Application Of Polarization Camera Based On Pixel Polarization Array In Bionic Polarization Navigation

In nature,many insects can perceive the polarization mode of sky light and use this information for precise navigation.Researchers imitate the navigation behavior of insects to carry out a lot of bionic polarized light navigation research.The polarization navigation sensors currently used in bionic polarization navigation applications are point source photoelectric sensors and multi-channel image sensors,which cannot simultaneously meet the three requirements of high stability,real-time imaging and high integration.With the rapid development of nanotechnology,the newly developed polarization camera based on pixel polarization array can meet these three requirements at the same time.Therefore,this paper studies the application of polarization camera based on pixel polarization array in the field of bionic polarization navigation.This topic mainly carries out the following three aspects of research work:(1)In order to solve the problem of spatial resolution loss in the four polarized pictures of 0°,45°,90° and 135° extracted from the original pictures taken by the polarized camera,a bilinear interpolation algorithm and a gradient-based interpolation algorithm are used for resolution compensation.The two interpolation methods are studied and the performance of the interpolation algorithms is compared.According to Marius' s law,a calibration experiment is carried out on the pixel polarization camera,and the calibration Mueller matrix of the polarization camera is calculated.(2)In this paper,the method of extracting the sun's azimuth angle information is studied.For the problem that the meridian can not be recognized by the polarization angle picture calculated by Stokes vector,the coordinate axis conversion is carried out by the coordinate system redrawing method to obtain the polarization angle picture with the meridian as the symmetry axis.Two methods for calculating the solar azimuth angle are studied,namely the solar vector estimation method based on the feature vector and the solar azimuth calculation method based on the image,and the180° ambiguity of the solar azimuth angle is solved by using the degree of polarization distribution.(3)An effective navigation test device was built to carry out the heading angle test experiment,and the heading angle accuracy test comparison between the two solar azimuth calculation methods was carried out.The experimental results show that the image-based heading angle calculation method has an average error of 0.332°and a maximum error of 0.882°;the feature vector-based sun vector calculation method has an average heading angle error of 2.304° and a maximum error of 3.372°.Study the relationship between the solar altitude angle and the accuracy of polarized navigation,and carry out heading angle test experiments at different time periods.The trend of the heading angle error curve obtained by the experiment is consistent with the theory of the single-point scattering Rayleigh model,and it also proves that the morning and evening moments are the best time periods for applying bionic polarization navigation.