中文摘要：

本论文研究主一从自适应卡尔曼滤波器的设计及其在动基座传递对准中的应用．对于舰载惯性导航系统，利用速度加角速度匹配方案能够实现快速对准，然而该方案对船体挠曲变形比较敏感，若处理不当将造成对准精度下降．本文将挠曲变形视为对准过程中观测量的不确定性干扰噪声进行处理，并且利用方差匹配策略设计了主一从自适应滤波器，这两个滤波器并行运算，其中主滤波器用于估计惯性导航系统的状态，从滤波器用于估计噪声的统计特性．仿真结果表明，在对准模型存在未知的随机系统噪声时，所设计的滤波器能够快速且准确地估计出失准角，符合传递对准在快速性和精度方面的需求．

英文摘要：

A master-slave adaptive filter is designed and its application to shipboard transfer alignment is studied. Although the velocity and angular rate matching scheme offers a rapid transfer-alignment for shipboard inertial navigation systems, it is sensitive to ship flexure. The flexure is modeled as a time-varying noise, of which the stochastic properties are estimated by an adaptive filter. Two Kalman filters are designed, one is the master filter for estimating the dynamic states, the other is for estimating the measurement noise statistics based on the master filter residuals. These two filters exchange information and run simultaneously. Simulation results show that the misalignment angles of slave navigation system can be estimated rapidly and accurately by the proposed filter.