Lyapunov-Based Visual Servo Tracking Control Via A Quatemion Formulation

The singularities in visual servo control associated with the image Jacobian and the error system containing trigonometric terms make the controller fail in some particular configurations. Many applications may be restricted by these singularities. In this paper, a quaternion parameterization is used to formulate the visual servo tracking control problem to eliminate the singularities. As in the previous work by Chen et al. (2005), the tracking problem is encoded by a sequence of images (i.e., a video). Adaptive tracking controllers are developed for both the camera-in-hand and camera-to-hand problem. A quaternion-based Lyapunov function is developed to facilitate the control design and the stability analysis