Development and Performance Analysis of UKF Models for Satellite Position and Attitude Determination

Abstract

This theses focuses on the development of models of satellite physics and attitude sensors, adaption of environmental models to be used in a highly optimised UKF algorithm for satellite attitude and orbit determination and the performance analysis of the system based on simulation data. During the writing of this theses a basic framework was set up to adapt the UKF algorithm for use in a satellite attitude determination environment and the star tracker, gyro and sun sensor models were developed. Several environmental models were adapted to work with the sensor models. Regarding orbital dynamics, a satellite communication ground station based orbital position correction method was proposed that, according to the knowledge of the author, has not been used on small satellites before. The method utilises the precise measurement of the communication signal return time to apply a correction factor to the orbital location of the satellite. To test the method, a ranging sensor model was devised. All sensors and the sensor fusion capability of the UKF were tested under varying operating conditions, which included different sensor combinations, sensor errors and different attitude manoeuvres. The required simulation data was generated using a simulation environment developed by the author. It contains the necessary physics and environmental models to produce realistic data. All models and the UKF algorithm were found to work well and the results match the accuracy configurations of the different sensors. The ground station ranging based orbital correction was shown to work well with considerably larger errors than would be encountered in actual orbit.