A Kendama learning robot based on bi-directional theory

Hiroyuki Miyamoto, Stefan Schaal, Francesca Gandolfo, Hiroaki Gomi, Yasuharu Koike, Rieko Osu, Eri Nakano, Yasuhiro Wada, Mitsuo Kawato

Research output: Contribution to journalReview article

99 Citations (Scopus)

Abstract

A general theory of movement-pattern perception based on bi-directional theory for sensory-motor integration can be used for motion capture and learning by watching in robotics. We demonstrate our methods using the game of Kendama, executed by the SARCOS Dextrous Slave Arm, which has a very similar kinematic structure to the human arm. Three ingredients have to be integrated for the successful execution of this task. The ingredients are (1) to extract via-points from a human movement trajectory using a forward-inverse relaxation model, (2) to treat via-points as a control variable while reconstructing the desired trajectory from all the via-points, and (3) to modify the via-points for successful execution. In order to test the validity of the via-point representation, we utilized a numerical model of the SARCOS arm, and examined the behavior of the system under several conditions.

Original languageEnglish
Pages (from-to)1281-1302
Number of pages22
JournalNeural Networks
Volume9
Issue number8
DOIs
Publication statusPublished - 1996 Nov
Externally publishedYes

Fingerprint

Robot learning
Trajectories
Learning
Robotics
Biomechanical Phenomena
Numerical models
Kinematics

Keywords

  • dynamic optimization
  • task-level learning
  • teaching by showing

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Miyamoto, H., Schaal, S., Gandolfo, F., Gomi, H., Koike, Y., Osu, R., ... Kawato, M. (1996). A Kendama learning robot based on bi-directional theory. Neural Networks, 9(8), 1281-1302. https://doi.org/10.1016/S0893-6080(96)00043-3

A Kendama learning robot based on bi-directional theory. / Miyamoto, Hiroyuki; Schaal, Stefan; Gandolfo, Francesca; Gomi, Hiroaki; Koike, Yasuharu; Osu, Rieko; Nakano, Eri; Wada, Yasuhiro; Kawato, Mitsuo.

In: Neural Networks, Vol. 9, No. 8, 11.1996, p. 1281-1302.

Research output: Contribution to journalReview article

Miyamoto, H, Schaal, S, Gandolfo, F, Gomi, H, Koike, Y, Osu, R, Nakano, E, Wada, Y & Kawato, M 1996, 'A Kendama learning robot based on bi-directional theory', Neural Networks, vol. 9, no. 8, pp. 1281-1302. https://doi.org/10.1016/S0893-6080(96)00043-3
Miyamoto H, Schaal S, Gandolfo F, Gomi H, Koike Y, Osu R et al. A Kendama learning robot based on bi-directional theory. Neural Networks. 1996 Nov;9(8):1281-1302. https://doi.org/10.1016/S0893-6080(96)00043-3
Miyamoto, Hiroyuki ; Schaal, Stefan ; Gandolfo, Francesca ; Gomi, Hiroaki ; Koike, Yasuharu ; Osu, Rieko ; Nakano, Eri ; Wada, Yasuhiro ; Kawato, Mitsuo. / A Kendama learning robot based on bi-directional theory. In: Neural Networks. 1996 ; Vol. 9, No. 8. pp. 1281-1302.
@article{f236767fad214b0ea9abeed188af6069,
title = "A Kendama learning robot based on bi-directional theory",
abstract = "A general theory of movement-pattern perception based on bi-directional theory for sensory-motor integration can be used for motion capture and learning by watching in robotics. We demonstrate our methods using the game of Kendama, executed by the SARCOS Dextrous Slave Arm, which has a very similar kinematic structure to the human arm. Three ingredients have to be integrated for the successful execution of this task. The ingredients are (1) to extract via-points from a human movement trajectory using a forward-inverse relaxation model, (2) to treat via-points as a control variable while reconstructing the desired trajectory from all the via-points, and (3) to modify the via-points for successful execution. In order to test the validity of the via-point representation, we utilized a numerical model of the SARCOS arm, and examined the behavior of the system under several conditions.",
keywords = "dynamic optimization, task-level learning, teaching by showing",
author = "Hiroyuki Miyamoto and Stefan Schaal and Francesca Gandolfo and Hiroaki Gomi and Yasuharu Koike and Rieko Osu and Eri Nakano and Yasuhiro Wada and Mitsuo Kawato",
year = "1996",
month = "11",
doi = "10.1016/S0893-6080(96)00043-3",
language = "English",
volume = "9",
pages = "1281--1302",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier Limited",
number = "8",

}

TY - JOUR

T1 - A Kendama learning robot based on bi-directional theory

AU - Miyamoto, Hiroyuki

AU - Schaal, Stefan

AU - Gandolfo, Francesca

AU - Gomi, Hiroaki

AU - Koike, Yasuharu

AU - Osu, Rieko

AU - Nakano, Eri

AU - Wada, Yasuhiro

AU - Kawato, Mitsuo

PY - 1996/11

Y1 - 1996/11

N2 - A general theory of movement-pattern perception based on bi-directional theory for sensory-motor integration can be used for motion capture and learning by watching in robotics. We demonstrate our methods using the game of Kendama, executed by the SARCOS Dextrous Slave Arm, which has a very similar kinematic structure to the human arm. Three ingredients have to be integrated for the successful execution of this task. The ingredients are (1) to extract via-points from a human movement trajectory using a forward-inverse relaxation model, (2) to treat via-points as a control variable while reconstructing the desired trajectory from all the via-points, and (3) to modify the via-points for successful execution. In order to test the validity of the via-point representation, we utilized a numerical model of the SARCOS arm, and examined the behavior of the system under several conditions.

AB - A general theory of movement-pattern perception based on bi-directional theory for sensory-motor integration can be used for motion capture and learning by watching in robotics. We demonstrate our methods using the game of Kendama, executed by the SARCOS Dextrous Slave Arm, which has a very similar kinematic structure to the human arm. Three ingredients have to be integrated for the successful execution of this task. The ingredients are (1) to extract via-points from a human movement trajectory using a forward-inverse relaxation model, (2) to treat via-points as a control variable while reconstructing the desired trajectory from all the via-points, and (3) to modify the via-points for successful execution. In order to test the validity of the via-point representation, we utilized a numerical model of the SARCOS arm, and examined the behavior of the system under several conditions.

KW - dynamic optimization

KW - task-level learning

KW - teaching by showing

UR - http://www.scopus.com/inward/record.url?scp=0030297195&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030297195&partnerID=8YFLogxK

U2 - 10.1016/S0893-6080(96)00043-3

DO - 10.1016/S0893-6080(96)00043-3

M3 - Review article

AN - SCOPUS:0030297195

VL - 9

SP - 1281

EP - 1302

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

IS - 8

ER -