摘要
针对滤波稳定性问题,提出了一种改进的衰减记忆自适应滤波算法。通过引入衰减记忆滤波矩阵,根据残差序列输出的互不相关性,在线自适应地调整衰减因子,从而使衰减记忆滤波工作在最佳状态。将该算法应用于惯导系统的传递对准过程,仿真结果表明在模型和噪声统计特性的先验信息不准确时,该算法优于传统的卡尔曼滤波。
A new adaptive estimation method of Kalman filter fading factor is presented from the stability of Kalman filter.The fading filter matrix is introduced,and the fading factor is adjusted adaptively on line based on the uncorrelated character of residual,so the fading filter works in the optimal state.The adaptive fading filter is applied in transfer alignment of inertial navigation systems（INS）,the result shows the proposed algorithm is better than other algorithm when the prior data of model and noise character is unaccurate.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2010年第12期2648-2651,共4页
Systems Engineering and Electronics
基金
国家自然科学基金(60874112)
十一五预研项目(51309060401)资助课题
关键词
卡尔曼滤波
自适应滤波
惯导系统
传递对准
衰减因子
Kalman filtering
adaptive filtering
inertial navigation system
transfer alignment
fading factor
