Motion Adaptation Based on Learning the Manifold of Task and Dynamic Movement Primitive Parameters

Dynamic movement primitives (DMP) are motion building blocks suitable for real-world tasks. We suggest a methodology for learning the manifold of task and DMP parameters, which facilitates runtime adaptation to changes in task requirements while ensuring predictable and robust performance. For effic...

Full description

Saved in:
Bibliographic Details
Published in:Robotica 2021-07, Vol.39 (7), p.1299-1315
Main Authors: Cohen, Yosef, Bar-Shira, Or, Berman, Sigal
Format: Article
Language:English
Subjects:
Adaptation
Artificial neural networks
Kernels
Machine learning
Manifolds (mathematics)
Movement
Parameters
Principal components analysis
Teaching methods
Throwing
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Dynamic movement primitives (DMP) are motion building blocks suitable for real-world tasks. We suggest a methodology for learning the manifold of task and DMP parameters, which facilitates runtime adaptation to changes in task requirements while ensuring predictable and robust performance. For efficient learning, the parameter space is analyzed using principal component analysis and locally linear embedding. Two manifold learning methods: kernel estimation and deep neural networks, are investigated for a ball throwing task in simulation and in a physical environment. Low runtime estimation errors are obtained for both learning methods, with an advantage to kernel estimation when data sets are small.
ISSN:0263-5747
1469-8668
DOI:10.1017/S0263574720001186