Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-16T02:46:11.785Z Has data issue: false hasContentIssue false

Speed adaptation for self-improvement of skills learned from user demonstrations

Published online by Cambridge University Press:  15 June 2015

Rok Vuga*
Affiliation:
Humanoid and Cognitive Robotics Lab, Department of Automatics, Biocybernetics and Robotics, Jožef Stean Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia. E-mails: bojan.nemec@ijs.si, ales.ude@ijs.si
Bojan Nemec
Affiliation:
Humanoid and Cognitive Robotics Lab, Department of Automatics, Biocybernetics and Robotics, Jožef Stean Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia. E-mails: bojan.nemec@ijs.si, ales.ude@ijs.si
Aleš Ude
Affiliation:
Humanoid and Cognitive Robotics Lab, Department of Automatics, Biocybernetics and Robotics, Jožef Stean Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia. E-mails: bojan.nemec@ijs.si, ales.ude@ijs.si
*
*Corresponding author. E-mail: rok.vuga@ijs.si
Rights & Permissions [Opens in a new window]

Summary

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the 'Save PDF' action button.

The paper addresses the problem of speed adaptation of movements subject to environmental constraints. Our approach relies on a novel formulation of velocity profiles as an extension of dynamic movement primitives (DMP). The framework allows for compact representation of non-uniformly accelerated motion as well as simple modulation of the movement parameters. In the paper, we evaluate two model free methods by which optimal parameters can be obtained: iterative learning control (ILC) and policy search based reinforcement learning (RL). The applicability of each method is discussed and evaluated on two distinct cases, which are hard to model using standard techniques. The first deals with hard contacts with the environment while the second process involves liquid dynamics. We find ILC to be very efficient in cases where task parameters can be easily described with an error function. On the other hand, RL has stronger convergence properties and can therefore provide a solution in the general case.

Information

Type
Articles
Copyright
Copyright © Cambridge University Press 2015