Abstract:

First of all, the dissertation calibrates the gyro model. Then the attitude quaternion from charge coupled device star sensor’s output and the other attitude quaternion that carrier relative inertial space of  SINS’s output are used to construct the observation, a model predictive filter(MPF) and an extended Kalman filter(EKF) are integrated for use. In this method, parts of inertial element errors are regarded as the model error and are used to estimate the part of the solution. And the algorithm could be implemented on line for estimating both filter measurements and state trajectories. Finally, the algorithm is simulated in MATLAB. It shows the effectiveness and accuracy of this technique, which has a strong practical significance. So it takes great conveniences for application and maintenance.