Local Three-Dimensional Wind Estimation for Fixed-Wing UAVs via a Physics-Informed GRU with Measurement-Noise-Adaptive Kalman Smoothing

Reliable local three-dimensional (3D) wind estimates are important for fixed-wing UAV flight under wind disturbances, but low-cost platforms lack direct 3D flow sensing. We propose PIRNN-AKF, which combines a physics-informed gated recurrent unit (PI-GRU) with a measurement-noise-adaptive Kalman smoother. PI-GRU embeds the wind triangle and attitude rotations in an airspeed-closure loss and predicts covariance scales for adaptive fusion. In PX4/JSBSim simulations with Dryden turbulence, five-seed PI-GRU attains a 3D RMSE of 0.208±0.003 m/s and a direction MAE of 3.22° on Test-ID. PIRNN-AKF attains 0.544±0.007 m/s and 1.97° on Test-OOD, reducing RMSE by 37.4% relative to Vanilla GRU; KalmanNet yields lower OOD RMSE but weaker ID accuracy and direction estimation. Session-aware evaluation shows that the AKF reduces OOD jitter by 24.8% while preserving RMSE. In ten frozen HITL sessions of the final 41-input, 100-step model, ID/OOD RMSEs are 0.569±0.054/0.832±0.318 m/s, the companion-side processing p95 is 20.4±0.5 ms, and the achieved loop rate is 83.2±1.6 Hz with 5.6±0.7% deadline misses. These results support 50 Hz feasibility under simulation truth; real-flight accuracy and closed-loop benefits remain untested.