Extended Kalman Filter Based Estimations for Satellite Attitude Control System

This paper mainly focuses on the maneuver of the satellite in orbit. A non-linear multi-inputs multi-outputs model has been derived from Newton-Euler equations of motion. The dynamics is presented with control methodologies allowing the Extended Kalman Filter (EKF) to iteratively provide improved data sets with zero errors. As the system is distracted from the atmospheric swings which are random hence the problem of stochastic disturbance is furnished. A set of differential equations of two dimensional Ito stochastic type is used for modeling the said disturbances (before t = 4s is recorded). The attitude parameters are recorded in RT-LAB setup with the Extended Kalman Filter (EKF) providing adequately superior estimation outcome which thereby makes the filter more appealing. With the presence of Gaussian noise in both dimension and system, Extended Kalman Filter gives the correct estimates. It’s collaboration with hardware setup RT-LAB is commendable. Hence, an Extended Kalman Filter which deals with such nonlinear models proves to be a higher choice for achieving best online results. A comparison reflecting the tracking and stable control of the satellite for the designed advanced adaptive robust controller (AARC) for two situations is plotted. The priority of making the system stable in the presence of stochastic disturbance is also visited. Also, the use of three different values of the confounding variables revealed that the control weighting line is completely diminished thereby boosting the tracking when the satellite is in orbit. Moreover, the previous research involves methods to improve satellite communication on ground station, this paper deals with exact positioning of concerned satellite attitude parameters and its validation tested experimentally on OPAL-RT hardware. To sum up, the development of advanced adaptive robust controllers have encouraged the stability and accuracy of systems considering the varying atmospheric conditions. The simulation results predict perfect tracking of output with respect to the desired set-point in the presence of stochastic disturbance for the proposed controller.