UBC Theses and Dissertations
Fault detection and diagnosis in a spacecraft Attitude Determination system Pirmoradi, Fatemeh
Spacecraft systems need increased on board autonomy to detect the occurred faults (e.g., faults in sensors and actuators), isolate the faulty components, and effectively handle their operation in the presence of such faults. In this research a scheme of fault detection and diagnosis is developed for spacecraft Attitude Determination (AD) sensors along with an algorithm for attitude determination. These together can provide a significant degree of autonomy since faults can be handled without ground interaction and intervention. The attitude determination system is a key component of the attitude control system of a spacecraft. Hence, improvements to its accuracy and reliability contribute directly to its performance and the success of the spacecraft mission. An integrated AD system that includes rate gyros and vector sensors (Sun sensor and magnetometer) is developed. Measurement data from all sensors are fused by utilizing a linearize Kalman filter, which is designed based on system kinematics, to provide attitude estimation and the values of gyro bias. The designed estimator is shown to provide more accurate estimates of the attitude angles than the measurements obtained directly from sensors. This results in the removal of erroneous sensor measurements, and prevents unbounded sensor measurement errors. Bias-free data are then used in the Fault Detection and Diagnosis (FDD) scheme.