An analytical fault detection algorithm using Euler angles-numerical and practical results

This paper presents a new scheme for fault detection in satellite Attitude Determination (AD) sensors. The proposed method is based on derivation of all possible rotations between orbital and body frames and comparison of Euler angles provided by them. Accordingly, significant variations in the variance of the obtained Euler angles sets utilized as a benchmark for fault detection. The features which have been achieved by this philosophy make it possible to provide an analytical approach for fault detection without imposing of any additional sensors, mass, power consumption and costs. Also, the outlined method is a model-free based approach which allows combination of this algorithm with the model-based algorithms that used in the attitude control subsystem. So it provides the detection of sensor faults independent of the faulty conditions of actuators. The numerical simulations presented in this paper demonstrate the accuracy and robustness of the designed algorithms. Also, designed algorithms are implemented using the hardware in the loop test bed together with manufactured space sensors to evaluate their functions in a practical situation.