Integrated Visual Servoing Solution to Quadrotor Stabilization and Attitude Estimation Using a Pan and Tilt Camera

This paper presents an integrated visual servoing solution to the problem of quadrotor attitude estimation and stabilization based solely on image feedback from a pan and tilt camera and biased rate gyros. The solution comprises a control law for the quadrotor position and attitude, a control law for the pan and tilt camera platform, and an estimator for the vehicle attitude, whose interconnection stabilizes the quadrotor in hover above certain terrain landmarks. Lateral-longitudinal stabilization is achieved with a nested saturation control law by feedback of the image measurements, estimated attitude, and corrected angular velocity measurements and the vehicle is stabilized vertically by means of image feedback. Resorting to the input-to-state stability properties of the controllers, the quadrotor's position and attitude are shown to converge to the desired equilibrium point and, in the end, the interconnected system's closed loop is proven robust to observer estimation errors. Additionally, the pan and tilt camera is actively actuated to keep the relevant features visible in the image field of view for most operating conditions. The robustness and performance of the proposed control and estimation architecture is illustrated through both simulation and experimental results.