Date of Award

Fall 2021

Publication Type


Degree Name



Mechanical, Automotive, and Materials Engineering

First Advisor

R. Razavi-Far

Second Advisor

J. Ahmed

Third Advisor



Attitude determination and control system (ADCS), Fault detection and isolation (FDI), Machine Learning, Reaction Wheel, Satellite, Spacecraft



Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


With the increasing number of satellite launches throughout the years, it is only natural that an interest in the safety and monitoring of these systems would increase as well. However, as a system becomes more complex it becomes difficult to generate a high-fidelity model that accurately describes all the system components. With such constraints using data-driven approaches becomes a more feasible option. One of the most commonly used actuators in spacecraft is known as the reaction wheel. If these reaction wheels are not maintained or monitored, it could result in mission failure and unwarranted costs. That is why fault detection and isolation, which is detecting anomalies in real-time and finding the root cause of the failure, is crucial.

This work proposes a novel approach for a data-driven machine learning technique for detecting and isolating multiple in-phase faults in nonlinear systems, with a case-study for an in-orbit closed loop-controlled satellite with reaction wheels as actuators. The proposed method uses a hierarchical approach with automated feature ex-traction, feature reduction and model selection. The method is also studied on three different datasets and three configurations. The results yield a performance accuracy of 98.91%, 97.87%, and 98.02% for all three configurations, respectively. Further-more, sensitivity analysis which includes missing values, missing sensors, and noise, are applied against the proposed method to test its robustness.