Adaptive Robust Motion Control Of Uncertain Manipulators Through Immersion And Invariance Adaptive Visual Servoing

The control objective of image-based eye-to-hand robot manipulator systems is to track a desired trajectory precisely in the presence of nonlinearities and uncertainties. This paper improves the dynamic performance of such a manipulator through a cascade control strategy of using a high-level controller with visual servoing for reference position command generation and a low-level controller for precise motion tracking in the presence of uncertainties. Specifically, an adaptive robust controller (ARC) is developed at the low-level to better handle the effect of both parametric uncertainties and uncertain nonlinearities for an improved robust motion tracking performance. The high-level controller uses the Immersion and Invariance (I&I) approach to provide reference position commands based on visual feedback. Simulation results are presented to illustrate the better robust tracking performance of the proposed control scheme in the presence of uncertainties.