Ultraviolet vs. Visible Skylight Polarization Measurements

Abstract

Concerning autonomous navigation in GNSS-denied environments, insects like bees, ants, locusts and butterflies are certainly the cup winners. Bees and some ants are known to be sensitive to polarized light, both in ultraviolet (UV) and visible spectra, and to use it for their outside moving heading calculation. Therefore, we chose to compare UV and visible skylight polarization measurements in order to understand how UV could be used to improve heading estimation by processing the skylight's angle of polarization patterns. We then used a commercial polarimetric camera sensitive to visible polarized light and a UV-sensitive camera enhanced with polarized light measurement ability.

SkyPole: a geolocation algorithm based on polarized vision without using astronomical ephemerides

“…14 Then, Rayleigh single scattering model was simulated using the OpenSky simulator. 11 For each set of three images, we computed the position of the celestial pole using the proposed algorithm, without the least square step, which was skipped. We observe that some errors occurs near the poles or the equator.…”

“…To simplify the training task, we trained our network on simulated images only, using the OpenSky simulator. 11 Radial invariance segmented images were generated using the analytical equation of this set (see supplemental information 4 ). Real images where pre-processed using a least square method, in order to match a simulated image of the DoLP, based on Rayleigh single scattering model, to the real image.…”

Passive Polarized Vision for Autonomous Vehicles: A Review

“…Challenges remain. By and large, skylight polarization in the UV range has seldom been exploited for robotic applications, none of the commercial or experimental devices detect the surrounding light in a full panoramic view as insects do, and none of them can work in cloudy or extreme weather conditions [ 268 ].…”