adaptability.

Key words: unscented Kalman filter(UKF), target tracking, gated recurrent unit(GRU), self-attention mechanism, universal, interacting multiple model(IMM)

摘要:

针对传统算法对高机动目标精准建模难、普适性差的问题,提出了一种基于门控循环单元(gated recurrent unit, GRU)预测的改进无迹卡尔曼滤波(unscented kalman filter, UKF)普适目标跟踪算法。将自注意力机制引入到传统GRU模型中,使得模型能够更好地从历史观测数据中挖掘出目标的运动学方程,解决人工难以对机动目标运动精准建模问题;将学到的模型应用到UKF算法中,实现对目标的跟踪。仿真实验表明,相比于传统GRU模型以及UKF、交互多模型(interacting multiple model, IMM)等经典算法,所提算法具有更好的跟踪精度以及适应性。

关键词: 无迹卡尔曼滤波, 目标跟踪, 门控循环单元, 自注意力机制, 普适, 交互多模型

CLC Number: