中文摘要：

针对弹载SINS／GPS系统的空中对准，提出一种四元数模约束条件下的非线性滤波初始对准算法。首先，利用坐标变换和四元数姿态描述，将传统的强非线性滤波对准问题转化为一个二阶弱非线性滤波问题；其次，采用二阶扩展卡尔曼滤波（EKF）对二阶非线性部分进行处理，得到一种简洁的滤波对准方案；最后，推导了四元数模约束条件下滤波算法的最优实现，及反馈四元数估计结果时的闭环滤波形式。利用车载MEMS IMU／GPS系统，进行了初始对准的地面试验，结果表明，在车辆弱机动条件下，对准算法能够实现姿态和惯性器件误差的快速估计，实现惯导系统的对准。

英文摘要：

A norm constraint quaternion nonlinear filtering in-motion alignment method is proposed in this paper for missile-borne SINS/GPS systems. Based on coordinate transformation and quaternion description, the traditional strong nonlinear alignment problem is simplified to a weak second order nonlinear one. Then the second order extended Kalman filter（EKF） is used to process the weak nonlinear part and a simple alignment algorithm is obtained. Finally, the optimized elosed-loop implementation of the proposed method is deduced under the quaternion norm constraint. Alignment experiment indicates that under weak maneuver of the land vehicle, the attitude and sensor errors can still converge fast.