26 January 2016 Autonomous celestial navigation based on Earth ultraviolet radiance and fast gradient statistic feature extraction
Author Affiliations +
Proceedings Volume 9796, Selected Papers of the Photoelectronic Technology Committee Conferences held November 2015; 97961R (2016) https://doi.org/10.1117/12.2228944
Event: Selected Proceedings of the Chinese Society for Optical Engineering Conferences held November 2015, 2015, Various, China
Abstract
To meet the requirement of autonomous orbit determination, this paper proposes a fast curve fitting method based on earth ultraviolet features to obtain accurate earth vector direction, in order to achieve the high precision autonomous navigation. Firstly, combining the stable characters of earth ultraviolet radiance and the use of transmission model software of atmospheric radiation, the paper simulates earth ultraviolet radiation model on different time and chooses the proper observation band. Then the fast improved edge extracting method combined Sobel operator and local binary pattern (LBP) is utilized, which can both eliminate noises efficiently and extract earth ultraviolet limb features accurately. And earth’s centroid locations on simulated images are estimated via the least square fitting method using part of the limb edges. Taken advantage of the estimated earth vector direction and earth distance, Extended Kalman Filter (EKF) is applied to realize the autonomous navigation finally. Experiment results indicate the proposed method can achieve a sub-pixel earth centroid location estimation and extremely enhance autonomous celestial navigation precision.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Shan Lu, Hanmo Zhang, "Autonomous celestial navigation based on Earth ultraviolet radiance and fast gradient statistic feature extraction", Proc. SPIE 9796, Selected Papers of the Photoelectronic Technology Committee Conferences held November 2015, 97961R (26 January 2016); doi: 10.1117/12.2228944; https://doi.org/10.1117/12.2228944
PROCEEDINGS
8 PAGES


SHARE
Back to Top