Research On High Dynamic Motion Star Positioning Technology Of Star Sensor

As an important part of spacecraft navigation system,star sensor is a high-precision instrument for attitude measurement.With the rapid development of space industry,high dynamic capability of spacecraft in orbit has become a major demand for space missions.In a high dynamic environment,the high angular velocity of the spacecraft make the star sensor move too fast relative to the observing star,leading to the degradation of the star image,such as motion tailing of star points or signal-to-noise ratio decline,which results in the decline of the star location performance and the limited dynamic performance.Therefore,it is very important to study the star point location technology of star sensor with high dynamic.In high dynamic situations,the background noise such as fixed pixel noise,pixel non-uniformity noise and dark signal non-uniformity is an important factor affecting the sensitivity of the star sensor,and motion blur causes the tailing of the star point,which leads to the decrease of the accuracy of the star centroid extraction,or even the failure of extraction.This paper mainly focuses on improving the star location accuracy of star sensor in high dynamic environment.The main content is divided into the following parts:(1)The basic working principle and main technical indexes of the star sensor are introduced,the factors affecting the dynamic performance of the star sensor are analyzed,and the research status of background noise suppression and moving star tailing removal at home and abroad is summarized.(2)The characteristics of static star imaging are analyzed.On this basis,the influence of triaxial rotation on star dynamic imaging is analyzed,and the star dynamic imaging model is deduced.Where the star map background noise from and the principle of star map degradation are analyzed.Finally,the centroid location principle of star point and the common centroid calculation method are introduced.(3)For the problem that signal-to-noise ratio decline under the condition of high dynamic,this paper introduced background noise suppression correction methods.Aiming at the star sensor with compact resources,a dynamic adaptive on-orbit correction method is proposed.The simulation experiment verifies that the method can adapt to the change of the angular velocity of the star sensor,which greatly saves the cost of space resources and removes the background noise of the star map in real timeand the signal-to-noise ratio of the star map be improved.(4)Aiming at the problem of star-point motion tailing under high dynamic conditions,two steps: motion blur kernel parameter estimation and star-point reconstruction were studied respectively.About blur kernel estimation,according to the demand of the star sensor,this paper proposed a fast blur kernel estimation method based on single frame map,using PCA in double spectrum domain to estimate motion angle.And this paper proposed a adaptive adjustment method to estimate the motion blur lenghth.The proposed method is verified by experiment can quickly get accurate blur kernel parameters,meeting the needs of star sensor.For star reconstruction,a region restoration method based on hyperLaplacian restoration algorithm is proposed which performs local restoration processing on star image and saves space resource costs and improves the real-time performance of the algorithm.(5)The overall algorithm of high dynamic star locating of star sensors is designed.The centroid error of the star extracted by the proposed method is greatly reduced and the number of star extracted is increased,which verifies that the proposed method can effectively improve the star location performance of the star sensor under high dynamic conditions,and it is of great significance to improve the dynamic performance of the star sensor.