In-hand object shape identification using invariant proprioceptive signatures

Alex Vásquez, Zhanat Kappassov, Véronique Perdereau

Research output: Chapter in Book/Report/Conference proceedingChapter

3 Citations (Scopus)

Abstract

Most modern approaches for tactile object recognition with robotic hands do not use proprioceptive data. In those that do, a limited number of objects with similar shapes is recognized. Furthermore, Self-Organizing Maps (SOM) based on raw values of joint angles/torques are frequently implemented which requires large sets of training data. In this paper, we present an approach based only on joint angles of a robotic hands to identify the shape of an object regardless its size and position within the hand. A representation of the joint angles is created to endow the robotic hand with proprioception. Support Vector Machine (SVM) is implemented for shape identification using patterns or signatures generated on this representation when the objects are grasped. To illustrate the scope of this method, tests are performed on five shapes present in common objects. Both SVM and SOM trained with signatures were at least 10% more accuracy than the ones trained with raw values of joint angles. Training sets are reduced at least 85% with respect to other works. An accuracy of 94% was obtained on large ranges of dimensions of the shapes and positions.
Original languageEnglish
Title of host publicationIEEE International Conference on Intelligent Robots and Systems
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages965-970
Number of pages6
ISBN (Print)9781509037629
DOIs
Publication statusPublished - Nov 28 2016

Publication series

NameIEEE International Conference on Intelligent Robots and Systems
Volume2016-November

Fingerprint

End effectors
Self organizing maps
Support vector machines
Object recognition
Torque

Keywords

  • Proprioception
  • Robotic hand
  • Shape identification

Cite this

Vásquez, A., Kappassov, Z., & Perdereau, V. (2016). In-hand object shape identification using invariant proprioceptive signatures. In IEEE International Conference on Intelligent Robots and Systems (pp. 965-970). (IEEE International Conference on Intelligent Robots and Systems; Vol. 2016-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IROS.2016.7759166

In-hand object shape identification using invariant proprioceptive signatures. / Vásquez, Alex; Kappassov, Zhanat; Perdereau, Véronique.

IEEE International Conference on Intelligent Robots and Systems. Institute of Electrical and Electronics Engineers Inc., 2016. p. 965-970 (IEEE International Conference on Intelligent Robots and Systems; Vol. 2016-November).

Research output: Chapter in Book/Report/Conference proceedingChapter

Vásquez, A, Kappassov, Z & Perdereau, V 2016, In-hand object shape identification using invariant proprioceptive signatures. in IEEE International Conference on Intelligent Robots and Systems. IEEE International Conference on Intelligent Robots and Systems, vol. 2016-November, Institute of Electrical and Electronics Engineers Inc., pp. 965-970. https://doi.org/10.1109/IROS.2016.7759166
Vásquez A, Kappassov Z, Perdereau V. In-hand object shape identification using invariant proprioceptive signatures. In IEEE International Conference on Intelligent Robots and Systems. Institute of Electrical and Electronics Engineers Inc. 2016. p. 965-970. (IEEE International Conference on Intelligent Robots and Systems). https://doi.org/10.1109/IROS.2016.7759166
Vásquez, Alex ; Kappassov, Zhanat ; Perdereau, Véronique. / In-hand object shape identification using invariant proprioceptive signatures. IEEE International Conference on Intelligent Robots and Systems. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 965-970 (IEEE International Conference on Intelligent Robots and Systems).
@inbook{95f5c6efb5e14b5994eeea5fb627a1c8,
title = "In-hand object shape identification using invariant proprioceptive signatures",
abstract = "Most modern approaches for tactile object recognition with robotic hands do not use proprioceptive data. In those that do, a limited number of objects with similar shapes is recognized. Furthermore, Self-Organizing Maps (SOM) based on raw values of joint angles/torques are frequently implemented which requires large sets of training data. In this paper, we present an approach based only on joint angles of a robotic hands to identify the shape of an object regardless its size and position within the hand. A representation of the joint angles is created to endow the robotic hand with proprioception. Support Vector Machine (SVM) is implemented for shape identification using patterns or signatures generated on this representation when the objects are grasped. To illustrate the scope of this method, tests are performed on five shapes present in common objects. Both SVM and SOM trained with signatures were at least 10{\%} more accuracy than the ones trained with raw values of joint angles. Training sets are reduced at least 85{\%} with respect to other works. An accuracy of 94{\%} was obtained on large ranges of dimensions of the shapes and positions.",
keywords = "Proprioception, Robotic hand, Shape identification",
author = "Alex V{\'a}squez and Zhanat Kappassov and V{\'e}ronique Perdereau",
year = "2016",
month = "11",
day = "28",
doi = "10.1109/IROS.2016.7759166",
language = "English",
isbn = "9781509037629",
series = "IEEE International Conference on Intelligent Robots and Systems",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "965--970",
booktitle = "IEEE International Conference on Intelligent Robots and Systems",
address = "United States",

}

TY - CHAP

T1 - In-hand object shape identification using invariant proprioceptive signatures

AU - Vásquez, Alex

AU - Kappassov, Zhanat

AU - Perdereau, Véronique

PY - 2016/11/28

Y1 - 2016/11/28

N2 - Most modern approaches for tactile object recognition with robotic hands do not use proprioceptive data. In those that do, a limited number of objects with similar shapes is recognized. Furthermore, Self-Organizing Maps (SOM) based on raw values of joint angles/torques are frequently implemented which requires large sets of training data. In this paper, we present an approach based only on joint angles of a robotic hands to identify the shape of an object regardless its size and position within the hand. A representation of the joint angles is created to endow the robotic hand with proprioception. Support Vector Machine (SVM) is implemented for shape identification using patterns or signatures generated on this representation when the objects are grasped. To illustrate the scope of this method, tests are performed on five shapes present in common objects. Both SVM and SOM trained with signatures were at least 10% more accuracy than the ones trained with raw values of joint angles. Training sets are reduced at least 85% with respect to other works. An accuracy of 94% was obtained on large ranges of dimensions of the shapes and positions.

AB - Most modern approaches for tactile object recognition with robotic hands do not use proprioceptive data. In those that do, a limited number of objects with similar shapes is recognized. Furthermore, Self-Organizing Maps (SOM) based on raw values of joint angles/torques are frequently implemented which requires large sets of training data. In this paper, we present an approach based only on joint angles of a robotic hands to identify the shape of an object regardless its size and position within the hand. A representation of the joint angles is created to endow the robotic hand with proprioception. Support Vector Machine (SVM) is implemented for shape identification using patterns or signatures generated on this representation when the objects are grasped. To illustrate the scope of this method, tests are performed on five shapes present in common objects. Both SVM and SOM trained with signatures were at least 10% more accuracy than the ones trained with raw values of joint angles. Training sets are reduced at least 85% with respect to other works. An accuracy of 94% was obtained on large ranges of dimensions of the shapes and positions.

KW - Proprioception

KW - Robotic hand

KW - Shape identification

U2 - 10.1109/IROS.2016.7759166

DO - 10.1109/IROS.2016.7759166

M3 - Chapter

SN - 9781509037629

T3 - IEEE International Conference on Intelligent Robots and Systems

SP - 965

EP - 970

BT - IEEE International Conference on Intelligent Robots and Systems

PB - Institute of Electrical and Electronics Engineers Inc.

ER -