Transfer entropy for feature extraction in physical human-robot interaction

Detecting perturbations from low-cost sensors

Erik Berger, David Müller, David Vogt, Bernhard Jung, Hani Ben Amor

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

5 Citations (Scopus)

Abstract

In physical human-robot interaction, robot behavior must be adjusted to forces applied by the human interaction partner. For measuring such forces, special-purpose sensors may be used, e.g. force-torque sensors, that are however often heavy, expensive and prone to noise. In contrast, we propose a machine learning approach for measuring external perturbations of robot behavior that uses commonly available, low-cost sensors only. During the training phase, behavior-specific statistical models of sensor measurements, so-called perturbation filters, are constructed using Principal Component Analysis, Transfer Entropy and Dynamic Mode Decomposition. During behavior execution, perturbation filters compare measured and predicted sensor values for estimating the amount and direction of forces applied by the human interaction partner. Such perturbation filters can therefore be regarded as virtual force sensors that produce continuous estimates of external forces.

Original languageEnglish (US)
Title of host publicationIEEE-RAS International Conference on Humanoid Robots
PublisherIEEE Computer Society
Pages829-834
Number of pages6
Volume2015-February
ISBN (Print)9781479971749
DOIs
StatePublished - Feb 12 2015
Externally publishedYes
Event2014 14th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2014 - Madrid, Spain
Duration: Nov 18 2014Nov 20 2014

Other

Other2014 14th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2014
CountrySpain
CityMadrid
Period11/18/1411/20/14

Fingerprint

Human robot interaction
Feature extraction
Entropy
Sensors
Costs
Robots
Phase behavior
Principal component analysis
Learning systems
Torque
Decomposition

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Human-Computer Interaction
  • Electrical and Electronic Engineering

Cite this

Berger, E., Müller, D., Vogt, D., Jung, B., & Ben Amor, H. (2015). Transfer entropy for feature extraction in physical human-robot interaction: Detecting perturbations from low-cost sensors. In IEEE-RAS International Conference on Humanoid Robots (Vol. 2015-February, pp. 829-834). [7041459] IEEE Computer Society. https://doi.org/10.1109/HUMANOIDS.2014.7041459

Transfer entropy for feature extraction in physical human-robot interaction : Detecting perturbations from low-cost sensors. / Berger, Erik; Müller, David; Vogt, David; Jung, Bernhard; Ben Amor, Hani.

IEEE-RAS International Conference on Humanoid Robots. Vol. 2015-February IEEE Computer Society, 2015. p. 829-834 7041459.

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

Berger, E, Müller, D, Vogt, D, Jung, B & Ben Amor, H 2015, Transfer entropy for feature extraction in physical human-robot interaction: Detecting perturbations from low-cost sensors. in IEEE-RAS International Conference on Humanoid Robots. vol. 2015-February, 7041459, IEEE Computer Society, pp. 829-834, 2014 14th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2014, Madrid, Spain, 11/18/14. https://doi.org/10.1109/HUMANOIDS.2014.7041459
Berger E, Müller D, Vogt D, Jung B, Ben Amor H. Transfer entropy for feature extraction in physical human-robot interaction: Detecting perturbations from low-cost sensors. In IEEE-RAS International Conference on Humanoid Robots. Vol. 2015-February. IEEE Computer Society. 2015. p. 829-834. 7041459 https://doi.org/10.1109/HUMANOIDS.2014.7041459
Berger, Erik ; Müller, David ; Vogt, David ; Jung, Bernhard ; Ben Amor, Hani. / Transfer entropy for feature extraction in physical human-robot interaction : Detecting perturbations from low-cost sensors. IEEE-RAS International Conference on Humanoid Robots. Vol. 2015-February IEEE Computer Society, 2015. pp. 829-834
@inproceedings{3ee9a89d0b1c4aa3911f3898e6d39fd2,
title = "Transfer entropy for feature extraction in physical human-robot interaction: Detecting perturbations from low-cost sensors",
abstract = "In physical human-robot interaction, robot behavior must be adjusted to forces applied by the human interaction partner. For measuring such forces, special-purpose sensors may be used, e.g. force-torque sensors, that are however often heavy, expensive and prone to noise. In contrast, we propose a machine learning approach for measuring external perturbations of robot behavior that uses commonly available, low-cost sensors only. During the training phase, behavior-specific statistical models of sensor measurements, so-called perturbation filters, are constructed using Principal Component Analysis, Transfer Entropy and Dynamic Mode Decomposition. During behavior execution, perturbation filters compare measured and predicted sensor values for estimating the amount and direction of forces applied by the human interaction partner. Such perturbation filters can therefore be regarded as virtual force sensors that produce continuous estimates of external forces.",
author = "Erik Berger and David M{\"u}ller and David Vogt and Bernhard Jung and {Ben Amor}, Hani",
year = "2015",
month = "2",
day = "12",
doi = "10.1109/HUMANOIDS.2014.7041459",
language = "English (US)",
isbn = "9781479971749",
volume = "2015-February",
pages = "829--834",
booktitle = "IEEE-RAS International Conference on Humanoid Robots",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Transfer entropy for feature extraction in physical human-robot interaction

T2 - Detecting perturbations from low-cost sensors

AU - Berger, Erik

AU - Müller, David

AU - Vogt, David

AU - Jung, Bernhard

AU - Ben Amor, Hani

PY - 2015/2/12

Y1 - 2015/2/12

N2 - In physical human-robot interaction, robot behavior must be adjusted to forces applied by the human interaction partner. For measuring such forces, special-purpose sensors may be used, e.g. force-torque sensors, that are however often heavy, expensive and prone to noise. In contrast, we propose a machine learning approach for measuring external perturbations of robot behavior that uses commonly available, low-cost sensors only. During the training phase, behavior-specific statistical models of sensor measurements, so-called perturbation filters, are constructed using Principal Component Analysis, Transfer Entropy and Dynamic Mode Decomposition. During behavior execution, perturbation filters compare measured and predicted sensor values for estimating the amount and direction of forces applied by the human interaction partner. Such perturbation filters can therefore be regarded as virtual force sensors that produce continuous estimates of external forces.

AB - In physical human-robot interaction, robot behavior must be adjusted to forces applied by the human interaction partner. For measuring such forces, special-purpose sensors may be used, e.g. force-torque sensors, that are however often heavy, expensive and prone to noise. In contrast, we propose a machine learning approach for measuring external perturbations of robot behavior that uses commonly available, low-cost sensors only. During the training phase, behavior-specific statistical models of sensor measurements, so-called perturbation filters, are constructed using Principal Component Analysis, Transfer Entropy and Dynamic Mode Decomposition. During behavior execution, perturbation filters compare measured and predicted sensor values for estimating the amount and direction of forces applied by the human interaction partner. Such perturbation filters can therefore be regarded as virtual force sensors that produce continuous estimates of external forces.

UR - http://www.scopus.com/inward/record.url?scp=84945190350&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84945190350&partnerID=8YFLogxK

U2 - 10.1109/HUMANOIDS.2014.7041459

DO - 10.1109/HUMANOIDS.2014.7041459

M3 - Conference contribution

SN - 9781479971749

VL - 2015-February

SP - 829

EP - 834

BT - IEEE-RAS International Conference on Humanoid Robots

PB - IEEE Computer Society

ER -