基于偏振光的无人机航姿LM优化解算方法

LM Optimization Calculate Method of UAV Attitude Based on Polarized Light

针对基于惯导和GPS/磁力计的导航系统易受到电磁干扰及诱导的问题,提出了一种基于偏振光和微惯导组合导航的无人机姿态求解算法。利用大气偏振矢量获取航向信息,搭配惯导元件实现列文伯格-马夸尔特(Levenberg-Marquardt,LM)优化的三维姿态角解算方法。实验结果表明:所提算法能有效降低噪声干扰,保持良好的静态稳定性;同时,相较于梯度法和牛顿法等常规算法,该方法具有更好的动态解算精度。

Aiming at the problem that the navigation system based on inertial navigation and GPS/magnetometer is susceptible to electromagnetic interference and induction,a UAV attitude solving algorithm based on the combined navigation of polarized light and micro-inertial navigation is proposed.The heading information is obtained by using the atmospheric polarization vector,and the three-dimensional attitude angle solution method optimized by Levenberg-Marquardt(LM)is implemented with the inertial navigation element.The experimental results show that the proposed algorithm can effectively reduce noise interference and maintain good static stability;at the same time,compared with conventional algorithms such as gradient method and Newton method,it has better dynamic solution accuracy.