论文部分内容阅读
对采用磁强计和太阳敏感器的卫星姿态模型,应用无迹卡尔曼滤波(UKF)算法确定卫星姿态。由四元数描述姿态,在算法中将其转换成旋转矢量,另设计了四元数求均值的迭代算法,解决了四元数在UKF算法中的处理。仿真结果表明:广义卡尔曼滤波(EKF)算法对初值依赖性较大,UKF算法不受初值影响且精度高于EKF,具快速性、精度高和稳定性强等优点。
For satellite attitude models using magnetometers and sun sensors, Unscented Kalman Filter (UKF) algorithm is used to determine satellite attitude. The quaternion is used to describe the pose, which is converted into a rotation vector in the algorithm. Another iterative algorithm for calculating the mean of the quaternion is designed and the quaternion is dealt with in the UKF algorithm. The simulation results show that the generalized Kalman filter (EKF) algorithm is more dependent on the initial value, the UKF algorithm is not affected by the initial value and the accuracy is higher than EKF, which has the advantages of fastness, high precision and strong stability.