About the role#
Reliable power systems are essential for deep space exploration. Many spacecraft faults occur that were not part of the original design, creating challenges for detection and control. This project explores how machine learning can improve fault management and forecasting for NASA space power systems.
What you'll need#
- Background in engineering, computer science, or a similar field.
- Proficiency in MATLAB, Python, and C/C++.
- Experience developing machine learning algorithms.
- Strong mathematical foundation in linear algebra, differential equations, and optimization.
- Solid communication and teamwork skills.
- Experience with Simulink is preferred.
- Experience with advanced control theory and dynamical systems is preferred.
What you'll do#
- Derive physics-informed machine learning techniques to support fault diagnostics and control of space-based power systems.
- Develop forecasting capabilities across various time horizons.
- Work full-time on-site at the Ames Research Center.
- Participate in weekly meetings with your mentor.
Location & details#
- Location: Moffett Field, California.
- Term: Summer 2027 (June 7 to August 13).
- Work Modality: On-site.
- Citizenship: US Citizen only.
- Education: Open to undergraduate juniors, seniors, Master's students, doctoral students, and post-doctoral researchers.


