Prediction and imitation of other's motions by reusing own forward-inverse model in robots

Tetsuya Ogata, Ryunosuke Yokoya, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

This paper proposes a model that enables a robot to predict and imitate the motions of another by reusing its body forward-inverse model. Our model includes three approaches: (i) projection of a self-forward model for predicting phenomena in the external environment (other individuals), (ii) "triadic relation" that is mediation by a physical object between self and others, (iii) introduction of infant imitation by a parent. The Recurrent Neural Network with Parametric Bias (RNNPB) model is used as the robot's self forward-inverse model. A group of hierarchical neural networks are attached to the RNNPB model as "conversion modules". Experiments demonstrated that a robot with our model could imitate a human's motions by translating the viewpoint. It could also discriminate known/unknown motions appropriately, and associate whole motion dynamics from only one motion snap image.

Original languageEnglish
Title of host publicationProceedings - IEEE International Conference on Robotics and Automation
Pages4144-4149
Number of pages6
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event2009 IEEE International Conference on Robotics and Automation, ICRA '09 - Kobe
Duration: 2009 May 122009 May 17

Other

Other2009 IEEE International Conference on Robotics and Automation, ICRA '09
CityKobe
Period09/5/1209/5/17

Fingerprint

Robots
Recurrent neural networks
Neural networks
Experiments

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Ogata, T., Yokoya, R., Tani, J., Komatani, K., & Okuno, H. G. (2009). Prediction and imitation of other's motions by reusing own forward-inverse model in robots. In Proceedings - IEEE International Conference on Robotics and Automation (pp. 4144-4149). [5152363] https://doi.org/10.1109/ROBOT.2009.5152363

Prediction and imitation of other's motions by reusing own forward-inverse model in robots. / Ogata, Tetsuya; Yokoya, Ryunosuke; Tani, Jun; Komatani, Kazunori; Okuno, Hiroshi G.

Proceedings - IEEE International Conference on Robotics and Automation. 2009. p. 4144-4149 5152363.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ogata, T, Yokoya, R, Tani, J, Komatani, K & Okuno, HG 2009, Prediction and imitation of other's motions by reusing own forward-inverse model in robots. in Proceedings - IEEE International Conference on Robotics and Automation., 5152363, pp. 4144-4149, 2009 IEEE International Conference on Robotics and Automation, ICRA '09, Kobe, 09/5/12. https://doi.org/10.1109/ROBOT.2009.5152363
Ogata T, Yokoya R, Tani J, Komatani K, Okuno HG. Prediction and imitation of other's motions by reusing own forward-inverse model in robots. In Proceedings - IEEE International Conference on Robotics and Automation. 2009. p. 4144-4149. 5152363 https://doi.org/10.1109/ROBOT.2009.5152363
Ogata, Tetsuya ; Yokoya, Ryunosuke ; Tani, Jun ; Komatani, Kazunori ; Okuno, Hiroshi G. / Prediction and imitation of other's motions by reusing own forward-inverse model in robots. Proceedings - IEEE International Conference on Robotics and Automation. 2009. pp. 4144-4149
@inproceedings{55b86d5900d047e68be42b5695b17dac,
title = "Prediction and imitation of other's motions by reusing own forward-inverse model in robots",
abstract = "This paper proposes a model that enables a robot to predict and imitate the motions of another by reusing its body forward-inverse model. Our model includes three approaches: (i) projection of a self-forward model for predicting phenomena in the external environment (other individuals), (ii) {"}triadic relation{"} that is mediation by a physical object between self and others, (iii) introduction of infant imitation by a parent. The Recurrent Neural Network with Parametric Bias (RNNPB) model is used as the robot's self forward-inverse model. A group of hierarchical neural networks are attached to the RNNPB model as {"}conversion modules{"}. Experiments demonstrated that a robot with our model could imitate a human's motions by translating the viewpoint. It could also discriminate known/unknown motions appropriately, and associate whole motion dynamics from only one motion snap image.",
author = "Tetsuya Ogata and Ryunosuke Yokoya and Jun Tani and Kazunori Komatani and Okuno, {Hiroshi G.}",
year = "2009",
doi = "10.1109/ROBOT.2009.5152363",
language = "English",
isbn = "9781424427895",
pages = "4144--4149",
booktitle = "Proceedings - IEEE International Conference on Robotics and Automation",

}

TY - GEN

T1 - Prediction and imitation of other's motions by reusing own forward-inverse model in robots

AU - Ogata, Tetsuya

AU - Yokoya, Ryunosuke

AU - Tani, Jun

AU - Komatani, Kazunori

AU - Okuno, Hiroshi G.

PY - 2009

Y1 - 2009

N2 - This paper proposes a model that enables a robot to predict and imitate the motions of another by reusing its body forward-inverse model. Our model includes three approaches: (i) projection of a self-forward model for predicting phenomena in the external environment (other individuals), (ii) "triadic relation" that is mediation by a physical object between self and others, (iii) introduction of infant imitation by a parent. The Recurrent Neural Network with Parametric Bias (RNNPB) model is used as the robot's self forward-inverse model. A group of hierarchical neural networks are attached to the RNNPB model as "conversion modules". Experiments demonstrated that a robot with our model could imitate a human's motions by translating the viewpoint. It could also discriminate known/unknown motions appropriately, and associate whole motion dynamics from only one motion snap image.

AB - This paper proposes a model that enables a robot to predict and imitate the motions of another by reusing its body forward-inverse model. Our model includes three approaches: (i) projection of a self-forward model for predicting phenomena in the external environment (other individuals), (ii) "triadic relation" that is mediation by a physical object between self and others, (iii) introduction of infant imitation by a parent. The Recurrent Neural Network with Parametric Bias (RNNPB) model is used as the robot's self forward-inverse model. A group of hierarchical neural networks are attached to the RNNPB model as "conversion modules". Experiments demonstrated that a robot with our model could imitate a human's motions by translating the viewpoint. It could also discriminate known/unknown motions appropriately, and associate whole motion dynamics from only one motion snap image.

UR - http://www.scopus.com/inward/record.url?scp=70350366588&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350366588&partnerID=8YFLogxK

U2 - 10.1109/ROBOT.2009.5152363

DO - 10.1109/ROBOT.2009.5152363

M3 - Conference contribution

AN - SCOPUS:70350366588

SN - 9781424427895

SP - 4144

EP - 4149

BT - Proceedings - IEEE International Conference on Robotics and Automation

ER -