Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

A New Approach to Estimate the Apparent Mass of Collaborative Robot Manipulators

: Herbster, S.; Behrens, R.; Elkmann, N.


Siciliano, B.:
Experimental Robotics : The 17th International Symposium, International Symposium on Experimental Robotics, ISER 2020, in Malta
Cham: Springer Nature, 2021 (Springer Proceedings in Advanced Robotics 19)
ISBN: 978-3-030-71150-4 (Print)
ISBN: 978-3-030-71151-1 (Online)
ISBN: 978-3-030-71152-8
ISBN: 978-3-030-71153-5
International Symposium on Experimental Robotics (ISER) <2020, Malta>
Fraunhofer IFF ()

Collaborative robots have finally found their way into the factories as companions to human workers for releasing them from repetitive and exhausting tasks. Internal torque sensors or tactile skins make these robots sensitive to external forces, which is a critical requirement for acting safely, as their direct force feedback allows them to react instantaneously on dangerous collisions with humans. Biomechanical limit values give robot users a reference to determine the injury potential of their robots. In some cases, those limits are available as energy limits. A valuable advantage of such limits is that any user can easily convert them into velocity constraints if the apparent mass of the robot is known. The apparent mass is a virtual quantity that projects the inertia properties of the robot to a directional point mass. In our paper, we show why the currently-available method to determine the apparent mass needs to be extended. Our findings reveal that the impact conditions determine whether the drive inertia must be taken into account or the apparent mass may only be calculated based on the dynamic mass parameters of the link. The verification relies on data from collision tests with two real collaborative robots. Our results fit well to the results we obtained from the simulation models, why our proposed modification seems to be necessary in order to achieve reliable velocity constraints for operating collaborative robots safely.