Servocompensation of disturbances in robotic systems

Abstract

Robot manipulators are susceptible to external disturbances that may alter their behaviour so significantly that certain specifications cannot be maintained. Additive signals that satisfy a linear differential equation with constant coefficients are the type of disturbance that may enter a manipulator and corrupt its behaviour. From linear servocompensator theory one knows that linear disturbances can be asymptotically rejected if the linear plant satisfies certain criteria. However, for robot manipulators, which are extremely non–linear, the former criteria cannot be applied. It is shown here that linear disturbances can be asymptotically rejected by employing a standard linear servocompensator. In linear system theory the proof that the error e → 0 as t → ∞ is shown via the Laplace transform. The Volterra series is utilized here to verify the asymptotic convergence of the error. Digital computer simulations are carried out on a two–link planar robot to demonstrate the effectiveness of the linear servoco...