Browse Teach Curate

Physics-Informed Gaussian Processes for Aircraft Stability-Derivative Estimation

A method that places a known aerodynamic model as the mean function of a Gaussian process so the data-driven model only learns deviations from established physics, then differentiates the smooth GP posterior at trim conditions to recover stability derivatives and short-period dynamics from arbitrary, non-standardized flight-test data with uncertainty quantification.

KITT Text

Physics-informed Gaussian Processes for Safe Envelope Expansion D. Isaiah Harp* †, Joshua Ott* ‡

This paper estimates aircraft aerodynamic stability derivatives from arbitrary, non-standardized flight-test data using a Gaussian process (a distribution over functions defined by a mean function an…