为解决目标机动导致相关纯方位跟踪算法无法使用的问题，提出一种基于变异数分析的纯方位目标机动辨识方法。利用目标机动前后方位序列回归系数发生变化的特点，结合统计学中变异数分析方法，将纯方位目标机动辨识问题转化为统计学中的变点检测问题，通过假设检验方法判定目标是否机动。该方法直接利用量测的方位序列作为纯方位目标机动辨识的唯一输入，不依赖于纯方位目标跟踪算法解算结果。理论分析与仿真计算验证了该方法的可行性与有效性，为工程应用中提高相关跟踪算法的适用性提供一种技术途径。
A bearing-only target maneuvering identification method based on variance analysis is proposed to address the issue of target maneuvering causing the inability of relevant bearing-only tracking algorithms. By utilizing the characteristics of changes in the regression coefficients of the bearing sequence before and after target maneuvering, combined with the analysis of variance method in statistics, the identification problem of bearing-only target maneuvering is transformed into a change point detection problem in statistics. Hypothesis testing is used to determine whether the target is maneuvering. This method directly utilizes the measured bearing sequence as the unique input for maneuvering identification of bearing-only targets, without relying on the tracking algorithm solution results. Theoretical analysis and simulation calculations have verified the feasibility and effectiveness of this method.Provide a technical approach to improve the adaptability of relevant tracking algorithms in engineering application.
2025,47(8): 119-125 收稿日期：2024-6-12
DOI：10.3404/j.issn.1672-7649.2025.08.020
分类号：E917；TN953
基金项目：装备预研领域基金资助项目（3020902020101）；装备预研领域基金资助项目（22-TQ02-01-ZT-01-001）
作者简介：张亚斌(1993-),男,硕士,工程师,研究方向为水下目标运动分析
参考文献：
[1] 司广宇. 潜艇指控系统理论与应用[M]. 北京: 电子工业出版社, 2018.
[2] 袁富宇, 肖碧琴, 刘凯, 等. 关于TMA的Taylor级数法的分析[J]. 指挥控制与仿真, 2022, 44(1): 38-43.
YUAN F Y, XIAO B Q, LIU K, et al. Analysis of Taylor series method for TMA[J]. Command Control & Simulation, 2022, 44(1): 38-43.
[3] 王越. 非线性滤波在水下机动目标跟踪技术中的应用[D]. 哈尔滨: 哈尔滨工程大学, 2017.
[4] 李岩昊, 袁富宇. 一种基于航向估计的多平台纯方位目标转向机动检测[J]. 指挥控制与仿真, 2023, 45(3): 126-133.
LI Y H, YUAN F Y. A multi-platform bearing-only target steering maneuver detection algorithm based on heading estimation[J]. Command Control & Simulation, 2023, 45(3): 126-133.
[5] 刘凯, 苗艳, 袁富宇. 用于纯方位机动目标跟踪的机动探测方法[J]. 指挥控制与仿真, 2006, 28(2): 30-34.
LIU K, MIAO Y, YUAN F Y. Maneuver detection algorithm used for bearing-only maneuver target tracking[J]. Command Control & Simulation, 2006, 28(2): 30-34.
[6] 肖碧琴. 纯方位小幅机动目标定位跟踪技术研究[J]. 电光与控制, 2017, 24(9): 27-30.
XIAO B Q. Locating and tracking of target with bearing-only small maneuvering[J]. Electronics Optics & Control, 2017, 24(9): 27-30.
[7] 梅鹏, 孙振新, 马莎莎. 基于机动检测的水下目标自适应跟踪算法[J]. 舰船科学技术, 2022, 44(6): 114-120.
MEI P, SUN Z X, MA S S. Adaptive tracking algorithm for underwater target based on maneuver detection[J]. Ship Science and Technology, 2022, 44(6): 114-120.
[8] 李鑫悦. 基于自归一化方法的时空变点检测[D]. 成都: 西南财经大学, 2023.
[9] 王国琴, 吴有富, 许婷, 等. 一种基于贝叶斯的广义Pareto分布变点估计[J]. 遵义师范学院学报, 2024, 26(1): 108-112.
WANG G Q, WU Y F, XU T, et al. A bayesian based generalized Pareto distribution change point estimation[J]. Journal of Zunyi National University, 2024, 26(1): 108-112.
[10] 张思, 金浩, 杨云峰. 基于Bootstrap抽样的厚尾自回归过程结构变点检测[J]. 华中师范大学学报(自然科学版), 2023, 57(4): 483-493.
ZHANG S, JIN H, YANG Y F. Structural change-point detection for heavy-tailed autoregressive process based on Bootstrap sampling[J]. Journal of Central China Normal University (Nature Science Edition), 2023, 57(4): 483-493.
[11] 吴常铖, 严余超, 曹青青, 等. 基于ANOVA的BP神经网络的最优肌电信号测量位置选择[J]. 南京信息工程大学学报(自然科学版), 2019, 11(2): 173-179.
WU C C, YAN Y C, CAO Q Q, et al. EMG measurement position optimization based on ANOVA and BP neural network[J]. Journal of Nanjing University of Information Science and Technology(Natural Science Edition), 2019, 11(2): 173-179.
[12] 赵海峰, 高天驰. 基于ANOVA的特征选择方法及其在齿轮箱故障诊断中的应用[J]. 内燃机与配件, 2017, 57(6): 110-113.
ZHAO H F, GAO T C. Feature selection method based on ANOVA and its application in gearbox fault diagnosis[J]. Internal Combustion Engine & Parts, 2017, 57(6): 110-113.
