摘要
针对在被动方式下进行水下目标跟踪容易导致滤波发散和收敛精度不高的问题 ,介绍了一种改进的自适应推广卡尔曼滤波算法。它能够在线估计虚拟观测噪声的统计特性 ,从而克服了观测模型线性化误差带来的不良影响。同时 ,通过引入修正增益函数 ,克服了由于观测噪声的统计特性不能精确已知而导致的滤波不稳定问题。仿真结果表明 ,不管是滤波精度还是收敛速度 ,都优于原来的自适应推广卡尔曼滤波算法。
Concerning the problem of instability and low accuracy of passive filter in underwater target tracking,a modified adaptive extended kalman filter(MAEKF) algorithm is presented.Owing to estimating the statistics of virtual noise on line,it overcomes the bad affect caused by linearization of nonlinear observation model.Meanwhile with the help of modified gain function,the problem of instability caused by the unknowm of statistics of measurement noise is solved,which greatly improves the filter rate and accuracy.
出处
《火力与指挥控制》
CSCD
2000年第3期13-16,27,共5页
Fire Control & Command Control
基金
九五国防科技预研重点项目! (44.6.1 .3)
关键词
自适应推广卡尔曼滤波
水下目标跟踪
虚拟观测噪声
修正增益函数
adaptive extended kalman filter
underwater target tracking
virtual observation noise
modified gain function
