Predicting object dynamics from visual images through active sensing experiences

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

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

7 Citations (Scopus)

Abstract

Prediction of dynamic features is an important task for determining the manipulation strategies of an object. This paper presents a technique for predicting dynamics of objects relative to the robot's motion from visual images. During the learning phase, the authors use Recurrent Neural Network with Parametric Bias (RNNPB) to self-organize the dynamics of objects manipulated by the robot into the PB space. The acquired PB values, static images of objects, and robot motor values are input into a hierarchical neural network to link the static images to dynamic features (PB values). The neural network extracts prominent features that induce each object dynamics. For prediction of the motion sequence of an unknown object, the static image of the object and robot motor value are input into the neural network to calculate the PB values. By inputting the PB values into the closed loop RNNPB, the predicted movements of the object relative to the robot motion are calculated sequentially. Experiments were conducted with the humanoid robot Robovie-IIs pushing objects at different heights. Reducted grayscale images and shoulder pitch angles were input into the neural network to predict the dynamics of target objects. The results of the experiment proved that the technique is efficient for predicting the dynamics of the objects.

Original languageEnglish
Title of host publicationProceedings - IEEE International Conference on Robotics and Automation
Pages2501-2506
Number of pages6
DOIs
Publication statusPublished - 2007
Externally publishedYes
Event2007 IEEE International Conference on Robotics and Automation, ICRA'07 - Rome
Duration: 2007 Apr 102007 Apr 14

Other

Other2007 IEEE International Conference on Robotics and Automation, ICRA'07
CityRome
Period07/4/1007/4/14

Fingerprint

Robots
Neural networks
Recurrent neural networks
Experiments

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering

Cite this

Nishide, S., Ogata, T., Tani, J., Komatani, K., & Okuno, H. G. (2007). Predicting object dynamics from visual images through active sensing experiences. In Proceedings - IEEE International Conference on Robotics and Automation (pp. 2501-2506). [4209459] https://doi.org/10.1109/ROBOT.2007.363841

Predicting object dynamics from visual images through active sensing experiences. / Nishide, Shun; Ogata, Tetsuya; Tani, Jun; Komatani, Kazunori; Okuno, Hiroshi G.

Proceedings - IEEE International Conference on Robotics and Automation. 2007. p. 2501-2506 4209459.

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

Nishide, S, Ogata, T, Tani, J, Komatani, K & Okuno, HG 2007, Predicting object dynamics from visual images through active sensing experiences. in Proceedings - IEEE International Conference on Robotics and Automation., 4209459, pp. 2501-2506, 2007 IEEE International Conference on Robotics and Automation, ICRA'07, Rome, 07/4/10. https://doi.org/10.1109/ROBOT.2007.363841
Nishide S, Ogata T, Tani J, Komatani K, Okuno HG. Predicting object dynamics from visual images through active sensing experiences. In Proceedings - IEEE International Conference on Robotics and Automation. 2007. p. 2501-2506. 4209459 https://doi.org/10.1109/ROBOT.2007.363841
Nishide, Shun ; Ogata, Tetsuya ; Tani, Jun ; Komatani, Kazunori ; Okuno, Hiroshi G. / Predicting object dynamics from visual images through active sensing experiences. Proceedings - IEEE International Conference on Robotics and Automation. 2007. pp. 2501-2506
@inproceedings{b54331afe7c644df9e888a7a8e8c22e2,
title = "Predicting object dynamics from visual images through active sensing experiences",
abstract = "Prediction of dynamic features is an important task for determining the manipulation strategies of an object. This paper presents a technique for predicting dynamics of objects relative to the robot's motion from visual images. During the learning phase, the authors use Recurrent Neural Network with Parametric Bias (RNNPB) to self-organize the dynamics of objects manipulated by the robot into the PB space. The acquired PB values, static images of objects, and robot motor values are input into a hierarchical neural network to link the static images to dynamic features (PB values). The neural network extracts prominent features that induce each object dynamics. For prediction of the motion sequence of an unknown object, the static image of the object and robot motor value are input into the neural network to calculate the PB values. By inputting the PB values into the closed loop RNNPB, the predicted movements of the object relative to the robot motion are calculated sequentially. Experiments were conducted with the humanoid robot Robovie-IIs pushing objects at different heights. Reducted grayscale images and shoulder pitch angles were input into the neural network to predict the dynamics of target objects. The results of the experiment proved that the technique is efficient for predicting the dynamics of the objects.",
author = "Shun Nishide and Tetsuya Ogata and Jun Tani and Kazunori Komatani and Okuno, {Hiroshi G.}",
year = "2007",
doi = "10.1109/ROBOT.2007.363841",
language = "English",
isbn = "1424406021",
pages = "2501--2506",
booktitle = "Proceedings - IEEE International Conference on Robotics and Automation",

}

TY - GEN

T1 - Predicting object dynamics from visual images through active sensing experiences

AU - Nishide, Shun

AU - Ogata, Tetsuya

AU - Tani, Jun

AU - Komatani, Kazunori

AU - Okuno, Hiroshi G.

PY - 2007

Y1 - 2007

N2 - Prediction of dynamic features is an important task for determining the manipulation strategies of an object. This paper presents a technique for predicting dynamics of objects relative to the robot's motion from visual images. During the learning phase, the authors use Recurrent Neural Network with Parametric Bias (RNNPB) to self-organize the dynamics of objects manipulated by the robot into the PB space. The acquired PB values, static images of objects, and robot motor values are input into a hierarchical neural network to link the static images to dynamic features (PB values). The neural network extracts prominent features that induce each object dynamics. For prediction of the motion sequence of an unknown object, the static image of the object and robot motor value are input into the neural network to calculate the PB values. By inputting the PB values into the closed loop RNNPB, the predicted movements of the object relative to the robot motion are calculated sequentially. Experiments were conducted with the humanoid robot Robovie-IIs pushing objects at different heights. Reducted grayscale images and shoulder pitch angles were input into the neural network to predict the dynamics of target objects. The results of the experiment proved that the technique is efficient for predicting the dynamics of the objects.

AB - Prediction of dynamic features is an important task for determining the manipulation strategies of an object. This paper presents a technique for predicting dynamics of objects relative to the robot's motion from visual images. During the learning phase, the authors use Recurrent Neural Network with Parametric Bias (RNNPB) to self-organize the dynamics of objects manipulated by the robot into the PB space. The acquired PB values, static images of objects, and robot motor values are input into a hierarchical neural network to link the static images to dynamic features (PB values). The neural network extracts prominent features that induce each object dynamics. For prediction of the motion sequence of an unknown object, the static image of the object and robot motor value are input into the neural network to calculate the PB values. By inputting the PB values into the closed loop RNNPB, the predicted movements of the object relative to the robot motion are calculated sequentially. Experiments were conducted with the humanoid robot Robovie-IIs pushing objects at different heights. Reducted grayscale images and shoulder pitch angles were input into the neural network to predict the dynamics of target objects. The results of the experiment proved that the technique is efficient for predicting the dynamics of the objects.

UR - http://www.scopus.com/inward/record.url?scp=36348929898&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36348929898&partnerID=8YFLogxK

U2 - 10.1109/ROBOT.2007.363841

DO - 10.1109/ROBOT.2007.363841

M3 - Conference contribution

SN - 1424406021

SN - 9781424406029

SP - 2501

EP - 2506

BT - Proceedings - IEEE International Conference on Robotics and Automation

ER -