Probabilistic consolidation of grasp experience

Yasemin Bekiroglu; Andreas Damianou; Renaud Detry; Johannes A Stork; Danica Kragic; Carl Henrik Ek

doi:10.1109/ICRA.2016.7487133

Probabilistic consolidation of grasp experience

Yasemin Bekiroglu, Andreas Damianou, Renaud Detry, Johannes A Stork, Danica Kragic, Carl Henrik Ek

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

6 Citations (Scopus)

342 Downloads (Pure)

Abstract

We present a probabilistic model for joint representation of several sensory modalities and action parameters in a robotic grasping scenario. Our non-linear probabilistic latent variable model encodes relationships between grasp-related parameters, learns the importance of features, and expresses confidence in estimates. The model learns associations between stable and unstable grasps that it experiences during an exploration phase. We demonstrate the applicability of the model for estimating grasp stability, correcting grasps, identifying objects based on tactile imprints and predicting tactile imprints from object-relative gripper poses. We performed experiments on a real platform with both known and novel objects, i.e., objects the robot trained with, and previously unseen objects. Grasp correction had a 75% success rate on known objects, and 73% on new objects. We compared our model to a traditional regression model that succeeded in correcting grasps in only 38% of cases.

Original language	English
Title of host publication	2016 IEEE International Conference on Robotics and Automation (ICRA 2016)
Subtitle of host publication	Proceedings of a meeting held 16-21 May 2016, Stockholm, Sweden
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	193-200
Number of pages	8
ISBN (Electronic)	9781467380263
ISBN (Print)	9781467380270
DOIs	https://doi.org/10.1109/ICRA.2016.7487133
Publication status	Published - Aug 2016

Access to Document

10.1109/ICRA.2016.7487133

Full-text PDF (accepted author manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://ieeexplore.ieee.org/document/7487133/. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 14.4 MB

Fingerprint Dive into the research topics of 'Probabilistic consolidation of grasp experience'. Together they form a unique fingerprint.

Cite this

Bekiroglu, Y., Damianou, A., Detry, R., Stork, J. A., Kragic, D., & Ek, C. H. (2016). Probabilistic consolidation of grasp experience. In 2016 IEEE International Conference on Robotics and Automation (ICRA 2016): Proceedings of a meeting held 16-21 May 2016, Stockholm, Sweden (pp. 193-200). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ICRA.2016.7487133

Bekiroglu, Yasemin ; Damianou, Andreas ; Detry, Renaud ; Stork, Johannes A ; Kragic, Danica ; Ek, Carl Henrik. / Probabilistic consolidation of grasp experience. 2016 IEEE International Conference on Robotics and Automation (ICRA 2016): Proceedings of a meeting held 16-21 May 2016, Stockholm, Sweden. Institute of Electrical and Electronics Engineers (IEEE), 2016. pp. 193-200

@inproceedings{8db82ef9cc2949b0be2536573e38346b,

title = "Probabilistic consolidation of grasp experience",

abstract = "We present a probabilistic model for joint representation of several sensory modalities and action parameters in a robotic grasping scenario. Our non-linear probabilistic latent variable model encodes relationships between grasp-related parameters, learns the importance of features, and expresses confidence in estimates. The model learns associations between stable and unstable grasps that it experiences during an exploration phase. We demonstrate the applicability of the model for estimating grasp stability, correcting grasps, identifying objects based on tactile imprints and predicting tactile imprints from object-relative gripper poses. We performed experiments on a real platform with both known and novel objects, i.e., objects the robot trained with, and previously unseen objects. Grasp correction had a 75% success rate on known objects, and 73% on new objects. We compared our model to a traditional regression model that succeeded in correcting grasps in only 38% of cases.",

author = "Yasemin Bekiroglu and Andreas Damianou and Renaud Detry and Stork, {Johannes A} and Danica Kragic and Ek, {Carl Henrik}",

year = "2016",

month = aug,

doi = "10.1109/ICRA.2016.7487133",

language = "English",

isbn = "9781467380270",

pages = "193--200",

booktitle = "2016 IEEE International Conference on Robotics and Automation (ICRA 2016)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

}

Bekiroglu, Y, Damianou, A, Detry, R, Stork, JA, Kragic, D & Ek, CH 2016, Probabilistic consolidation of grasp experience. in 2016 IEEE International Conference on Robotics and Automation (ICRA 2016): Proceedings of a meeting held 16-21 May 2016, Stockholm, Sweden. Institute of Electrical and Electronics Engineers (IEEE), pp. 193-200. https://doi.org/10.1109/ICRA.2016.7487133

Probabilistic consolidation of grasp experience. / Bekiroglu, Yasemin; Damianou, Andreas; Detry, Renaud; Stork, Johannes A; Kragic, Danica; Ek, Carl Henrik.

2016 IEEE International Conference on Robotics and Automation (ICRA 2016): Proceedings of a meeting held 16-21 May 2016, Stockholm, Sweden. Institute of Electrical and Electronics Engineers (IEEE), 2016. p. 193-200.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Probabilistic consolidation of grasp experience

AU - Bekiroglu, Yasemin

AU - Damianou, Andreas

AU - Detry, Renaud

AU - Stork, Johannes A

AU - Kragic, Danica

AU - Ek, Carl Henrik

PY - 2016/8

Y1 - 2016/8

N2 - We present a probabilistic model for joint representation of several sensory modalities and action parameters in a robotic grasping scenario. Our non-linear probabilistic latent variable model encodes relationships between grasp-related parameters, learns the importance of features, and expresses confidence in estimates. The model learns associations between stable and unstable grasps that it experiences during an exploration phase. We demonstrate the applicability of the model for estimating grasp stability, correcting grasps, identifying objects based on tactile imprints and predicting tactile imprints from object-relative gripper poses. We performed experiments on a real platform with both known and novel objects, i.e., objects the robot trained with, and previously unseen objects. Grasp correction had a 75% success rate on known objects, and 73% on new objects. We compared our model to a traditional regression model that succeeded in correcting grasps in only 38% of cases.

AB - We present a probabilistic model for joint representation of several sensory modalities and action parameters in a robotic grasping scenario. Our non-linear probabilistic latent variable model encodes relationships between grasp-related parameters, learns the importance of features, and expresses confidence in estimates. The model learns associations between stable and unstable grasps that it experiences during an exploration phase. We demonstrate the applicability of the model for estimating grasp stability, correcting grasps, identifying objects based on tactile imprints and predicting tactile imprints from object-relative gripper poses. We performed experiments on a real platform with both known and novel objects, i.e., objects the robot trained with, and previously unseen objects. Grasp correction had a 75% success rate on known objects, and 73% on new objects. We compared our model to a traditional regression model that succeeded in correcting grasps in only 38% of cases.

U2 - 10.1109/ICRA.2016.7487133

DO - 10.1109/ICRA.2016.7487133

M3 - Conference Contribution (Conference Proceeding)

SN - 9781467380270

SP - 193

EP - 200

BT - 2016 IEEE International Conference on Robotics and Automation (ICRA 2016)

PB - Institute of Electrical and Electronics Engineers (IEEE)

ER -