Applied robust control for vibration suppression in parallel robots

Abstract

Automation within the range of handling and assembly applications requires qualified solutions due to the complex technological processes. The long-term goals are the reduction of cycle time and an increase of the process quality. They can be achieved by innovative concepts which are based on parallel kinematics, realizing higher speeds and accelerations with constant accuracy compared to conventional serial robot structures. High accelerations are equivalent to high forces in the starting and deceleration phase of the trajectory. The vibrations of the robot structure induced thereby and the following decaying procedure are unwanted and time-consuming in handling and assembly applications, in particular during accurate placement of components. In the context of German DFG Collaborative Research Center 562 'Robotic Systems for Handling and Assembly' a parallel robot with two degrees of freedom, made of CFRP components, was built-up at the Institute of Composite Structures and Adaptive Systems, DLR, Germany. The specialty of this robot is the integrated vibration suppression introduced by active rods, that are driven by piezoceramic stacks. These active rods are addressed by a robust controller, which generates the suitable control variable using measurements of the oscillations of the effector. The robot is not a time-invariant system and therefore its vibration characteristics changes depending on the position, the loading condition and the way the robot was assembled. These facts make high demands on the robust controller, which must output suitable signals to the actuators for the suppression of vibrations in each condition, without becoming unstable. In this article the parallel robot and its components are presented. A special focus is put on the design of the robust controller for vibration suppression. Furthermore, the strategies used for the employment of the control in the entire work space of the robot are shown. A further topic is the system identification of the plant, which must be accomplished fast and reliably with a variant system like this. Finally the effectiveness of the concepts and procedures presented here is shown with experimental data.