DocumentCode
567234
Title
Incremental learning of primitive skills from demonstration of a task
Author
Lee, Sang Hyoung ; Kim, Hyung Kyu ; Suh, Il Hong
Author_Institution
Dept. of Electron. & Comput. Eng., Hanyang Univ., Seoul, South Korea
fYear
2011
fDate
8-11 March 2011
Firstpage
185
Lastpage
186
Abstract
In this work, we propose methods for automatically generating primitive skills from the demonstration of a task. Additionally, we propose methods for improving existing primitive skills, and for automatically and incrementally adding new primitive skills. To validate our proposed methods, we present the experimental results of a human-like robot handling three gestures and a task of making coffee.
Keywords
humanoid robots; learning (artificial intelligence); human-like robot; incremental learning; primitive skills; Computers; Educational institutions; Hidden Markov models; Humans; Principal component analysis; Robots; Trajectory; KL-divergence; hidden Markov model; incremental learning; primitive skill; threshold model;
fLanguage
English
Publisher
ieee
Conference_Titel
Human-Robot Interaction (HRI), 2011 6th ACM/IEEE International Conference on
Conference_Location
Lausanne
ISSN
2167-2121
Print_ISBN
978-1-4673-4393-0
Electronic_ISBN
2167-2121
Type
conf
Filename
6281287
Link To Document