论文部分内容阅读
针对四元数在飞行器姿态描述时存在冗余的问题,设计了把修正罗德里格参数(MRPs)作为描述姿态的一种全姿态描述方法,将加速度计和磁强计作为观测量,提出了基于平方根无迹卡尔曼滤波器(SRUKF)的姿态估计算法,针对外界环境对加速度计和磁强计的干扰问题,将专家系统引入SRUKF算法中,根据加速度及磁场信息实时调整滤波器的量测噪声矩阵,有效地提高了算法的抗干扰性。实验结果表明,该算法成功使系统的静态精度姿态角小于0.05°,航向角小于0.1°;动态精度姿态角小于0.5°,航向角小于1°,优化效果得到了明显改善。
Aiming at the problem that the quaternion is redundant when the attitude of the aircraft is described, the modified Rodrigues parameters (MRPs) are described as a kind of all-attitude description method, and the accelerometer and the magnetometer are taken as the observational quantities Based on the square root unscented Kalman filter (SRUKF) attitude estimation algorithm, the expert system is introduced into the SRUKF algorithm according to the disturbance of the external environment to accelerometers and magnetometers. The filter measurements are adjusted in real time according to the acceleration and magnetic field information Noise matrix, which effectively improves the anti-interference of the algorithm. The experimental results show that the proposed algorithm can make the system static attitude angle less than 0.05 ° and the heading angle less than 0.1 °. The dynamic attitude accuracy is less than 0.5 ° and the heading angle is less than 1 °. The optimization results are obviously improved.