Wind gust estimation on a small VTOL UAV

This paper describes a Kalman filter based gust identification technique for estimating wind gusts on a small rotary-wing unmanned aerial vehicle (UAV) during flight. The algorithm is designed to estimate wind gusts using only inertial, position, and control actuation sensors (no air data measurements). The technique is demonstrated using flight tests of an off-the-shelf unmanned quad-rotor UAV. An accurate dynamic model of the aircraft was developed using system identification techniques, and the model is used in a Kalman filter to estimate the external wind disturbances. Testing was conducted in an indoor flying lab using a high speed fan to simulate gusts and a motion tracking system to provide accurate state measurements. The motion capture system was used for both closed loop control of the aircraft and in the gust estimation algorithms. The technique can be used to enhance gust rejection on small UAVs or to survey wind fields around objects to understand turbulent airwakes and to validate physics models of these flow fields.