基于残差自适应的纯方位伪线性卡尔曼滤波跟踪方法  

Residual adaptive bearing-only pseudo-linear Kalman filter tracking method

作　　者：陈启航 罗威[1] 谢晓乐 代涛 潘景辉 CHEN Qihang;LUO Wei;XIE Xiaole;DAI Tao;PAN Jinghui(Innovation Center,China Ship Development and Design Center,Wuhan 430064,China;School of Electronic Information,Wuhan University,Wuhan 430060,China)

机构地区：[1]中国舰船研究设计中心创新中心,湖北武汉430064 [2]武汉大学电子信息学院,湖北武汉430060

出　　处：《舰船科学技术》2025年第4期130-136,共7页Ship Science and Technology

基　　金：国家自然科学基金叶企孙重点项目(U2341228)。

摘　　要：在噪声复杂多变的海上环境上,针对传统卡尔曼滤波算法在被动声呐纯方位跟踪场景中存在的滤波发散问题,提出一种残差自适应的纯方位伪线性卡尔曼滤波算法。将观测残差引入到伪线性卡尔曼滤波中,改进直接用于自适应估计伪线性观测噪声方差,并通过SAM准则对滤波的过度补偿进行修正。结果表明,在观测噪声方差不匹配时,本文算法的稳定性和性能优于传统的偏差补偿伪线性卡尔曼滤波算法;在偏差补偿伪线性卡尔曼滤波算法发散情况下,所提算法与IEKF、EKF算法相比,位置方向上时间平均均方根误差(RTAMS)分别减少44.87%和64.88%,速度方向上时间平均均方根误差分别减少17.30%和30.99%,在改善伪线性卡尔曼滤波的稳定性的同时,增大了补偿算法的性能,可以为海上环境的机动平台目标跟踪提供参考意义。In a maritime environment with complex and changeable noise,a residual adaptive bearing-only pseudo-lin-ear Kalman filter algorithm is proposed to address the filter divergence problem of the traditional Kalman filter algorithm in passive sonar bearing-only tracking scenarios.The observation residual is introduced into the pseudo-linear Kalman filter,and the improvement is directly used to adaptively estimate the pseudo-linear observation noise variance,and the over-com-pensation of the filter is corrected through the SAM criterion.The results show that when the observation noise variance does not match,the stability and performance of the proposed algorithm are better than the traditional bias compensation pseudo-linear Kalman filter algorithm;when the bias compensation pseudo-linear Kalman filter algorithm diverges,the proposed al-gorithm is consistent with IEKF,Compared with the EKF algorithm,the time-averaged root mean square error(RTAMS)in the position direction is reduced by 44.87%and 64.88%respectively,and the time-averaged root mean square error in the speed direction is reduced by 17.30%and 30.99%respectively.In improving the pseudo-linear Kalman filter While improv-ing stability,it also increases the performance of the compensation algorithm,which can provide reference significance for mobile platform target tracking in the maritime environment.

关 键 词：纯方位跟踪 残差 角度选择策略 伪线性卡尔曼滤波 

分 类 号：P228.4[天文地球—大地测量学与测量工程]