Calibration and image processing method for polarized skylight sensor

Navigation and positioning technology is widely utilized in numerous aspects in human society, from travel in daily life to aerospace. However, conventional navigation methods such as global navigation satellite system (GNSS) and inertial navigation system (INS) have some inherent defects, such as sensitivity to external interference and error accumulation. Scientists have found that Cataglyphis (a kind of desert ant) and some other insects can capture the polarization direction of skylight to return home along a straight line from a few dozen meters away. Compared to conventional navigation methods, the bionic polarized skylight sensor overcomes these inherent defects. This paper develops a real-time bionic polarized skylight sensor. This sensor only uses a monocular polarized camera equipped with internally installed polarizing films, instead of the complex multi-camera structure using independent polarizing films which need to be adjusted by hand. In order to extract the polarization information from images, a processing algorithm for the images obtained by the polarized camera is introduced. We also conduct an outdoor experiment to verify the Rayleigh scattering theory and the feasible of both device and algorithm. The polarized skylight pattern measured in experiment is consistent with the theoretical atmospheric polarization model and the result shows that the polarized skylight sensor with the calibration and image processing algorithm performs fine robustness.