Unmanned Aerial Vehicle Attitude Measurement Based on Adaptive Federated Filtering Algorithm

In order to realize high-precision tracking of unmanned aerial vehicles (UAV), an adaptive federal filtering algorithm with outlier detection is proposed in this paper. First, in the information allocation stage, an adaptive factor allocation scheme is designed for the information detected by the motion processing sensor MPU6050 and the magnetometer, which improves the efficiency of information allocation; second, in the information fusion stage, in order to reduce the influence of the error data information on the results, an outlier detection algorithm is proposed, which determines whether the data is outlier or not through the theory of Dempster-Shafe(D-S) evidence; and finally, a linear minimum variance estimation for fusion to get more accurate estimation results of attitude solving. Simulations and experiments verify that the algorithm can solve the more accurate UAV attitude information.