Robust control of an eye-in-hand robot based on affine models

In this paper, the problem of controlling the position of a robot camera with respect to an object is addressed by carefully choosing the system state representation and applying nonlinear control theory. An image-based state space representation of the robot camera-object interaction model is used, assuming affine shape transformations in the image space and local linear approximation of visible object's surface. These assumptions permit to greatly simplify the mathematical model of the visual interaction. The image-based visual system is stabilized by using Lyapunov direct method and the control law ensures asymptotic stability in case of ea:act model and state measurement. Robustness analysis is also carried out. Experimental results with a PUMA 560 robot in eye-in-hand configuration validate the theoretical framework both in terms of system convergence and control robustness.