Image-based finite-time visual servoing of a quadrotor for tracking a moving target

This paper proposes an image-based visual servoing control method for a moving target of quadrotor UAV (QUAV). Firstly, the dynamic image model with parameters is established based on moment features in virtual camera plane. For unpredictability space, we use high-order differentiator to estimate state target. In order solve problem depth information caused by monocular camera, derive nonlinear finite-time linear velocity observer from plane, which can not only QUAV but also avoid measurement depth. Based above information, design global controller and Lyapunov theory prove stability system. Finally, numerical simulations verify convergence proposed scheme, ROS gazebo demonstrate improved performance scheme tracking error.