A Nonlinear Model-Based Wind Velocity Observer for Unmanned Aerial Vehicles

Abstract

This paper presents an exponentially stable nonlinear wind velocity observer for fixed-wing unmanned aerial vehicles (UAVs). The observer uses a model of the aircraft combined with a GNSS-aided inertial navigation system (INS). The INS uses an attitude observer together with a pitot static probe measuring dynamic pressure in the longitudinal direction as well as the airspeed. The observer is able to estimate the wind velocity and from this compute the relative velocity, which directly contains information about the angle of attack (AOA) and sideslip angle (SSA). The nonlinear observer is also able to estimate the scaling factor of the pitot static probe measurement and there are no requirements on persistence of excitation (PE) of the UAV maneuvers. The computational footprint is smaller than the conventional Kalman filter, which makes the algorithm well suited for embedded systems. The designed observer is proven exponentially stable under stable flight and through simulations it is verified that the estimates converge to the true values of a realistic wind velocity when there are no model errors.