UAV Autonomous Landing using Visual Servo Control based on Aerostack
Guanzheng Wang, Zhihong Liu, Xiangke Wang · Proceedings of the 3rd International Conference on Computer Science and Application Engineering · 2019
The autonomous landing for unmanned aerial vehicles (UAVs) plays an important role in driving the unmanned systems towards practical applications. However, the traditional position-based landing control method relies heavily on satellite navigation and positioning systems, which cannot be applied in environments with weak satellite signals such as indoors and valleys. In order to solve these problems, this paper proposes a visual servo control method for the UAV autonomous landing based on Aerostack software framework. The method takes the real-time on-board camera image as input and extracts tag features on the ground. In addition, the autonomous landing visual servo law is then constructed to obtain the velocity control signals which are able to guide the drone to land accurately. The experimental results in the Gazebo simulation environment show that the proposed algorithm can achieve robust and accurate autonomous landing without satellite navigation signals. Most specifically, the distance deviation is less than 0.1 meters and the yaw deviation is less than 0.1 radians.