Data-Driven Discovery of Predictive Aeroheating Models for EDL Missions

Validated aerothermodynamic prediction is identified as one of the highest priority technical gaps that must be addressed for both future human Mars and robotic Entry; Descent; and Landing missions. Despite a wealth of wind-tunnel tests and theoretical analyses spanning 60 years; the effect of roughness on aeroheating predictions remains poorly understood and reliable predictive models have not yet been developed. We propose a novel Machine Learning framework that can distill physical laws from noisy wind-tunnel data which can be infused into NASA’s aerothermal prediction tools. The proposed framework combines Deep Learning and Symbolic Regression to learn mappings from high-dimensional feature spaces to algebraic; predictive model expressions; capable of extrapolating beyond wind-tunnel conditions. If successful; the project could revolutionize how we predict turbulent roughness heating augmentation and more broadly; how we distill wind-tunnel data over a range of applications.

This project has the potential to revolutionize how we predict turbulent heating augmentation due to roughness effects by automatically discovering hidden physical models from noisy wind-tunnel data through a novel Deep Learning and Symbolic Regression approach. If successful; the project will provide a radical new tool for distilling wind-tunnel data across a range of application domains and could lead to new and exciting avenues of aerothermodynamic research at NASA.

This is early/mid-stage (TRL 2) — 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.