Image-based fixed-time visual servoing control for UAV landing on a moving platform with visibility constraints

Image-based visual servoing control for UAV landing poses significant challenges due to the UAV’s under-actuated characteristics, lack of relative velocity measurements, and visibility constraints. In this paper, a visibility constrained fixed-time control method is proposed for UAV landing on a moving ground vehicle. Firstly, the image features are extracted by using the perspective projection model and their dynamics are accordingly derived. Subsequently, a barrier function based backstepping controller is developed to eliminate image feature errors in fixed-time, while ensuring the visual target is always confined within the camera’s field of view. In particular, the relative velocity between the two vehicles and time-varying disturbance components are estimated by a fixed-time adaptive velocity observer and compensated in the controller. The key advantage of the proposed method lies in its well balance of control requirements between achieving precise landing and not violating the visibility constraints. Finally, comparative Gazebo simulations are conducted to verify the satisfactory control performance.