Experimental analysis of a hybrid drive train implemented in a novel industrial robot approach
Conference: ISR Europe 2022 - 54th International Symposium on Robotics
06/20/2022 - 06/21/2022 at Munich
Proceedings: ISR Europe 2022
Pages: 8Language: englishTyp: PDF
Authors:
Hamann, Tobias; Hansen, Stephan; Boehlmann, Christian; Moeller, Christian (Fraunhofer IFAM, Stade, Germany)
Bitterolf, David (Siemens AG, Erlangen, Germany)
Seifried, Robert; Hintze, Wolfgang (MUM and IPMT, TUHH, Hamburg, Germany)
Abstract:
It is currently admitted that the path accuracy of industrial robots is not better than one millimeter. One source of inaccuracy originates in the low stiffness of the transmission link in the drivetrain, usually including a gearbox. The resulting low resonance frequency limits the bandwidth of the position controller loop. This leads to a higher tracking error and a lack of path accuracy in trajectories with high curvature. This paper presents an approach to address this issue by combining a gearbox motor with an additional torque motor on the load side. Direct control with the torque motor enables faster reference tracking and the combination of drives allows higher dam ping by the controllers. However, this special drivetrain is rarely scientifically or systematically analyzed. Furthermore, proposals for controller adjustments and commissioning with industrial hardware are non-existent. In this paper, the basic characteristics of common servo motor gearbox drives are described and the effects of the add itional torque motor are analyzed. This includes the commissioning in industrial hardware structure in terms of a cascaded P/PI -controller architecture and the description of adjustable tuning parameters as well as their influence on the path accuracy. The focus hereby is to improve the tracking behavior of the robot axis by increasing the damping, which enables - as a consequence - the increaseof the bandwidth. Further details about the experimental setup and investigation with the robotic manipulator aregiven. Finally, experimental results on an industrial robot prototype equipped with hybrid drives are presented.