Virtual Reality for Demonstrating Tool Pose

Research output: Contribution to conference without publisher/journalConference abstract for conferenceResearchpeer-review

44 Downloads (Pure)

Abstract

We investigate the viability of collecting demonstrations, for learning robot control, in Virtual Reality (VR). If acceptable demonstrations can be collected offline using VR, problems with affecting the production line can be avoided and large amounts of realistic data from the actual production environment can be used.
We show that the relatively low detail data that can be captured from the production line using a RGB-D camera provides sufficient information for an expert to perform acceptable demonstrations in Virtual-Reality. However, performance is not yet on par with what can be achieved through real world observations. We propose several future improvements that will help close the gap.
Original languageEnglish
Publication date5 Oct 2018
Publication statusPublished - 5 Oct 2018
EventAutomating Robot Experiments: Manipulation and Learning - Madrid, Spain
Duration: 5 Oct 20185 Oct 2018
http://www.cs.cmu.edu/~lerrelp/AutomatingRobotExperiments/

Workshop

WorkshopAutomating Robot Experiments
CountrySpain
CityMadrid
Period05/10/201805/10/2018
Internet address

Fingerprint

Virtual reality
Demonstrations
Robot learning
Cameras

Cite this

Philipsen, M. P., Wu, H., & Moeslund, T. B. (2018). Virtual Reality for Demonstrating Tool Pose. Abstract from Automating Robot Experiments, Madrid, Spain.
Philipsen, Mark Philip ; Wu, Haiyan ; Moeslund, Thomas B. / Virtual Reality for Demonstrating Tool Pose. Abstract from Automating Robot Experiments, Madrid, Spain.
@conference{0d2a24b618dc481d8503c1a1a95ae493,
title = "Virtual Reality for Demonstrating Tool Pose",
abstract = "We investigate the viability of collecting demonstrations, for learning robot control, in Virtual Reality (VR). If acceptable demonstrations can be collected offline using VR, problems with affecting the production line can be avoided and large amounts of realistic data from the actual production environment can be used.We show that the relatively low detail data that can be captured from the production line using a RGB-D camera provides sufficient information for an expert to perform acceptable demonstrations in Virtual-Reality. However, performance is not yet on par with what can be achieved through real world observations. We propose several future improvements that will help close the gap.",
keywords = "VR, Robot, RGB-D, demonstration",
author = "Philipsen, {Mark Philip} and Haiyan Wu and Moeslund, {Thomas B.}",
year = "2018",
month = "10",
day = "5",
language = "English",
note = "Automating Robot Experiments : Manipulation and Learning ; Conference date: 05-10-2018 Through 05-10-2018",
url = "http://www.cs.cmu.edu/~lerrelp/AutomatingRobotExperiments/",

}

Philipsen, MP, Wu, H & Moeslund, TB 2018, 'Virtual Reality for Demonstrating Tool Pose', Automating Robot Experiments, Madrid, Spain, 05/10/2018 - 05/10/2018.

Virtual Reality for Demonstrating Tool Pose. / Philipsen, Mark Philip; Wu, Haiyan; Moeslund, Thomas B.

2018. Abstract from Automating Robot Experiments, Madrid, Spain.

Research output: Contribution to conference without publisher/journalConference abstract for conferenceResearchpeer-review

TY - ABST

T1 - Virtual Reality for Demonstrating Tool Pose

AU - Philipsen, Mark Philip

AU - Wu, Haiyan

AU - Moeslund, Thomas B.

PY - 2018/10/5

Y1 - 2018/10/5

N2 - We investigate the viability of collecting demonstrations, for learning robot control, in Virtual Reality (VR). If acceptable demonstrations can be collected offline using VR, problems with affecting the production line can be avoided and large amounts of realistic data from the actual production environment can be used.We show that the relatively low detail data that can be captured from the production line using a RGB-D camera provides sufficient information for an expert to perform acceptable demonstrations in Virtual-Reality. However, performance is not yet on par with what can be achieved through real world observations. We propose several future improvements that will help close the gap.

AB - We investigate the viability of collecting demonstrations, for learning robot control, in Virtual Reality (VR). If acceptable demonstrations can be collected offline using VR, problems with affecting the production line can be avoided and large amounts of realistic data from the actual production environment can be used.We show that the relatively low detail data that can be captured from the production line using a RGB-D camera provides sufficient information for an expert to perform acceptable demonstrations in Virtual-Reality. However, performance is not yet on par with what can be achieved through real world observations. We propose several future improvements that will help close the gap.

KW - VR

KW - Robot

KW - RGB-D

KW - demonstration

M3 - Conference abstract for conference

ER -

Philipsen MP, Wu H, Moeslund TB. Virtual Reality for Demonstrating Tool Pose. 2018. Abstract from Automating Robot Experiments, Madrid, Spain.