The focus of this project is on the flight testing of analytical foundations of machine learning algorithms and their intersection with parameter estimation in control. Machine learning algorithms will be implemented alongside controllers to update a system’ s regulating control law as the dynamics of the system change. The resulting closed-loop system will not only be robust to modeling and environmental uncertainty but will learn parameters that minimize a performance objective function. These algorithms will be simulated in MATLAB/Simulink and then tested on an aerial vehicle through collaboration with Aurora Flight Sciences.

I am participating in SuperUROP to gain experience researching controls before I attend graduate school and continue to do more controls research. SuperUROP will provide me with the planning, writing, and presentation skills I will need to effectively communicate my work whether it be in graduate school or afterwards. Furthermore, the project I am working on brings together machine learning and controls, which could transform both fields.