Star Centroid Estimation And Star Identification Of High Accuracy Star Tracker

Star tracker is one of the most important spacecraft attitude sensors which can provide the absolute 3-axis attitude of a spacecraft utilizing star observations. It is widely used in the spacecraft attitude determination system (ADS) and autonomous attitude control system (ACS). It is very important to improve the accuracy of star tracker to fulfill the drastically increasing requirements for satellite payload pointing accuracy. Star centroid estimation and star identification of high accuracy star tracker are studied in this thesis. The method and technology of star centroid estimation and star identification to improve the accuracy of star tracker are explored.A comprehensive discussion is given on the accuracy of star centroid estimation and the identification rate of star identification. The error model of star centroid estimation for star tracker is established. A high accuracy systematical error compensation algorithm based on the error model of star centroid estimation is proposed. An all-sky autonomous star identification method based on Fourier transform correlation is proposed. The main works are listed as follows:1. The thesis begin with reviewing the development history and the updated achievements of star tracker. Based on the basic error model of star tracker, the influence of star centroid estimation and star identification on the attitude measurement accuracy of star tracker is analyzed. A comparative analysis is given on different star centroiding algorithms and star identification algorithms.2. A theoretical analysis of the systematic error of subpixel centroid estimation algorithm utilizing frequency domain analysis under the consideration of sampling frequency limitation and sampling window limitation is presented. Explicit expression of systematic error of centroid estimation is obtained, and the dependence of systematic error on Gaussian width of star image, actual star centroid location and the number of sampling pixels is derived.3. An unbiased subpixel centroid estimation algorithm of point image is proposed through the compensation of the systematic error of center of mass (CoM) method. The Cramér-Rao lower bound (CRLB) on centroid estimation variances is derived under the photon shot noise condition and is utilized to evaluate the proposed algorithm. Numerical analysis shows that the proposed centroid estimator attains the required lower bound, thus the proposed algorithm can be asserted as a minimum variance estimator. Simulation results indicated that the estimator can attain subpixel accuracy closely to 1/100 pixel when 1000 photons are detected. Some important issues for practical applications of compensated center of mass method are discussed, such as interpixel crosstalk, additive noises and multidimensional extension. High accuracy star centroiding experiment is performed to verify this algorithm. 4. The physical band of star tracker attitude estimation accuracy is analyzed under photon shot noise condition and the CRLB on star image centroid estimation variances. The results indicate that, with the typical design parameters, the conservative accuracy of star tracker attitude estimation is about 0.02", which is much higher than the accuracy of modern high accuracy star trackers.5. With respect to the requirements for star identification of star tracker, influence of number of reference stars, number of sensor stars, star brightness noise and star position noise on traditional star identification algorithms is studied in detail. An all-sky autonomous star identification method based on Fourier transform correlation is proposed. It improves the success rate and reduces the wrong identification rate remarkably, and it is robust to star brightness noise and star position noise. The algorithm obtains an identification rate of 99.93% from simulation where the star position noise is 1 pixel and the apparent brightness noise is 0.4 units stellar magnitude.6. A star identification method utilizing optical correlator is studied. The correlation operation of star identification is realized through joint transform optical correlator to reduce time consuming drastically. The time of all-sky star identification can be reduced to 80 millisecond. A coaxial joint transform optical correlator and a diarylethene film-based optical correlator are proposed to reduce the physic length of optical correlator.Through the research of this thesis, star centroiding accuracy model is established; a high accuracy systematic error compensation algorithm and an all-sky autonomous star identification method based on Fourier transform correlation are proposed. Simulations and experiments are performed to verify these algorithms. The research results establish basic error model for the development of star tracker, and can be used as a basic direction of the star tracker parameter designing and star information processing.