Reserch Of Celestrial Positioning Technology With Thin Field Of View

In recent years,with the continuous deepening space exploration,the efficient acquisition and deep excavation of spatial information has become an important research area in the field of space exploration in various countries.As a essential brunch of celestial navigation technology,celestial positioning technology has extensive practical value in military and civil fields.The celestial positioning technique takes natural celestial objects as reference standard,and uses the ground-based optoelectronics measuring equipment to calibrate the spatial targets in highly precision.The space objects of celestial positioning technology are mainly divided into two brunches: artificial spacecraft and orbits debris.With the increasing frequency of space missions such as space launches,the outer space of the earth is gradually developing into a new territory of military struggle.In this situation,the celestial positioning technology for artificial spacecraft has become one of the key technologies of the aerospace measurement and control system,meanwhile has been widely used in the fields of reconnaissance,detection,tracking and location prediction.Frequently space missions leads to exponential growth in of space debris,space debris threatens to satellites and normal operation of space station is becoming more and more severe,a large number of uncatalogued space debris sufficiently cause davastated disaster.To plan a long-term and a effective utilize of space,the celestial positioning technology provides an important data basis.Therefore,to provide a low cost,high precision,independent from exchange information with the outer environment,undisturbed by other systems of the celestial positioning technology has become a practical problem for scholars in various countries.As one of the important branches of celestial positioning technology,the narrow field of view(FOV)celestial positioning has a small FOV(usually around 4°x4°)characteristics,under this circumstances,due to the limitation of distribution of the stars,it`s easily to appear less observable guiding stars than it is required or even no guiding star in a direction of certain optical axis,and ultimately led to failure of celestial positioning.In view of this,this paper based on the narrow FOV celestial positioning technology research and practical significance and application prospects,based on three aspects of star database selection,star pattern recognition and ground-based optoelectronic surveillance,this paper makes theoretical research and practical application of celestial positioning technology,in which the main research results and innovative work are as follows:(1)Aiming at the problem of poor uniformity of the guidance star catalogue and the "hole" in the FOV when applying traditional method of star catalogue construction to the condition of narrow FOV(4°x4°),this paper proposes a method of constructing a guidance star catalogue combining Boltzmann entropy and support vector machine.In order to reduce the star density in the neighborhood of each star in the original star catalogue,improve the uniformity and reliability of the navigation star catalogue,this paper introduce the magnitude of star in the original catalogue,the number of star points in its neighborhood and the Boltzmann entropy into the feature vectors of the star point,and complete guidance star classification under the minimum structure risk by the support vector machine method.Finally,a guidance star catalogue suitable for narrow FOV is constructed,and its performance is simulated and analyzed.The experimental results show that under the condition of narrow FOV,this method combines Boltzmann entropy and support vector machine to construct the guidance star database,compared with the traditional threshold filtering method(VMT),its star catalogue capacity reduced to the original 66.85%,the standard deviation reduced to the original 78.16%,and "hole " in guidance star catalogue also got inhibited.(2)Aiming at the problem of the low recognition rate caused by the disturbance of position noise when applying traditional star pattern recognition algorithm,due to the interference in the construction of star pattern,this paper proposes a fast radial companion star pattern recognition algorithm with adding compensation code into radial feature vector.The algorithm constructs a feature vector based on the characteristic of the radial companion star in the form of a bit vector,and adds the information of the angular distance between the companion stars and the compensation information of the position noise to the feature vector,which effectively reduces the capacity of the guidance star catalogue and improves the stability and the recognition rate of the pattern recognition algorithm.Finally,based on the characteristics of the bit vectors,the first match between the observation star and the guidance star is completed by using the method of least similarity difference,and the unique recognition of the observation star is completed based on the correlation of the location information of the star in the same field of view.The simulation results show that the success rate of pattern recognition is 97.8% in the case of position noise as1 pixels,and the success rate of star pattern recognition is 96.4% in the case of 0.46 Mv magnitude noise.When apply to the real star imag,the success rate of star pattern recognition algorithm reaches 94.2%.Compared with the traditional triangle algorithm and the radial feature pattern recognition algorithm with no compensation code,the algorithm has improved in different degree in recognition success rate and time.(3)In view of the significant increase in orbital satellites and debris,a large number of uncataloged debris is sufficient to cause davasteted disaster,and the problem of long-term space utilize needs to be reconsidered.In this paper,the optical reflection characteristics of the space objects are analyzed in detail,the signal-to-noise ratio(SNR)in different modes is compared and the optimal design of the optoelectronic surveillance is given;meanwhile the observation mode of the ground-based geosynchronous orbit(GEO)objects is designed,which is suitable for the narrow field of vision.Aiming at the problem of faint object recognition in star image,this paper proposes a new method for image processing of low SNR object recognition based on optimization principle.The experimental results show that compared with the traditional difference frame method the objecrt can be recognized under condition of SNR more than 3.09.(4)For narrow FOV,in certain optical axis direction it not enough guide stars can be observed in the sky area,a narrow celestial positioning method with prior information is proposed in this paper,with mapping relationship between CCD coordinates and celestial coordinates obtained by using correlated frames introduced to the current frame of the celestial positioning.The experimental results show that the proposed method can obtain the position information of spatial target with low measurement error under the condition of only one guide star.For the LOW object,the measurement error of the right ascension and the declination is less than 5 arc-seconds,and the measurement error of the right ascension and declination of the GEO object is less than 2 arc-seconds.