An Autonomous Positioning Method Utilizing Feature Extraction From Polarized Moonlight

Precisely position-determining in challenging nighttime environments (e.g., limited visibility, electromagnetic interference, and signal deception) presents an intriguing and complex problem. Inspired by nocturnal insects’ navigation mechanism, an autonomous positioning method utilizing feature extraction from polarized moonlight is presented. Significant errors in the observed moonlight polarization pattern are addressed. To improve the single Rayleigh scattering model (SRSM) for positioning application, the E-vector error is modeled. A polarization feature extraction approach is developed such that the light interference and device noises in low-light intensity environments can be tackled. Moreover, an autonomous positioning method is proposed without relying on external sensors. Finally, the experimental result illustrates the feasibility of this bioinspired global autonomous nighttime positioning method. The test results display location errors within tens of kilometers, accompanied by a latitude accuracy of 0.62° (
$1\sigma $
) and a longitude accuracy of 0.02° (
$1\sigma $
).