基于ANFIS自调整EKF的微卫星姿态确定研究

针对太阳敏感器、磁强计以及惯性陀螺组成的微卫星姿态确定系统，设计了有陀螺测量和基于四元数差分法的无陀螺测量两种广义卡尔曼滤波器（EKF）。由于EKF设计复杂，通常参数计算和调整依赖地面系统，现应用自适应神经模糊推理（ANFTS）进行EKF的参数自调整；使用四元数避免了欧拉角法的奇异问题；采用高斯-牛顿误差最小法显著减少直接使用EKF的计算量。设计仿真数据进行算法验证，结果表明该方法能成功地得到姿态估计，ANFIS可以提高EKF精度，增强鲁棒性，而且便于最优的在轨自主导航。

Micro-Satellite Attitude Determination Based on ANFIS Self-Tuning EKF

For a micro-satellite attitude determination system that used sun sensor,magnetometer and inertial gyro,this paper developed two extended Kalman filtering(EKF) algorithms,One of which incorporates rate measurement while the other relays on differencing quaternion.Because it is complex to design the EKF,and the parameters is tuned and calculated rely on ground support,so this paper used the Adaptive Neuro-Fuzzy Inference System(ANFIS) to tune the parameters of EKF by itself.Quaternion is used rather than Eular angles to avoid the problem of singularity.Gauss-Newton error minimization is used to reduce the computational requirement significantly.Simulated data are designed to test the algorithms.The result illustrates that the method presented here can estimate the attitude successfully.The ANFIS can enhance the accuracy and robustness of EKF,and it is convenient for the autonomy navigate onboard