基于局部大气偏振特性的仿生导航方法

目前主流的偏振导航算法中，思路通常为通过对全天域偏振图像的获取拟合出太阳子午线或中性点的位置，进而判断导航方向。现提出一种新的导航方向获取算法，可在小视场条件下，通过对太阳方向矢量的拟合判断导航方向。在此基础上进行了计算机仿真和室外实验，利用偏振相机获取了天顶区域大气偏振分布模式；利用斯托克斯矢量分析法解算出了大气偏振模式的分布情况，包括其中偏振度和偏振角的信息，初步证实了解算算法在实际情况下的可行性和精度，导航精度可达到0.2°左右。

Bionic navigation based on local atmospheric polarization

In the current mainstream polarization navigation algorithms, it is often necessary to obtain the polarization image of the whole sky, the solar meridian, and the neutral point before judging the navigation direction. In this paper, a new navigation direction acquisition algorithm is proposed, which can determine the navigation direction by fitting the sun direction vector in a small field of view. On this basis, computer simulation and outdoor experiments are carried out. The polarization distribution mode of the atmosphere in the zenith area is obtained by the polarization camera, and the distribution of the atmospheric polarization mode calculated by the Stokes vector analysis method, including the information of polarization degree and polarization angle. The feasibility and accuracy of the algorithm in the actual situation are verified, navigation accuracy can reach about 0.2°.