Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction

Saskia Golz; Christian Osendorfer; Sami Haddadin

doi:10.1109/ICRA.2015.7139726

Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction

Saskia Golz, Christian Osendorfer, Sami Haddadin

Leibniz Universität Hannover

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

62 Scopus citations

Abstract

Detecting and interpreting contacts is a crucial aspect of physical Human-Robot Interaction. In order to discriminate between intended and unintended contact types, we derive a set of linear and non-linear features based on physical contact model insights and from observing real impact data that may even rely on proprioceptive sensation only. We implement a classification system with a standard non-linear Support Vector Machine and show empirically both in simulations and on a real robot the high accuracy in off- as well as on-line settings of the system. We argue that these successful results are based on our feature design derived from first principles.

Original language	English
Title of host publication	2015 IEEE International Conference on Robotics and Automation, ICRA 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3788-3794
Number of pages	7
Edition	June
ISBN (Electronic)	9781479969234
DOIs	https://doi.org/10.1109/ICRA.2015.7139726
State	Published - 29 Jun 2015
Externally published	Yes
Event	2015 IEEE International Conference on Robotics and Automation, ICRA 2015 - Seattle, United States Duration: 26 May 2015 → 30 May 2015

Publication series

Name	Proceedings - IEEE International Conference on Robotics and Automation
Number	June
Volume	2015-June
ISSN (Print)	1050-4729

Conference

Conference	2015 IEEE International Conference on Robotics and Automation, ICRA 2015
Country/Territory	United States
City	Seattle
Period	26/05/15 → 30/05/15

Access to Document

10.1109/ICRA.2015.7139726

Cite this

Golz, S., Osendorfer, C., & Haddadin, S. (2015). Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction. In 2015 IEEE International Conference on Robotics and Automation, ICRA 2015 (June ed., pp. 3788-3794). Article 7139726 (Proceedings - IEEE International Conference on Robotics and Automation; Vol. 2015-June, No. June). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICRA.2015.7139726

Golz, Saskia ; Osendorfer, Christian ; Haddadin, Sami. / Using tactile sensation for learning contact knowledge : Discriminate collision from physical interaction. 2015 IEEE International Conference on Robotics and Automation, ICRA 2015. June. ed. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 3788-3794 (Proceedings - IEEE International Conference on Robotics and Automation; June).

@inproceedings{d657591753af4204b2af11424f168491,

title = "Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction",

abstract = "Detecting and interpreting contacts is a crucial aspect of physical Human-Robot Interaction. In order to discriminate between intended and unintended contact types, we derive a set of linear and non-linear features based on physical contact model insights and from observing real impact data that may even rely on proprioceptive sensation only. We implement a classification system with a standard non-linear Support Vector Machine and show empirically both in simulations and on a real robot the high accuracy in off- as well as on-line settings of the system. We argue that these successful results are based on our feature design derived from first principles.",

author = "Saskia Golz and Christian Osendorfer and Sami Haddadin",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 2015 IEEE International Conference on Robotics and Automation, ICRA 2015 ; Conference date: 26-05-2015 Through 30-05-2015",

year = "2015",

month = jun,

day = "29",

doi = "10.1109/ICRA.2015.7139726",

language = "English",

series = "Proceedings - IEEE International Conference on Robotics and Automation",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "June",

pages = "3788--3794",

booktitle = "2015 IEEE International Conference on Robotics and Automation, ICRA 2015",

edition = "June",

}

Golz, S, Osendorfer, C & Haddadin, S 2015, Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction. in 2015 IEEE International Conference on Robotics and Automation, ICRA 2015. June edn, 7139726, Proceedings - IEEE International Conference on Robotics and Automation, no. June, vol. 2015-June, Institute of Electrical and Electronics Engineers Inc., pp. 3788-3794, 2015 IEEE International Conference on Robotics and Automation, ICRA 2015, Seattle, United States, 26/05/15. https://doi.org/10.1109/ICRA.2015.7139726

Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction. / Golz, Saskia; Osendorfer, Christian; Haddadin, Sami.
2015 IEEE International Conference on Robotics and Automation, ICRA 2015. June. ed. Institute of Electrical and Electronics Engineers Inc., 2015. p. 3788-3794 7139726 (Proceedings - IEEE International Conference on Robotics and Automation; Vol. 2015-June, No. June).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Using tactile sensation for learning contact knowledge

T2 - 2015 IEEE International Conference on Robotics and Automation, ICRA 2015

AU - Golz, Saskia

AU - Osendorfer, Christian

AU - Haddadin, Sami

PY - 2015/6/29

Y1 - 2015/6/29

N2 - Detecting and interpreting contacts is a crucial aspect of physical Human-Robot Interaction. In order to discriminate between intended and unintended contact types, we derive a set of linear and non-linear features based on physical contact model insights and from observing real impact data that may even rely on proprioceptive sensation only. We implement a classification system with a standard non-linear Support Vector Machine and show empirically both in simulations and on a real robot the high accuracy in off- as well as on-line settings of the system. We argue that these successful results are based on our feature design derived from first principles.

AB - Detecting and interpreting contacts is a crucial aspect of physical Human-Robot Interaction. In order to discriminate between intended and unintended contact types, we derive a set of linear and non-linear features based on physical contact model insights and from observing real impact data that may even rely on proprioceptive sensation only. We implement a classification system with a standard non-linear Support Vector Machine and show empirically both in simulations and on a real robot the high accuracy in off- as well as on-line settings of the system. We argue that these successful results are based on our feature design derived from first principles.

UR - http://www.scopus.com/inward/record.url?scp=84938224603&partnerID=8YFLogxK

U2 - 10.1109/ICRA.2015.7139726

DO - 10.1109/ICRA.2015.7139726

M3 - Conference contribution

AN - SCOPUS:84938224603

T3 - Proceedings - IEEE International Conference on Robotics and Automation

SP - 3788

EP - 3794

BT - 2015 IEEE International Conference on Robotics and Automation, ICRA 2015

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 26 May 2015 through 30 May 2015

ER -

Golz S, Osendorfer C, Haddadin S. Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction. In 2015 IEEE International Conference on Robotics and Automation, ICRA 2015. June ed. Institute of Electrical and Electronics Engineers Inc. 2015. p. 3788-3794. 7139726. (Proceedings - IEEE International Conference on Robotics and Automation; June). doi: 10.1109/ICRA.2015.7139726

Using tactile sensation for learning contact knowledge: Discriminate collision from physical interaction

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this