Motion generation based on reliable predictability using self-organized object features

Shun Nishide, Tetsuya Ogata, Jun Tani, Toru Takahashi, Kazunori Komatani, Hiroshi G. Okuno

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

Abstract

Predictability is an important factor for determining robot motions. This paper presents a model to generate robot motions based on reliable predictability evaluated through a dynamics learning model which self-organizes object features. The model is composed of a dynamics learning module, namely Recurrent Neural Network with Parametric Bias (RNNPB), and a hierarchical neural network as a feature extraction module. The model inputs raw object images and robot motions. Through bi-directional training of the two models, object features which describe the object motion are self-organized in the output of the hierarchical neural network, which is linked to the input of RNNPB. After training, the model searches for the robot motion with high reliable predictability of object motion. Experiments were performed with the robot's pushing motion with a variety of objects to generate sliding, falling over, bouncing, and rolling motions. For objects with single motion possibility, the robot tended to generate motions that induce the object motion. For objects with two motion possibilities, the robot evenly generated motions that induce the two object motions.

Original languageEnglish
Title of host publicationIEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings
Pages3453-3458
Number of pages6
DOIs
Publication statusPublished - 2010
Externally publishedYes
Event23rd IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Taipei
Duration: 2010 Oct 182010 Oct 22

Other

Other23rd IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010
CityTaipei
Period10/10/1810/10/22

Fingerprint

Robots
Recurrent neural networks
Neural networks
Feature extraction
Experiments

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Control and Systems Engineering

Cite this

Nishide, S., Ogata, T., Tani, J., Takahashi, T., Komatani, K., & Okuno, H. G. (2010). Motion generation based on reliable predictability using self-organized object features. In IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings (pp. 3453-3458). [5652609] https://doi.org/10.1109/IROS.2010.5652609

Motion generation based on reliable predictability using self-organized object features. / Nishide, Shun; Ogata, Tetsuya; Tani, Jun; Takahashi, Toru; Komatani, Kazunori; Okuno, Hiroshi G.

IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings. 2010. p. 3453-3458 5652609.

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

Nishide, S, Ogata, T, Tani, J, Takahashi, T, Komatani, K & Okuno, HG 2010, Motion generation based on reliable predictability using self-organized object features. in IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings., 5652609, pp. 3453-3458, 23rd IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010, Taipei, 10/10/18. https://doi.org/10.1109/IROS.2010.5652609
Nishide S, Ogata T, Tani J, Takahashi T, Komatani K, Okuno HG. Motion generation based on reliable predictability using self-organized object features. In IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings. 2010. p. 3453-3458. 5652609 https://doi.org/10.1109/IROS.2010.5652609
Nishide, Shun ; Ogata, Tetsuya ; Tani, Jun ; Takahashi, Toru ; Komatani, Kazunori ; Okuno, Hiroshi G. / Motion generation based on reliable predictability using self-organized object features. IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings. 2010. pp. 3453-3458
@inproceedings{1daa83ae39354439a290caaca52ad8f1,
title = "Motion generation based on reliable predictability using self-organized object features",
abstract = "Predictability is an important factor for determining robot motions. This paper presents a model to generate robot motions based on reliable predictability evaluated through a dynamics learning model which self-organizes object features. The model is composed of a dynamics learning module, namely Recurrent Neural Network with Parametric Bias (RNNPB), and a hierarchical neural network as a feature extraction module. The model inputs raw object images and robot motions. Through bi-directional training of the two models, object features which describe the object motion are self-organized in the output of the hierarchical neural network, which is linked to the input of RNNPB. After training, the model searches for the robot motion with high reliable predictability of object motion. Experiments were performed with the robot's pushing motion with a variety of objects to generate sliding, falling over, bouncing, and rolling motions. For objects with single motion possibility, the robot tended to generate motions that induce the object motion. For objects with two motion possibilities, the robot evenly generated motions that induce the two object motions.",
author = "Shun Nishide and Tetsuya Ogata and Jun Tani and Toru Takahashi and Kazunori Komatani and Okuno, {Hiroshi G.}",
year = "2010",
doi = "10.1109/IROS.2010.5652609",
language = "English",
isbn = "9781424466757",
pages = "3453--3458",
booktitle = "IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings",

}

TY - GEN

T1 - Motion generation based on reliable predictability using self-organized object features

AU - Nishide, Shun

AU - Ogata, Tetsuya

AU - Tani, Jun

AU - Takahashi, Toru

AU - Komatani, Kazunori

AU - Okuno, Hiroshi G.

PY - 2010

Y1 - 2010

N2 - Predictability is an important factor for determining robot motions. This paper presents a model to generate robot motions based on reliable predictability evaluated through a dynamics learning model which self-organizes object features. The model is composed of a dynamics learning module, namely Recurrent Neural Network with Parametric Bias (RNNPB), and a hierarchical neural network as a feature extraction module. The model inputs raw object images and robot motions. Through bi-directional training of the two models, object features which describe the object motion are self-organized in the output of the hierarchical neural network, which is linked to the input of RNNPB. After training, the model searches for the robot motion with high reliable predictability of object motion. Experiments were performed with the robot's pushing motion with a variety of objects to generate sliding, falling over, bouncing, and rolling motions. For objects with single motion possibility, the robot tended to generate motions that induce the object motion. For objects with two motion possibilities, the robot evenly generated motions that induce the two object motions.

AB - Predictability is an important factor for determining robot motions. This paper presents a model to generate robot motions based on reliable predictability evaluated through a dynamics learning model which self-organizes object features. The model is composed of a dynamics learning module, namely Recurrent Neural Network with Parametric Bias (RNNPB), and a hierarchical neural network as a feature extraction module. The model inputs raw object images and robot motions. Through bi-directional training of the two models, object features which describe the object motion are self-organized in the output of the hierarchical neural network, which is linked to the input of RNNPB. After training, the model searches for the robot motion with high reliable predictability of object motion. Experiments were performed with the robot's pushing motion with a variety of objects to generate sliding, falling over, bouncing, and rolling motions. For objects with single motion possibility, the robot tended to generate motions that induce the object motion. For objects with two motion possibilities, the robot evenly generated motions that induce the two object motions.

UR - http://www.scopus.com/inward/record.url?scp=78651497115&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651497115&partnerID=8YFLogxK

U2 - 10.1109/IROS.2010.5652609

DO - 10.1109/IROS.2010.5652609

M3 - Conference contribution

AN - SCOPUS:78651497115

SN - 9781424466757

SP - 3453

EP - 3458

BT - IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings

ER -