Research On The High Accuracy Subdivision Localization Algorithm Of Star Point

The star sensor is used to obtain the current observations data of the star,so as to determine the attitude information of the spacecraft.In the process of star sensor work,the star point extraction algorithm directly determines the accuracy of star map recognition,and it is the basic technical index to determine the attitude of spacecraft.In this paper,we based on the actual needs of star sensors,focuses on the star point extraction algorithm,and proposes a star point extraction algorithm based on PSF(Point Spread Function),and validates the algorithm.Specific work content and innovation include:(1)The coordinate transformation of star imaging and star energy distribution characteristics were analyzed,and according to the actual data and shooting star established partial normal distribution model and the energy distribution of star sensor model based on the energy distribution of saturated star.(2)In this paper,combined with the star energy distribution characteristics,we propose a PSF based correlation algorithm for star extraction.The algorithm includes the PSF correlation method based on partial normal distribution and the PSF correlation method based on supersaturation.These two PSF correlation methods have better applicability to the star point energy distribution as partial and sensor supersaturated star image.(3)In order to verify the validity of the proposed algorithm,we generate the simulated star images under the energy distribution of different star points according to the parameters of the star chart,which are used to verify the PSF correlation algorithm and compared with the traditional star point extraction algorithm.Simulation results show that:In the star images with noise,both the normal PSF correlation method and the supersaturated PSF correlation method have higher average positioning accuracy than the traditional centroid method,the square weighted centroid method and the Gauss surface fitting method,and the anti-noise performance and stability are better.In this paper,we focus on the algorithm of star point segmentation and localization.A new algorithm based on PSF correlation is proposed,which can effectively improve the positioning accuracy of the star center in the condition that the energy distribution of the star point is partial normal distribution and sensor supersaturation.