Nonlinear Robust Hybrid Control of Robotic Manipulators

Abstract

The success of a robot force and position (hybrid) control scheme relies extensively upon its robustness against uncertainties such as unknown external disturbance or modeling errors in the description of robot, sensor and environment. In this paper we propose a new nonlinear robust hybrid control scheme for robot motion control. The control input consists of a nonlinear and a linear part. The nonlinear input decouples a robot dynamics and gives a set of position and force equations in the hand or cartesian coordinates. The linear part applies the servomechanism theory to suppress position or force tracking error due to uncertainties. This nonlinear robust hybrid control scheme is applied to a two-joint SCARA type robot, and simulation results demonstrate excellent robustness properties and satisfactory hybrid control under severe modeling errors.