Visual Servoing of Flexible Manipulators With Unknown Camera Intrinsic Parameters and Vibration States

This article studies the visual servoing and vibration suppression of flexible manipulators using a fixed camera with uncalibrated intrinsic parameters. The dynamic equations of the flexible manipulator are decomposed into the slow and fast subsystems based on the singular perturbation theory. In the slow subsystem, the adaptive visual servoing controller is designed to regulate the positions of feature points on the image plane, where the unknown camera intrinsic parameters are estimated online. With respect to the fast subsystem, the unknown camera intrinsic parameters are regarded as the additional states of the fast subsystem, and the nonlinear state observer based on the Lie derivatives is developed to estimate the vibration states and camera parameters simultaneously from the visual feedback. The vision-based fast subsystem controller is proposed by using the estimated states to suppress the elastic vibration of the flexible manipulator. Both the stabilities of the slow and fast subsystems are proved based on the Lyapunov theory. Experimental results are provided to verify the effectiveness of the proposed control scheme.