Wind Shear On-Line Identification for Unmanned Aerial Systems

Abstract

An algorithm to perform the on line identification of the wind shear components suitable for the UAS characteristics has been implemented. The mathematical model of aircraft and wind shear in the augmented state space has been built without any restrictive assumption on the dynamic of wind shear. Because of the strong velocity variations typical of wind shear induce severe accelerations on the aircraft the wind shear effects have been modeled as external forces and moments applied on the aircraft. The identification problem addressed in this work has been solved by using the Filter error method approach. An Extended Kalman Filter has been developed to propagate state. It has been tuned by using a database of measurements through off-line identification of the process noise covariance matrix. Afterwards the implemented EKF has been employed to estimate onboard either aircraft state or turbulence, with significant savings in terms of time and computing resources. Robustness of implemented algorithm has been verified by means of several tests. The obtained results show the feasibility of the tuned up algorithm. In fact it is possible, by using a few numbers of low cost sensors, to estimate with a noticeable accuracy the augmented state vector. Besides a very short computation time is required to perform the augmented state estimation even by using low computation power.