Real-time approaches to the estimation of local wind velocity for a fixed-wing unmanned air vehicle

Abstract

Three real-time approaches to estimating local wind velocity for a fixed-wing unmanned air vehicle are presented in this study. All three methods work around the navigation equations with added wind components. The first approach calculates the local wind speed by substituting the ground speed and ascent rate data given by the Global Positioning System (GPS) into the navigation equations. The second and third approaches utilize the extended Kalman filter (EKF) and the unscented Kalman filter (UKF), respectively. The results show that, despite the nonlinearity of the navigation equations, the EKF performance is proven to be on a par with the UKF. A time-varying noise estimation method based on the Wiener filter is also discussed. Results are compared with the average wind speed measured on the ground. All three approaches are proven to be reliable with stated advantages and disadvantages.