捷联惯导系统静基座初始对准精度分析及仿真

在利用卡尔曼滤波器对捷联惯导系统（SINS）进行静基座初始对准中，由于系统的不完全可观测性，使得有些状态没有滤波效果，有些状态的估计精度受到限制。对SINS静基座初始对准卡尔曼滤波方程进行了可观测性分析，提出了状态降阶的处理方法，并得到了各状态估计的极限精度公式。最后进行了软件仿真，仿真结果表明：降阶滤波器和全降阶滤波器的估计精度基本相同，但足前者计算量更小，并且在滤波计算中能够消除不可观测状态的不利影响。

Initial Alignment Accuracy Analysis and Simulation of Strapdown Inertial Navigation System on a Stationary Base

In the process of initial alignment for an inertial navigation system (SINS) on a stationary base via Kalman filter, some states have no filtering effect as the system is not completely observable. Moreover, the accuracy of the states which have filtering effect are limited to some extend for the poor observability. In this paper, the observability for the initial alignment equations of SINS on a stationary base is analyzed, and the reduced - order Kalman filter is established with each state estimating accuracy limitation proposed. Simulation results show that the reduced - order Kalman filter has the some accuracy as the traditional full - order estamation method, but the former has less calculating consumption and is not sensitive to the bad effect of the unobservable states.