Investigation of torque controlled robots with flexible links using a flexible multibody simulation

Tom Prautzsch, Tobias F.C. Berninger, Daniel J. Rixen

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

1 Zitat (Scopus)

Abstract

The past two decades have brought forward a new class of robotic arms designed for close human-robot cooperation. Their lighter design has greater structural flexibilities compared to classical industrial robots, which leads to undesired vibrations at the tool center point. Joint torque feedback controllers have been designed to counteract effects of robot joint flexibility. The goal of this project is to study the impact of link flexibilities on the accuracy of robots with such torque controllers. The study is performed using a flexible multibody simulation that is experimentally validated with the torque controlled robot Franka Panda. It is shown that, for both a single link and a 6-axis robot model, the torque controllers can dampen structural vibrations within the joint space if the torque feedback loop has high enough bandwidth. However, modes orthogonal to the joint space cannot be dampened.

OriginalspracheEnglisch
Titel2021 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2021
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers Inc.
Seiten638-644
Seitenumfang7
ISBN (elektronisch)9781665441391
DOIs
PublikationsstatusVeröffentlicht - 12 Juli 2021
Veranstaltung2021 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2021 - Delft, Niederlande
Dauer: 12 Juli 202116 Juli 2021

Publikationsreihe

NameIEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM
Band2021-July

Konferenz

Konferenz2021 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2021
Land/GebietNiederlande
OrtDelft
Zeitraum12/07/2116/07/21

Fingerprint

Untersuchen Sie die Forschungsthemen von „Investigation of torque controlled robots with flexible links using a flexible multibody simulation“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren