Research on the autonomous shipboard landing algorithm of the VTOL aerial vehicles

Abstract

The objective of this thesis is to design the autonomous shipboard landing system for various VTOL Unmanned Aerial Vehicles using Time-Delay Control. The autonomous shipboard landing algorithm is composed of two parts: one is the guidance to generate velocity and heading command, and the other part is the controller to control the vehicle and follow the given command input. Designed guidance is divided into 4 steps depending on the relative distance. In addition, the different landing procedure is conducted according to the cross wind factor of the relative wind between the ship and the aircraft. Proposed guidance uses the distance error as feedback, so that the generated velocity commands ignore the attitude of the ship and the aerial vehicle. Two frames fixed at the ship and the aircraft are defined, respectively. Considering the attitude and two frames, the body velocity of the objective of the control, VTOL UAV, is obtained through velocity vector transformation. The velocity command is passed to the controller. In this thesis, TDC + PD compound controller is newly designed. The model of the VTOL aerial vehicle is usually changed depending on the flight velocity. The operation environment of the sea field has various kinds of unexpected factors. To obtain the robustness against model uncertainty and sea environment factors, Time-Delay Control which is one of the adaptive controls is designed. By combining with PD control, TDC compensate the weak point and highlight the strong points. Designed autonomous shipboard landing system is evaluated by X-Plane based simulation. Proposed Compound controller shows advanced performance than existing conventional controller. The sea operation environment is described in detail, and the average error of the touch-down point is less than 1m. Furthermore, proposed algorithm is applied to various types of the VTOL vehicles and proofed of its expandability. During the presence of the cross wind, the algorithm makes s...