A Model-Based Approach to Finding Substitute Tools in 3D Vision Data

Paulo Abelha Ferreira, Frank Guerin, Markus Schoeler

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

10 Citations (Scopus)
6 Downloads (Pure)

Abstract

A robot can feasibly be given knowledge of a set of tools for manipulation activities (e.g. hammer, knife, spatula). If the robot then operates outside a closed environment it is likely to face situations where the tool it knows is not available, but alternative unknown tools are present. We tackle the problem of finding the best substitute tool based solely on 3D vision data. Our approach has simple hand-coded models of known tools in terms of superquadrics and relationships among them. Our system attempts to fit these models to point clouds of unknown tools, producing a numeric value for how good a fit is. This value can be used to rate candidate substitutes. We explicitly control how closely each part of a tool must match our model, under direction from parameters of a target task. We allow bottom-up information from segmentation to dictate the sizes that should be considered for various parts of the tool. These ideas allow for a flexible matching so that tools may be superficially quite different, but similar in the way that matters. We evaluate our system's ratings relative to other approaches and relative to human performance in the same task. This is an approach to knowledge transfer, via a suitable representation and reasoning engine, and we discuss how this could be extended to transfer in planning.
Original languageEnglish
Title of host publication2016 IEEE International Conference on Robotics and Automation (ICRA 2016)
PublisherIEEE Press
ISBN (Print)9781467380270
DOIs
Publication statusPublished - 2016

Fingerprint

Robots
Hammers
Engines
Planning

Keywords

  • three-dimensional (3-D) displays
  • shape
  • data models
  • robots
  • computational modeling
  • solid modeling
  • numerical models

Cite this

Abelha Ferreira, P., Guerin, F., & Schoeler, M. (2016). A Model-Based Approach to Finding Substitute Tools in 3D Vision Data. In 2016 IEEE International Conference on Robotics and Automation (ICRA 2016) IEEE Press. https://doi.org/10.1109/ICRA.2016.7487400

A Model-Based Approach to Finding Substitute Tools in 3D Vision Data. / Abelha Ferreira, Paulo; Guerin, Frank; Schoeler, Markus.

2016 IEEE International Conference on Robotics and Automation (ICRA 2016). IEEE Press, 2016.

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

Abelha Ferreira, P, Guerin, F & Schoeler, M 2016, A Model-Based Approach to Finding Substitute Tools in 3D Vision Data. in 2016 IEEE International Conference on Robotics and Automation (ICRA 2016). IEEE Press. https://doi.org/10.1109/ICRA.2016.7487400
Abelha Ferreira P, Guerin F, Schoeler M. A Model-Based Approach to Finding Substitute Tools in 3D Vision Data. In 2016 IEEE International Conference on Robotics and Automation (ICRA 2016). IEEE Press. 2016 https://doi.org/10.1109/ICRA.2016.7487400
Abelha Ferreira, Paulo ; Guerin, Frank ; Schoeler, Markus. / A Model-Based Approach to Finding Substitute Tools in 3D Vision Data. 2016 IEEE International Conference on Robotics and Automation (ICRA 2016). IEEE Press, 2016.
@inproceedings{a2fa1250a0ed40efaea260cc14decd15,
title = "A Model-Based Approach to Finding Substitute Tools in 3D Vision Data",
abstract = "A robot can feasibly be given knowledge of a set of tools for manipulation activities (e.g. hammer, knife, spatula). If the robot then operates outside a closed environment it is likely to face situations where the tool it knows is not available, but alternative unknown tools are present. We tackle the problem of finding the best substitute tool based solely on 3D vision data. Our approach has simple hand-coded models of known tools in terms of superquadrics and relationships among them. Our system attempts to fit these models to point clouds of unknown tools, producing a numeric value for how good a fit is. This value can be used to rate candidate substitutes. We explicitly control how closely each part of a tool must match our model, under direction from parameters of a target task. We allow bottom-up information from segmentation to dictate the sizes that should be considered for various parts of the tool. These ideas allow for a flexible matching so that tools may be superficially quite different, but similar in the way that matters. We evaluate our system's ratings relative to other approaches and relative to human performance in the same task. This is an approach to knowledge transfer, via a suitable representation and reasoning engine, and we discuss how this could be extended to transfer in planning.",
keywords = "three-dimensional (3-D) displays, shape, data models, robots, computational modeling, solid modeling, numerical models",
author = "{Abelha Ferreira}, Paulo and Frank Guerin and Markus Schoeler",
note = "Paulo Abelha is supported by the Brazilian agency CAPES through the program Science without Borders. Thanks to the University of Aberdeen’s ABVenture Zone for equipment use. We got a lot of very helpful advice and assistance from the following people: Severin Fichtl, Dirk Kraft, Norbert Kruger, Karthik Varadarajan, Markus Vincze, Florentin ¨ Worg ¨ otter.",
year = "2016",
doi = "10.1109/ICRA.2016.7487400",
language = "English",
isbn = "9781467380270",
booktitle = "2016 IEEE International Conference on Robotics and Automation (ICRA 2016)",
publisher = "IEEE Press",

}

TY - GEN

T1 - A Model-Based Approach to Finding Substitute Tools in 3D Vision Data

AU - Abelha Ferreira, Paulo

AU - Guerin, Frank

AU - Schoeler, Markus

N1 - Paulo Abelha is supported by the Brazilian agency CAPES through the program Science without Borders. Thanks to the University of Aberdeen’s ABVenture Zone for equipment use. We got a lot of very helpful advice and assistance from the following people: Severin Fichtl, Dirk Kraft, Norbert Kruger, Karthik Varadarajan, Markus Vincze, Florentin ¨ Worg ¨ otter.

PY - 2016

Y1 - 2016

N2 - A robot can feasibly be given knowledge of a set of tools for manipulation activities (e.g. hammer, knife, spatula). If the robot then operates outside a closed environment it is likely to face situations where the tool it knows is not available, but alternative unknown tools are present. We tackle the problem of finding the best substitute tool based solely on 3D vision data. Our approach has simple hand-coded models of known tools in terms of superquadrics and relationships among them. Our system attempts to fit these models to point clouds of unknown tools, producing a numeric value for how good a fit is. This value can be used to rate candidate substitutes. We explicitly control how closely each part of a tool must match our model, under direction from parameters of a target task. We allow bottom-up information from segmentation to dictate the sizes that should be considered for various parts of the tool. These ideas allow for a flexible matching so that tools may be superficially quite different, but similar in the way that matters. We evaluate our system's ratings relative to other approaches and relative to human performance in the same task. This is an approach to knowledge transfer, via a suitable representation and reasoning engine, and we discuss how this could be extended to transfer in planning.

AB - A robot can feasibly be given knowledge of a set of tools for manipulation activities (e.g. hammer, knife, spatula). If the robot then operates outside a closed environment it is likely to face situations where the tool it knows is not available, but alternative unknown tools are present. We tackle the problem of finding the best substitute tool based solely on 3D vision data. Our approach has simple hand-coded models of known tools in terms of superquadrics and relationships among them. Our system attempts to fit these models to point clouds of unknown tools, producing a numeric value for how good a fit is. This value can be used to rate candidate substitutes. We explicitly control how closely each part of a tool must match our model, under direction from parameters of a target task. We allow bottom-up information from segmentation to dictate the sizes that should be considered for various parts of the tool. These ideas allow for a flexible matching so that tools may be superficially quite different, but similar in the way that matters. We evaluate our system's ratings relative to other approaches and relative to human performance in the same task. This is an approach to knowledge transfer, via a suitable representation and reasoning engine, and we discuss how this could be extended to transfer in planning.

KW - three-dimensional (3-D) displays

KW - shape

KW - data models

KW - robots

KW - computational modeling

KW - solid modeling

KW - numerical models

U2 - 10.1109/ICRA.2016.7487400

DO - 10.1109/ICRA.2016.7487400

M3 - Conference contribution

SN - 9781467380270

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

PB - IEEE Press

ER -