|MIT Department: Aeronautics and Astronautics
Faculty Mentor: Prof. Richard Linares
Undergraduate Institution: University of Nebraska-Lincoln
I am a Mechanical Engineering student with a Mathematics minor at the University of Nebraska-Lincoln. I am also an international student from Porto Alegre, Brazil. As an undergraduate student, I have been involved with multiple classes as teaching assistant, engineering tutoring, and research. My research focuses on determining how fighter jet dynamics changes when there are multiple nonlinear attachments coupled to the wings, and why such difference in behavior occurs. My career goals are to pursue a PhD in Aerospace Engineering, and to work with research and development or test engineering in the aerospace industry. In my free time I like to travel, play the piano, listen to music, and read about different scientific concepts.
Deep Artificial Neural Network Model for Improved Thermospheric
Guilherme M. Eymael1, Peng Mun Siew2 and Richard Linares2
1Department of Mechanical and Materials Engineering, University of Nebraska-Lincoln
2Department of Aeronautics and Astronautics, Massachusetts Institute of Technology
Given the rising number of satellites and space debris, there is an increasing demand for more efficient space traffic management. For that, it is critical to better determine and predict satellites’ orbits. Currently, the most significant source of error for satellites’ orbits prediction is the drag force calculation, due to inaccurate estimation of the thermospheric mass density. Therefore, we are working on a machine learning reduced-order model (ML-ROM) as a new approach for thermospheric density prediction to overcome the limitations imposed by other methods. Our model is trained via population based training and uses dynamic mode decomposition with control for data propagation in time. The results are then compared to the proper orthogonal decomposition reduced order model (POD-ROM) for validation. Our initial ML-ROM accuracy was found to be similar to the POD-ROM. Our ML-ROM’s inaccuracy gradually increases with time due to error propagation. Moreover, the most significant absolute percentage errors in the ML-ROM predictions correspond to times of high solar and geomagnetic activities. In conclusion, with better training and incorporation of a recurrent neural network for data propagation in time, our model has potential to allow for better thermospheric density prediction. This will decrease the probability of satellites/debris collisions, improving space traffic management.