Model-aided wind estimation method for a tail-sitter aircraft

Abstract

This paper presents a wind estimation method for a dual-rotor tail-sitter unmanned aerial vehicle with all flight phases. The large flightenvelope and the slipstream generated by the propellers introduceextra challenges to estimating the wind field during flight for dual-rotor tail-sitter aircraft. In this method, a synthetic wind measurementis proposed based on a low-fidelity aircraft model and operated as avirtual sensor. This synthetic wind measurement and the data from thepitot tube are fused with an extended Kalman filter. The simulationand experimental results of the developed estimation method showa good estimation of the wind speed and direction in the hoveringphase, transition, and cruising phases. The proposed wind estimationmethod was also tested in the hovering phase using the aerodynamiccoefficients of NACA 0012 airfoil and a flat plate instead of the vehicle’smodel to provide a compromise solution for vehicles with no preciseaerodynamic model.