中文摘要：

为了满足小型无人机自主控制系统对导航系统性能的要求,研究低成本的基于微机电系统的捷联惯性导航系统（MEMS-SINS）/全球定位系统（GPS）/磁强计组合导航系统。提出一种利用磁强计辅助MEMS-SINS的静基座初始姿态确定方法,采用四元数误差模型对MEMS-SINS/GPS/磁强计组合导航系统进行信息融合的建模,采用基于正交三角（QR）分解的平方根无色卡尔曼滤波（UKF）非线性估计方法对组合导航系统进行数据融合,克服由于计算机舍入误差引起的状态协方差阵的计算值失去非负定性甚至对称性,通过小型无人机的自主飞行试验,证实MEMS-SINS/GPS/磁强计组合导航算法满足小型无人机自主控制系统的要求。

英文摘要：

In order to achieve better precision and meet the reliability requirements of small unmanned air vehicle（UAV） navigation systems,this article presents an initial attitude determination method of strapdown inertial navigation system based on a micro electromechanical system（MEMS-SINS）/global positioning system（GPS）/magnetometer integrated navigation system.And the models of quaternion error and the orthogonal-triangular（QR） decomposition of unscented Kalman filter（QR-UKF） are used for data fusion of the MEMS-SINS/GPS/magnetometer integrated navigation system. Autonomous flight test results of small UAVs show that the MEMS-SINS/GPS/magnetometer integrated navigation system not only provides continuous and reliable navigation solutions to the guidance and flight control loop, but also has guarantees positive semi-definiteness of the state eovarianee.