Visual servoing for underactuated VTOL UAVs:a linear, homography-based framework

摘要

The paper concerns the control of vertical take-off and landing (VTOL) underactuated aerial vehicles (UAVs) in hover flight, on the basis of measurements provided by an onboard video camera. The objective is to stabilize the vehicle to the equilibrium pose associated with an image of a planar target, using a minimal sensor suite and poor knowledge of the environment. By using the homography matrix computed from the camera measurements of the target, stabilizing feedback laws are derived on the basis of the visual data and gyrometer measurements only. Explicit stability conditions on the control parameters are provided, showing that a proper tuning of the control parameters ensures a large robustness margin with only planar target and visibility assumptions, although the target size and orientation, the UAV position, linear velocity and orientation are unknown. Additional issues, such as the use of accelerometers to improve the UAV's positioning in the case of unmodeled dynamics (such as wind), are also considered.