The wind disturbance in flight induces measuring error of air data, including the true airspeed, angle of attack, and sideslip angle. To obtain the air data with better accuracy, a synthetic filtering system integrated with a customized wind model and an adaptive square root Kalman smoother (ASR-UKS) were proposed based on the flight data of civil aviation aircraft. The Gaussian process (GP) regression was first used to extract the random and rapid-changing turbulence that deteriorated the air-data measurements, and a parameterized turbulence model was built by autoregressive (AR) modeling. After this, a synthetic filtering system integrated with the recursive turbulence model was developed based on the inertial measurements in flight data. Finally, the ASR-UKS was designed to estimate the airspeed, angle of attack, and sideslip angle within the finite time span of flight data. Simulation results indicate that the customized wind model is able to recover in situ wind series experienced by the aircraft, and the synthetic filtering system is able to track the true value of air data and wind series well. Furthermore, the ASR-UKS, characterized by fully using the finite-length flight data and updating the noise covariance matrices adaptively, is able to reduce the estimation error and counteract the adverse effects of uncertain noise in flight data as well. A further test with real flight data indicates that the proposed method gives the refined estimation of airspeed, angle of attack, and sideslip angle in wind disturbance. (C) 2022 American Society of Civil Engineers.
展开▼
