International Finalist for NASA RASC-AL 2024 “AI-Powered Self-Replicating Probes” Research Competition.
Year
2023-2024
As part of a multidisciplinary team selected as International Finalists in NASA’s RASC-AL competition, I helped architect an autonomous spacecraft concept designed for sustainable long-duration operations in the Asteroid Belt.
Our design combined reusable VTVL landing systems, multi-modal propulsion (metallic and water-derived), dynamic resource extraction, and an AI-driven control protocol capable of interpreting and acting on mission objectives without ground intervention. I synthesized over 30 domain-specific research sources into a generative AI model (GPT) tailored to provide spacecraft commands as an proto-example of AI-embedded decision workflows for autonomous systems.
In parallel, I developed and validated a machine learning model to classify asteroids by compositional taxonomy (Bus–DeMeo SMASS) using Python and Scikit-Learn. By curating NASA spectral datasets and normalizing feature ranges, I trained a Random Forest classifier that achieved high multi-class performance (0.89 F1), effectively mapping sensor-measurable spectral features to Asteroid composition categories.
Disciplines
Machine Learning
Technical Documentation
Mathematical Modeling
Mechanical Design
Data Analysis
ConOps
R&D
