Koutras, Leonidas
Doulgeri, Zoe
2019-11-01
<p>Dynamic movement primitives (DMP) are an efficient way for learning<br>
and reproducing complex robot behaviors. A singularity free DMP formulation<br>
for orientation in the Cartesian space is proposed by Ude et al. in 2014 and has been<br>
largely adopted by the research community. In this work, we demonstrate the<br>
undesired oscillatory behavior that may arise when controlling the robot’s orien-<br>
tation with this formulation, producing a motion pattern highly deviant from the<br>
desired and highlight its source. A correct formulation is then proposed that alle-<br>
viates such problems while guaranteeing generation of orientation parameters that<br>
lie in SO(3). We further show that all aspects and advantages of DMP including<br>
ease of learning, temporal and spatial scaling and the ability to include coupling<br>
terms are maintained in the proposed formulation. Simulations and experiments<br>
with robot control in SO(3) are performed to demonstrate the performance of the<br>
proposed formulation and compare it with the previously adopted one.</p>
https://doi.org/10.5281/zenodo.3537791
oai:zenodo.org:3537791
eng
Zenodo
https://zenodo.org/communities/collaborate_project
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.3537790
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
CoRL, Conference on Robot Learning, Osaka, Japan, 30 October - 1 November 2019
Dynamic Movement Primitives
Control in Orientation Trajectories
A correct formulation for the Orientation Dynamic Movement Primitives for robot control in the Cartesian space
info:eu-repo/semantics/conferencePaper