Real-Time Attitude Estimation Algorithm of Flapping-Wing Air Vehicle Under Large Vibration

Flapping-wing air vehicle (FAV) is a new concept aircraft designed to imitate the flight of birds or insects based on the principle of bionics. Unlike conventional fixed-wing or rotorcraft, the real-time on-board attitude estimation of FAV is very difficult due to fuselage severe transient oscillation excited by high-frequency wing flapping. In this study, an attitude estimation algorithm combining second-order complementary filtering algorithm and Extended Kalman filtering algorithm (CEKF) are proposed. The Inertial Measurement Unit (IMU) data is first processed by second-order complementary filtering algorithm, and the resulting attitude value is processed as the input of the EKF to derive the FAV’s real-time attitude angle. Thro ugh the processing of two basic filtering algorithms, noise errors that could not be processed can be eliminated, thereby improving the accuracy of the attitude solution and achieving the purpose of real-time attitude solution. The proposed method is verified by MATLAB-based simulation. The accuracy, convergence and robustness of the algorithm are approved.