26 August 2011 An improved triangle star pattern recognition algorithm with high identification probability
Author Affiliations +
In this paper, an improved triangle star pattern recognition algorithm is proposed. There are two improvements in the algorithm compared with the classic triangle algorithm. Firstly, the adjacent measured stars in field of view are eliminated and the remaining measured stars are used to build candidate measured star triangles. So high attitude determination precision is obtained if a triangle is chosen from candidate measured star triangles to identify the three stars. Secondly, in the candidate measured star triangles, the triangle which has great difference in three angular distances is chosen as the measured star triangle waiting for identification, so as to mismatch between it and every guide star triangle due to angular distance match can be avoided. In the case of a certain position and magnitude error limit, simulation results indicate that the identification probability of this algorithm is much higher than that of the classic triangle algorithm, and is also higher than that of the triangle algorithm based on the star pattern of two angular distances and their angle. So the algorithm introduced in this paper can enhance the reliability of attitude determination of star trackers.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Lei Du, Lei Du, Yan Zhao, Yan Zhao, } "An improved triangle star pattern recognition algorithm with high identification probability", Proc. SPIE 8193, International Symposium on Photoelectronic Detection and Imaging 2011: Advances in Infrared Imaging and Applications, 819305 (26 August 2011); doi: 10.1117/12.895213; https://doi.org/10.1117/12.895213

Back to Top