Image-based Visual Servoing of Unmanned Aerial Vehicles for Variable Angle Target

This paper presents a double virtual plane method for a quadrotor unmanned aerial vehicle(UAV) using a monocular camera and an inertial measurement unit sensor(IMU). Using the above method and an image-based visual servoing controller, the UAV can achieve the tracking of the target both on the horizontal plane and on the inclined plane. Utilizing the image moments, the image features are defined and and adopted by the IBVS controller. The inclination angle of the target plane can be obtained through the image distortion in the horizontal virtual image plane. The closed-loop system is proved globally asymptotic stable by means of Lyapunov analysis. And the simulation of the method is completed, which can be used to prove the correctness and stability of the theory.