TU Darmstadt / ULB / TUprints

Demonstration based trajectory optimization for generalizable robot motions

Koert, Dorothea ; Maeda, Guilherme ; Lioutikov, Rudolf ; Neumann, Gerhard ; Peters, Jan (2022):
Demonstration based trajectory optimization for generalizable robot motions. (Postprint)
In: 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids), pp. 515-522,
Darmstadt, IEEE, International Conference on Humanoid Robots (Humanoids), Cancun, Mexico, 15.-17.11.2016, e-ISSN 2164-0580, ISBN 978-1-509-04719-2,
DOI: 10.26083/tuprints-00020544,
[Conference or Workshop Item]

[img] Text
Debato.pdf
Copyright Information: In Copyright.

Download (4MB)
Item Type: Conference or Workshop Item
Origin: Secondary publication service
Status: Postprint
Title: Demonstration based trajectory optimization for generalizable robot motions
Language: English
Abstract:

Learning motions from human demonstrations provides an intuitive way for non-expert users to teach tasks to robots. In particular, intelligent robotic co-workers should not only mimic human demonstrations but should also be able to adapt them to varying application scenarios. As such, robots must have the ability to generalize motions to different workspaces, e.g. to avoid obstacles not present during original demonstrations. Towards this goal our work proposes a unified method to (1) generalize robot motions to different workspaces, using a novel formulation of trajectory optimization that explicitly incorporates human demonstrations, and (2) to locally adapt and reuse the optimized solution in the form of a distribution of trajectories. This optimized distribution can be used, online, to quickly satisfy via-points and goals of a specific task. We validate the method using a 7 degrees of freedom (DoF) lightweight arm that grasps and places a ball into different boxes while avoiding obstacles that were not present during the original human demonstrations.

Book Title: 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids)
Place of Publication: Darmstadt
Publisher: IEEE
Collation: 8 Seiten
Classification DDC: 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik
600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften
Divisions: 20 Department of Computer Science > Intelligent Autonomous Systems
TU-Projects: EC/H2020|640554|SKILLS4ROBOTS
Event Title: International Conference on Humanoid Robots (Humanoids)
Event Location: Cancun, Mexico
Event Dates: 15.-17.11.2016
Date Deposited: 18 Nov 2022 13:57
Last Modified: 18 Nov 2022 13:57
DOI: 10.26083/tuprints-00020544
Corresponding Links:
URN: urn:nbn:de:tuda-tuprints-205443
URI: https://tuprints.ulb.tu-darmstadt.de/id/eprint/20544
PPN:
Export:
Actions (login required)
View Item View Item