Solar position acquisition method by structural similarity characteristics of skylight polarized pattern images

Obtaining solar position from the skylight polarized information is one of the key issues in the research of the polarization navigation. Through the statistical observation of the skylight polarized pattern images information, a method is proposed to obtain the solar position by utilizing characteristics of the skylight polarized pattern based on the structural similarity. Based on the characteristics of the skylight polarized pattern, the structural similarity function of the skylight polarized pattern is established for solving solar position. The bat optimization algorithm is utilized to optimally estimate the structural similarity function. And the theoretical skylight polarized pattern that best matches the measured skylight polarized pattern can be obtained. Then, the corresponding solar position can be calculated. The experimental results show that the average errors of azimuth and zenith angle estimation are 0.55° and 0.66° without interference, respectively. In the interference experiment, when the cloud area is less than 65%, the errors of azimuth and zenith angle are within 1.5°. It is proved by experimental results that the proposed method can obtain solar position in the measured skylight polarized pattern. And it has a certain anti-interference ability to the sky occlusion caused by clouds.