Learning a real time grasping strategy

Bidan Huang, Sahar El-Khoury, Miao Li, Joanna J. Bryson, Aude Billard

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

  • 21 Citations

Abstract

Real time planning strategy is crucial for robots working in dynamic environments. In particular, robot grasping tasks require quick reactions in many applications such as human-robot interaction. In this paper, we propose an approach for grasp learning that enables robots to plan new grasps rapidly according to the object’s position and orientation. This is achieved by taking a three-step approach. In the first step, we compute a variety of stable grasps for a given object. In the second step, we propose a strategy that learns a probability distribution of grasps based on the computed grasps. In the third step, we use the model to quickly generate grasps. We have tested the statistical method on the 9 degrees of freedom hand of the iCub humanoid robot and the 4 degrees of freedom Barrett hand. The average computation time for generating one grasp is less than 10 milliseconds. The experiments were run in Matlab on a machine with 2.8GHz processor.
LanguageEnglish
Title of host publicationIEEE International Conference on Robotics and Automation (ICRA) 2013
Pages593-600
DOIs
StatusPublished - 1 May 2013

Fingerprint

Robots
End effectors
Human robot interaction
Probability distributions
Statistical methods
Planning
Experiments

Cite this

Huang, B., El-Khoury, S., Li, M., Bryson, J. J., & Billard, A. (2013). Learning a real time grasping strategy. In IEEE International Conference on Robotics and Automation (ICRA) 2013 (pp. 593-600). DOI: 10.1109/ICRA.2013.6630634

Learning a real time grasping strategy. / Huang, Bidan; El-Khoury, Sahar; Li, Miao; Bryson, Joanna J.; Billard, Aude.

IEEE International Conference on Robotics and Automation (ICRA) 2013. 2013. p. 593-600.

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

Huang, B, El-Khoury, S, Li, M, Bryson, JJ & Billard, A 2013, Learning a real time grasping strategy. in IEEE International Conference on Robotics and Automation (ICRA) 2013. pp. 593-600. DOI: 10.1109/ICRA.2013.6630634
Huang B, El-Khoury S, Li M, Bryson JJ, Billard A. Learning a real time grasping strategy. In IEEE International Conference on Robotics and Automation (ICRA) 2013. 2013. p. 593-600. Available from, DOI: 10.1109/ICRA.2013.6630634
Huang, Bidan ; El-Khoury, Sahar ; Li, Miao ; Bryson, Joanna J. ; Billard, Aude. / Learning a real time grasping strategy. IEEE International Conference on Robotics and Automation (ICRA) 2013. 2013. pp. 593-600
@inproceedings{6fd2a85d61ed4874a33a925ba3a88f55,
title = "Learning a real time grasping strategy",
abstract = "Real time planning strategy is crucial for robots working in dynamic environments. In particular, robot grasping tasks require quick reactions in many applications such as human-robot interaction. In this paper, we propose an approach for grasp learning that enables robots to plan new grasps rapidly according to the object’s position and orientation. This is achieved by taking a three-step approach. In the first step, we compute a variety of stable grasps for a given object. In the second step, we propose a strategy that learns a probability distribution of grasps based on the computed grasps. In the third step, we use the model to quickly generate grasps. We have tested the statistical method on the 9 degrees of freedom hand of the iCub humanoid robot and the 4 degrees of freedom Barrett hand. The average computation time for generating one grasp is less than 10 milliseconds. The experiments were run in Matlab on a machine with 2.8GHz processor.",
author = "Bidan Huang and Sahar El-Khoury and Miao Li and Bryson, {Joanna J.} and Aude Billard",
year = "2013",
month = "5",
day = "1",
doi = "10.1109/ICRA.2013.6630634",
language = "English",
isbn = "978-1-4673-5641-1",
pages = "593--600",
booktitle = "IEEE International Conference on Robotics and Automation (ICRA) 2013",

}

TY - GEN

T1 - Learning a real time grasping strategy

AU - Huang,Bidan

AU - El-Khoury,Sahar

AU - Li,Miao

AU - Bryson,Joanna J.

AU - Billard,Aude

PY - 2013/5/1

Y1 - 2013/5/1

N2 - Real time planning strategy is crucial for robots working in dynamic environments. In particular, robot grasping tasks require quick reactions in many applications such as human-robot interaction. In this paper, we propose an approach for grasp learning that enables robots to plan new grasps rapidly according to the object’s position and orientation. This is achieved by taking a three-step approach. In the first step, we compute a variety of stable grasps for a given object. In the second step, we propose a strategy that learns a probability distribution of grasps based on the computed grasps. In the third step, we use the model to quickly generate grasps. We have tested the statistical method on the 9 degrees of freedom hand of the iCub humanoid robot and the 4 degrees of freedom Barrett hand. The average computation time for generating one grasp is less than 10 milliseconds. The experiments were run in Matlab on a machine with 2.8GHz processor.

AB - Real time planning strategy is crucial for robots working in dynamic environments. In particular, robot grasping tasks require quick reactions in many applications such as human-robot interaction. In this paper, we propose an approach for grasp learning that enables robots to plan new grasps rapidly according to the object’s position and orientation. This is achieved by taking a three-step approach. In the first step, we compute a variety of stable grasps for a given object. In the second step, we propose a strategy that learns a probability distribution of grasps based on the computed grasps. In the third step, we use the model to quickly generate grasps. We have tested the statistical method on the 9 degrees of freedom hand of the iCub humanoid robot and the 4 degrees of freedom Barrett hand. The average computation time for generating one grasp is less than 10 milliseconds. The experiments were run in Matlab on a machine with 2.8GHz processor.

UR - http://www.cs.bath.ac.uk/~jjb/ftp/HuangICRA13.pdf

UR - http://dx.doi.org/10.1109/ICRA.2013.6630634

U2 - 10.1109/ICRA.2013.6630634

DO - 10.1109/ICRA.2013.6630634

M3 - Conference contribution

SN - 978-1-4673-5641-1

SP - 593

EP - 600

BT - IEEE International Conference on Robotics and Automation (ICRA) 2013

ER -