The Research Of On-Orbit Calibration Algorithm Of Star Sensor

On-orbit calibration is an important part of the high-accuracy satellite ADCS(attitude determination and control system). Base on different sensors,the paper has designed several algorithms for calibrating the system often value error of star sensor.On the basis of presentation attitude descriptive methods,attitude kinematics,the least squares optimal estimation and kalman filter theory,studies the realizing principle of estimate star sensor errors,designs the on-orbit calibration algorithm,and then gives the simulation results.The topics presented in this paper are as follows:This paper has researched autonomous on-orbit calibration algorithm of star sensors with gyros.Since the gyros can get sequential attitude angles that it selects the satellite's kinematics as kalman filter state equation.The star sensor main error is misalignment.For correcting the star sensor errors effects,it's useful that calibration star sensors on-orbit to improve accuracy of the satellite attitude determination.This algorithm can improve star sensor's precision efficiently.It estimates the principal point offset and the focal length with the help of Kalman filters.Star sensor is calibrated in laboratory before launch,but the geometry may change by the strong acceleration of the launch and by thermal influence of the sun.The camera is designed to be autonomously self-calibrating.Star camera calibration plays an important role for the case of the spacecraft attitude estimation problem.A novel algorithm has been developed to estimate the principal point offset and focal length by using measurements independent of attitude.It has researched an arithmetic which to calibrate the star sensor on-obit that based on landmark sensor.The arithmetic is designed by using the result of image matching.Considering landmark measuring error,a novel algorithm is designed for filter the noise out,by use the EKE and combine the result of LS.This algorithm can improve star sensors'precision efficiently.Simulation results indicate the availability of the on-orbit calibration algorithm.Base on satellite payload,this paper has researched algorithm of on orbit calibration of star sensor.Using payload imaging the target star , three algorithms for on-orbit calibration of the system often value error of star sensor are developed.LS and kalman filter and LS estimates are used as "measurements" for a recursive Kalman filter to filter out the noise as a new filter . Then compare these algorithms after simulation.Simulation results indicate that the new algorithm is most availability.All simulation results indicate the availability and accuracy of the on-orbit calibration algorithm.