Extended and Unscented Kalman Filters for attitude estimation of an Unmanned Aerial Vehicle

Precise position and attitude estimation is necessary for guidance, navigation and control of small Unmanned Aerial Vehicles (UAVs). With limited payload capabilities and low-cost sensors it becomes necessary to implement robust estimation techniques that can handle varied amount of system nonlinearities and noise. With limited accuracy of such sensors, the need for efficient and accurate estimation methods become vital. This paper compares and contrasts Monte Carlo simulation results as obtained for an Extended Kalman Filter (EKF) and an Unscented Kalman Filter (UKF), towards the attitude estimation of a small UAV.