← Back to NASA Technology Projects
Completed TRL 6 (started at 4, targeting 6)
We propose to apply Machine Learning (ML) algorithms to improve retrievals by minimizing instrument bias effects, for both orbital and in-situ spectrometers. We anticipate ML will better discriminate against interferences and spurious instrument effects and drastically reduce data volumes compared to existing theoretical and empirical methods. The proposed work would enhance instrument (sensor) design, enable on-board data reduction, and would be adaptable to virtually any present or future planetary mission or instrument that would carry a spectrometer.
The ultimate science goal and payoff to NASA and GSFC would be to develop an in-house ML capability for planetary spectroscopy applications. This would be a widely applicable toolset that would benefit early spectrometer instrument design and optimization, in contrast to using ML to speed up only onboard data processing. In the design of an active (or passive) spectrometer for planetary applications we seek to minimize the required resources (volume, mass, power, complexity/risk, and cost) while still achieving the measurement precision and accuracy required to meet the science objectives. A problem related to spectrometer design is the correction of instrument bias effects in spectrometers. All spectroscopic instruments (in-situ, or remote sensing) have systematic and random errors that limit their accuracy and precision. Machine learning is ideally suited to correct spurious instrument effects (biases) without having precise knowledge of the underlying physical model.
Listed on TechPort itself — the most direct way to ask about this specific project.
This is early/mid-stage (TRL 6) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.
None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.