06.1 - Flight Dynamics and Control (Control & Modelling)
PHYSICS-INFORMED MACHINE LEARNING FOR FLIGHT DYNAMICS IDENTIFICATION OF THE SAAB 340B USING FLIGHT TEST DATA
A. Navarro¹, M. Alam¹, A. Elliott¹, E. Jinks, Siemens Digital Industries Software, United Kingdom; ¹Cranfield University, United Kingdom
A physics-informed machine learning framework is applied to identify the short-period longitudinal dynamics of the Saab 340 using flight test data. A hybrid neural architecture estimates reduced-order model parameters constrained by the equations of motion. Automated hyperparameter optimisation using Siemens HEEDS improves prediction accuracy relative to manual tuning, demonstrating enhanced modelling fidelity and workflow efficiency.
DOI: 10.71945/icas2026_0686