Attitude Estimation Algorithm of Flapping-Wing Micro Air Vehicle Based on Extended Kalman Filter

Due to its uniqueway of flying, the flapping mechanism of the flapping-wing micro air vehicle (FWMAV) will generate severe instantaneous oscillations which is far greater than traditional flight platforms, brings great challenge to its sensor data acquisition and attitude estimation. To overcome these drawbacks, this paper proposes a set of algorithms for data fusion and attitude estimation of FWMAV based on the Extended Kalman Filter (EKF) algorithm. The second-order low-pass filter and dynamic window smoothing filter are used to preprocess the data of the gyroscope and accelerometer. And an ellipsoid fitting algorithm is designed for the electronic compass to process Earth's magnetic field data. Finally, based on the EKF algorithm, this paper dynamically corrects the estimated attitude using data of the accelerometer and electronic compass, maintains the weights of different data in real time, and performs sensor data fusion. The experimental results demonstrate that, compared with the traditional algorithm, the algorithm proposed in this paper manifests an excellent function of filtering out platform jitter and estimating the real attitude.