Correlation Analysis Based Direct Identification of a Quadrotor Generalized Attitude Model Using Outdoor Flight Data

This paper studies the direct identification of a quadrotor generalized attitude model (GAM) by utilizing the collected outdoor flight data corrupted with strong random wind disturbance. In order to diminish the ill effects caused by such disturbing winds on the identification accuracy, an improved correlation computation scheme is designed and then the prediction error method is applied to estimate the GAM and its associated noise model. Finally, model validations are given to demonstrate the effectiveness of the proposed identification procedure and the resulting estimated GAM.