论文部分内容阅读
针对陀螺漂移增大这种渐变故障,提出了一种UKF(Unscented Kalman Filtering)结合姿态运动学方程进行角速率估计,从而进行陀螺故障预报的方法。用四元数表示姿态运动学方程,以卫星姿态角和陀螺角速率为状态量,太阳敏感器和地球敏感器确定的姿态角为观测量,创建了UKF滤波器模型。根据估计角速率与陀螺测量值产生的残差序列,提出陀螺故障预报方法。避免了动力学方程受星体惯量和控制力矩影响产生的误差以及EKF,PF滤波算法的不足。在MATLAB环境中进行了仿真,仿真结果表明,该算法可以及时准确的预报陀螺漂移增大故障,模型简单,易于构建,计算量小,具有很好的工程实用性。
Aiming at the gradual change of gyro drift, this paper proposes a method of estimating gyro angular rate by Unscented Kalman Filtering (UKF) combined with kinematic equation of attitude. Quaternion is used to represent the kinematic equations of attitude. The attitude angle determined by the attitude angle of the satellite and the gyroscope is used as the observer, and the UKF filter model is created. According to the residual sequence produced by the estimated angular velocity and gyroscope measurement, the gyroscope fault prediction method is proposed. Which avoids the error caused by the inertia and the control moment of the kinetic equation and the deficiencies of the EKF and PF filtering algorithms. The simulation results in MATLAB environment show that the proposed algorithm can predict gyroscope drifts and increase faults timely and accurately. The model is simple, easy to construct and has a small amount of calculation, and has good engineering practicability.