Spacecraft Attitude Determination For Multiple Fields Of View Star Sensors

Star sensor is widely used in a majority of spacecraft attitude determination missions due to the high accuracy. No matter what geostationary satellite(GEO) or deep space explorer, large scale space platform or small satellite, attitude determination systems almost adopt star sensor. The new generation multiple fields of view(FOVs) star sensor and the pertinent problems are studied, and the main results achieved in this dissertation are summarized as follows:The general plan of double FOVs star sensor is preliminarily designed. The principle of star map simulation is studied, and the navigational stars used for the single and double FOVs star sensor are selected from the Tycho-2 star catalogue. The first moment method of image centroiding is introduced, and the FOV identification technique is studied. The imaging principle of double FOVs star sensor is studied and the primary parameters are given, and the mehtod used to estimate the misalignment matrix of beam-splitting mirror is presented.The Pyramid star identification algorithm is studied. On the basis of the K-vector search technique, the Pyramid star identification is studied and improved for the structure characteristics of double FOVs star sensor. Simulation results indicate that the speed and success rate of the proposed star identification algorithm is high. The attitude determination algorithms for star sensors are studied. The several typical attitude determination algorithms using vector observations are discussed, and the numerical examples are given for the comparison of the performance. The quaternion extracting from rotation matrix and the averaging attitude matrices and quaternions are studied, which expands the application range of Wahba problem.The multiplicative extended Kalman filter (MEKF) for spacecraft attitude determination is developed. For the arbitrary reference frame, the linear quaternion kinematics equation and the dynamical model of three-component attitude errors are formulated, and gyro-aided and gyro-free attitude determination schemes are designed.The typical basic equation of the MEKF is developed, including the detailed models for vector measurements, quaternion measurements and Euler angle measurements. For the nonlinear estimation problem of gyro-free attitude determination system, the second-order nonlinear multiplicative filtering algorithm is designed. The discretization algorithm related to the continuous system is presented by using the quadratic form of matrix theory, and the integration of Kalman filter on computer is derived.