Bayesian tactile object recognition: learning and recognising objects using a new inexpensive tactile sensor

Tadeo Corradi, Peter Hall, Pejman Iravani

Research output: Contribution to conferencePaper

  • 3 Citations

Abstract

We present a Bayesian approach to tactile object recognition that improves on state-of-the-art in using single-touch events in two ways. First by improving recognition accuracy from about 90\% to about 95\%, using about half the number of touches. Second by reducing the number of touches needed for training from about 200 to about 60. In addition, we use a new tactile sensor that is less than one tenth of the cost of widely available sensors. The paper describes the sensor, the likelihood function used with the Naive Bayes classifier, and experiments on a set of ten real objects. We also provide preliminary results to test our approach for its ability to generalise to previously unencountered objects.

Conference

ConferenceIEEE Interational Conference on Robotics and Automation (ICRA) 2015
CountryUSA United States
CitySeattle
Period26/05/1530/05/15

Fingerprint

Object recognition
Sensors
Classifiers
Costs
Experiments

Keywords

  • tactile sensing
  • object recognition
  • Bayesian
  • Robotics

Cite this

Corradi, T., Hall, P., & Iravani, P. (2015). Bayesian tactile object recognition: learning and recognising objects using a new inexpensive tactile sensor. Paper presented at IEEE Interational Conference on Robotics and Automation (ICRA) 2015, Seattle, USA United States.

Bayesian tactile object recognition : learning and recognising objects using a new inexpensive tactile sensor. / Corradi, Tadeo; Hall, Peter; Iravani, Pejman.

2015. Paper presented at IEEE Interational Conference on Robotics and Automation (ICRA) 2015, Seattle, USA United States.

Research output: Contribution to conferencePaper

Corradi, T, Hall, P & Iravani, P 2015, 'Bayesian tactile object recognition: learning and recognising objects using a new inexpensive tactile sensor' Paper presented at IEEE Interational Conference on Robotics and Automation (ICRA) 2015, Seattle, USA United States, 26/05/15 - 30/05/15, .
Corradi T, Hall P, Iravani P. Bayesian tactile object recognition: learning and recognising objects using a new inexpensive tactile sensor. 2015. Paper presented at IEEE Interational Conference on Robotics and Automation (ICRA) 2015, Seattle, USA United States.
Corradi, Tadeo ; Hall, Peter ; Iravani, Pejman. / Bayesian tactile object recognition : learning and recognising objects using a new inexpensive tactile sensor. Paper presented at IEEE Interational Conference on Robotics and Automation (ICRA) 2015, Seattle, USA United States.
@conference{24ccfde93b2e42109c8c29509997043c,
title = "Bayesian tactile object recognition: learning and recognising objects using a new inexpensive tactile sensor",
abstract = "We present a Bayesian approach to tactile object recognition that improves on state-of-the-art in using single-touch events in two ways. First by improving recognition accuracy from about 90\{\%} to about 95\{\%}, using about half the number of touches. Second by reducing the number of touches needed for training from about 200 to about 60. In addition, we use a new tactile sensor that is less than one tenth of the cost of widely available sensors. The paper describes the sensor, the likelihood function used with the Naive Bayes classifier, and experiments on a set of ten real objects. We also provide preliminary results to test our approach for its ability to generalise to previously unencountered objects.",
keywords = "tactile sensing, object recognition, Bayesian, Robotics",
author = "Tadeo Corradi and Peter Hall and Pejman Iravani",
year = "2015",
month = "5",
language = "English",
note = "IEEE Interational Conference on Robotics and Automation (ICRA) 2015 ; Conference date: 26-05-2015 Through 30-05-2015",

}

TY - CONF

T1 - Bayesian tactile object recognition

T2 - learning and recognising objects using a new inexpensive tactile sensor

AU - Corradi,Tadeo

AU - Hall,Peter

AU - Iravani,Pejman

PY - 2015/5

Y1 - 2015/5

N2 - We present a Bayesian approach to tactile object recognition that improves on state-of-the-art in using single-touch events in two ways. First by improving recognition accuracy from about 90\% to about 95\%, using about half the number of touches. Second by reducing the number of touches needed for training from about 200 to about 60. In addition, we use a new tactile sensor that is less than one tenth of the cost of widely available sensors. The paper describes the sensor, the likelihood function used with the Naive Bayes classifier, and experiments on a set of ten real objects. We also provide preliminary results to test our approach for its ability to generalise to previously unencountered objects.

AB - We present a Bayesian approach to tactile object recognition that improves on state-of-the-art in using single-touch events in two ways. First by improving recognition accuracy from about 90\% to about 95\%, using about half the number of touches. Second by reducing the number of touches needed for training from about 200 to about 60. In addition, we use a new tactile sensor that is less than one tenth of the cost of widely available sensors. The paper describes the sensor, the likelihood function used with the Naive Bayes classifier, and experiments on a set of ten real objects. We also provide preliminary results to test our approach for its ability to generalise to previously unencountered objects.

KW - tactile sensing

KW - object recognition

KW - Bayesian

KW - Robotics

M3 - Paper

ER -