Autonomous Landing of a Rotor Unmanned Aerial Vehicle on a Boat Using Image-Based Visual Servoing
Lingjie Yang, Zhihong Liu, Xiangke Wang, Guanzheng Wang, Xinyu Hu, Yexun Xi · 2021 IEEE International Conference on Robotics and Biomimetics (ROBIO) · 2021
This paper proposes an image-based visual servoing (IBVS) method for a rotor unmanned aerial vehicle (UAV) to land on a boat with a downward-looking camera. The controller is designed in the image plane directly with the aid of image Jacobian matrix. In order to decouple the horizontal translation from the rotation of pitch angle and roll angle of the UAV, a virtual camera method is adopted to compensate the effect of UAV’s attitude changes. For the purpose of enabling the UAV to land from a high altitude, we design a nested 2D tag to ensure the detection within a large altitude range. However, due to the shake of the boat, the landing tag is no longer horizontal, which increases the difficulty of obtaining the desired position of the feature point. To solve the problem, we solely use the tag center in the image as the input, then the desired position of it will coincide with the image center of the virtual camera, irrespective of whether the tag is horizontal. Furthermore, Kalman filter is introduced to optimally estimate the position of the feature point with observation noise. Finally, simulation experiments based on Gazebo and the field experiments based on the deck simulation platform are conducted. The results show that our method can achieve a precise landing within errors of 0.2m from altitude around 50m.