Theory and Experiments on the Compliance Control of Redundant Manipulators

2. UNSTRUCTURED MODELING OF THE SECONDARY MANIPULA TOR The primary manipulator is assumed to have a trajectory controller for positioning its endpoint (i.e. the base of the secondary manipulator). Several linear and nonlinear control methods can be used to develop this trajectory controller. However, the concern here is the control and modeling of the secondary manipulator. The secondary manipulator motors are assu:med to have velocity controllers since most servo motors have rate controllers. In general, a robotic system with a velocity controller has a velocity that is a dynamic function of its input vector!, e, and of the force, d, imposed at its endpoint. The velocity controller for the secondary manipulator is assumed to have zero 'telocity input in the neighborhood of a particular configuration of the active end-effector. Thus, the active end-effector is not used for maneuvering parts, but is used at a particular known configuration. 'J:'herefore, the dynamic behavior of the secondary manipulator is expressed in terms of linear transfer function matrices, not to Sil11plify the problem, but because of the nature of end-effector operation. Let G and So be two ABSTRACT This work presents a control methodology for compliant motion in redundant robot manipulators. This control approach takes advantage of the redundancy in the robot's degrees of freedom: while a maximum six degrees of freedom of the robot control the robot's endpoint position, the remaining degrees of freedom impose an appropriate force on the environment. To verify the applicability of this control method, an active end-effector is mounted on an industrial robot to generate redundancy in the degrees of freedom. A set of experiments are described to demonstrate the use of this control method in constrained maneuvers. The stability of the robot and the environment is analyzed.