Estimation of turbulence and state based on EKF for a tandem Canard UAV

This paper deals with the state and turbulence estimation of a model describing the longitudinaldynamics of an Unmanned Aerial Vehicle (UAV). Due to both the high nonlinearities of the model and thestochastic nature of disturbances, an Extended Kalman Filter (EKF) is proposed. To allow the estimator to beemployed on low cost UAV systems, it is assumed that the aircraft is equipped with a low performance GPS,characterized by a relatively low refresh rate. The designed EKF is able to work efficiently in both turbulent andcalm atmosphere. In order to obtain information about the performances of the proposed estimator for controlpurposes, a control system, consisting of the EKF, a PID-type controller and the longitudinal dynamic model ofthe UAV, is implemented in Matlab-Simulink environment with the aim of verifying the effects of the estimationerrors on the tracking of the reference signals. The obtained results are fully satisfactory.