Tail-sitter unmanned aerial vehicles (UAVs) have broad application prospects since they merge advantages of both fixed-wing UAVs and rotary-wing UAVs. However, there exist great challenges in the transition maneuvers due to model un-certainties and external disturbances. Aimed at these problems, a robust transition controller based on incremental nonlinear dynamic inversion (INDI) is developed for a tail-sitter UAV in this paper. Different from existing works, the controller mainly concerns the transition pitch angle and altitude, because the altitude is more intuitive then flight speed in indicating whether a transition is successful or not. The robustness of the developed transition controller is analyzed with consideration of error terms exist in the closed-loop system, which were generally omitted in existing INDI flight control works. With some reasonable assumptions, it is proven that the tracking errors can converge into a specified neighbourhood of the origin in a finite time by choosing appropriate controller parameters. Numerical simulations demonstrate the robustness of the controller in handling model uncertainties and external disturbances.

This paper presents the development and hovering control of a tailsitter unmanned aerial vehicle (UAV) that merges long endurance and vertical takeoff and landing (VTOL) abilities. The designed tailsitter contains one flying-wing with two motors and two elevons. Vehicle aerodynamics and a six-degrees-of-freedom (6-DOF) model are especially developed for the tailsitter. To achieve a good performance in outdoor stationary hovering and accurate vertical flying, the active disturbance rejection control (ADRC) for attitude controller is proposed. With signals from extended state observer (ESO) and tracking differentiator (TD), ADRC decouples the system model into a controllable chain of integrators. Based on the decoupled system dynamics, the motion of tailsitter can be easily handled by developed position controller. Experimental results are presented to corroborate the effectiveness of the controller in disturbance rejection.