Active learning for vision-based robot grasping

The IE-ID3 algorithm is described which extends the Interval Estimation (IF) active learning approach from discrete to real-valued learning domains by combining IE with a classification tree learning algorithm (ID-3). Presented is a robot system which rapidly learns to select the grasp approach dire...

Full description

Saved in:
Bibliographic Details
Published in:Machine learning 1996, Vol.23 (2-3), p.251-278
Main Authors: Salganicoff, Marcos, Ungar, Lyle H., Bajcsy, Ruzena
Format: Article
Language:English
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:The IE-ID3 algorithm is described which extends the Interval Estimation (IF) active learning approach from discrete to real-valued learning domains by combining IE with a classification tree learning algorithm (ID-3). Presented is a robot system which rapidly learns to select the grasp approach directions using IE-ID3 given simplified superquadric shape approximations of objects. Initial results on a small set of objects show that a robot with a laser scanner system can rapidly learn to pick up new objects, and simulation studies show the superiority of the active learning approach for a simulated grasping task using larger sets of objects. Extensions of the approach and future areas of research incorporating more sophisticated perceptual and action representation are discussed.
ISSN:0885-6125
1573-0565
DOI:10.1007/BF00117446