Abstract | ||
---|---|---|
When learning forward and inverse kinematics maps of manipulators, usually little attention is paid to data-efficiency, i.e., the accuracy gained per action-outcome sample.This paper examines properties of popular (online) learning techniques and demonstrates that - regardless of the employed exploration strategy - the structure of kinematics mappings does not allow for a practically viable trade-off between the number of samples and the resulting approximation error for manipulators with more than a few DoFs - unless tailored parametric models are employed.We discuss suitable choices for these parametric models for both rigid and elastic discretely-actuated robots and compare their data-efficiency to that of popular exploratory learning approaches relying on non-parametric models. Our theoretical considerations are confirmed by various experimental results for inverse kinematics mappings of rigid and omnielastic manipulators. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/IROS.2018.8593833 | 2018 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) |
Field | DocType | ISSN |
Parametric model,Kinematics,Inverse kinematics,Computer science,Exploratory learning,Algorithm,Control engineering,Solid modeling,Robot,Approximation error | Conference | 2153-0858 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Daniel Kubus | 1 | 48 | 9.02 |
Rania Rayyes | 2 | 2 | 1.75 |
Jochen J. Steil | 3 | 910 | 87.50 |