Perception-Aware Model Predictive Visual Servoing for Fixed-Wing Unmanned Aerial Vehicles in Target Tracking

In this paper, a perception-aware model predictive control (MPC) method is proposed to improve the performance of image-based visual servoing (IBVS) controller. In this way, the dynamic constraints of UAVs and perception constraint of the camera can be taken into consideration, which cannot be solved by the classic IBVS. Besides, by integrating the control of UAV and pan-tilt camera, the target can be tracked and concurrently kept near the image center. Furthermore, the soft constraint method is used to reduce the computational complexity and avoid infeasible solution when calculating the MPC optimization problem. Finally, the effectiveness of the proposed method is verified based on software-in-the-loop simulations.