Search

This work addresses the visual tracking problem of robot manipulators with non-negligible dynamics using a fixed camera, when the camera-robot system parameters are uncertain. An adaptive strategy is developed for visual servoing systems based on the image-based look-and-move structure to allow the tracking of a 2D reference trajectory, without using image velocity measurements. The adaptive visual servoing problem free of image velocity information is formulated as a relative degree two MIMO adaptive control problem. As a solution, we employ a recently proposed Lyapunov/passivity-based adaptive control scheme based on the SDU factorization method. From a cascade control strategy, the resulting online camera calibration scheme is combined with a direct adaptive motion controller for the robot manipulator, which takes into account its uncertain nonlinear dynamics. The overall stability of the adaptive visual servoing system can be proved thanks to the explicit Lyapunov-like functions of both adaptation schemes.

The design of a variable structure model reference robust control without a prior knowledge of high frequency gain sign is presented. Based on an appropriate monitoring function, a switching scheme for some control signals is proposed. It is shown that after a finite number of switching, the tracking error converges to zero at least exponentially for plants with relative degree one or converges exponentially to a small residual set for plants with higher relative degree, and the input disturbance can be completely rejected without affecting the tracking performance.