A virtual demonstrator environment for robot imitation learning

Di Wei Huang; Garrett Katz; Joshua Langsfeld; Rodolphe Gentili; James Reggia

doi:10.1109/TePRA.2015.7219691

A virtual demonstrator environment for robot imitation learning

Di Wei Huang, Garrett Katz, Joshua Langsfeld, Rodolphe Gentili, James Reggia

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

9 Scopus citations

Abstract

To support studies in robot imitation learning, this paper presents a software platform, SMILE (Simulator for Maryland Imitation Learning Environment), specifically targeting tasks in which exact human motions are not critical. We hypothesize that in this class of tasks, object behaviors are far more important than human behaviors, and thus one can significantly reduce complexity by not processing human motions at all. As such, SMILE simulates a virtual environment where a human demonstrator can manipulate objects using GUI controls without body parts being visible to a robot in the same environment. Imitation learning is therefore based on the behaviors of manipulated objects only. A simple Matlab interface for programming a simulated robot is also provided in SMILE, along with an XML interface for initializing objects in the virtual environment. SMILE lowers the barriers for studying robot imitation learning by (1) simplifying learning by making the human demonstrator be a virtual presence and (2) eliminating the immediate need to purchase special equipment for motion capturing.

Original language	English (US)
Title of host publication	2015 IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015
Publisher	IEEE Computer Society
Volume	2015-August
ISBN (Electronic)	9781479987573, 9781479987573
DOIs	https://doi.org/10.1109/TePRA.2015.7219691
State	Published - Aug 24 2015
Externally published	Yes
Event	IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015 - Woburn, United States Duration: May 11 2015 → May 12 2015

Other

Other	IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015
Country/Territory	United States
City	Woburn
Period	5/11/15 → 5/12/15

Keywords

Graphical user interfaces
Joints
MATLAB
Robot kinematics
Robot sensing systems
XML

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Computer Science Applications
Computer Vision and Pattern Recognition
Electrical and Electronic Engineering

Access to Document

10.1109/TePRA.2015.7219691

Cite this

A virtual demonstrator environment for robot imitation learning. / Huang, Di Wei; Katz, Garrett; Langsfeld, Joshua; Gentili, Rodolphe; Reggia, James.

2015 IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015. Vol. 2015-August IEEE Computer Society, 2015. 7219691.

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

Huang, DW, Katz, G, Langsfeld, J, Gentili, R & Reggia, J 2015, A virtual demonstrator environment for robot imitation learning. in 2015 IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015. vol. 2015-August, 7219691, IEEE Computer Society, IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015, Woburn, United States, 5/11/15. https://doi.org/10.1109/TePRA.2015.7219691

@inproceedings{558a5b60f2f34d6eb2559ebaa65cbc0d,

title = "A virtual demonstrator environment for robot imitation learning",

abstract = "To support studies in robot imitation learning, this paper presents a software platform, SMILE (Simulator for Maryland Imitation Learning Environment), specifically targeting tasks in which exact human motions are not critical. We hypothesize that in this class of tasks, object behaviors are far more important than human behaviors, and thus one can significantly reduce complexity by not processing human motions at all. As such, SMILE simulates a virtual environment where a human demonstrator can manipulate objects using GUI controls without body parts being visible to a robot in the same environment. Imitation learning is therefore based on the behaviors of manipulated objects only. A simple Matlab interface for programming a simulated robot is also provided in SMILE, along with an XML interface for initializing objects in the virtual environment. SMILE lowers the barriers for studying robot imitation learning by (1) simplifying learning by making the human demonstrator be a virtual presence and (2) eliminating the immediate need to purchase special equipment for motion capturing.",

keywords = "Graphical user interfaces, Joints, MATLAB, Robot kinematics, Robot sensing systems, XML",

author = "Huang, {Di Wei} and Garrett Katz and Joshua Langsfeld and Rodolphe Gentili and James Reggia",

year = "2015",

month = aug,

day = "24",

doi = "10.1109/TePRA.2015.7219691",

language = "English (US)",

volume = "2015-August",

booktitle = "2015 IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015",

publisher = "IEEE Computer Society",

address = "United States",

note = "IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015 ; Conference date: 11-05-2015 Through 12-05-2015",

}

TY - GEN

T1 - A virtual demonstrator environment for robot imitation learning

AU - Huang, Di Wei

AU - Katz, Garrett

AU - Langsfeld, Joshua

AU - Gentili, Rodolphe

AU - Reggia, James

PY - 2015/8/24

Y1 - 2015/8/24

N2 - To support studies in robot imitation learning, this paper presents a software platform, SMILE (Simulator for Maryland Imitation Learning Environment), specifically targeting tasks in which exact human motions are not critical. We hypothesize that in this class of tasks, object behaviors are far more important than human behaviors, and thus one can significantly reduce complexity by not processing human motions at all. As such, SMILE simulates a virtual environment where a human demonstrator can manipulate objects using GUI controls without body parts being visible to a robot in the same environment. Imitation learning is therefore based on the behaviors of manipulated objects only. A simple Matlab interface for programming a simulated robot is also provided in SMILE, along with an XML interface for initializing objects in the virtual environment. SMILE lowers the barriers for studying robot imitation learning by (1) simplifying learning by making the human demonstrator be a virtual presence and (2) eliminating the immediate need to purchase special equipment for motion capturing.

AB - To support studies in robot imitation learning, this paper presents a software platform, SMILE (Simulator for Maryland Imitation Learning Environment), specifically targeting tasks in which exact human motions are not critical. We hypothesize that in this class of tasks, object behaviors are far more important than human behaviors, and thus one can significantly reduce complexity by not processing human motions at all. As such, SMILE simulates a virtual environment where a human demonstrator can manipulate objects using GUI controls without body parts being visible to a robot in the same environment. Imitation learning is therefore based on the behaviors of manipulated objects only. A simple Matlab interface for programming a simulated robot is also provided in SMILE, along with an XML interface for initializing objects in the virtual environment. SMILE lowers the barriers for studying robot imitation learning by (1) simplifying learning by making the human demonstrator be a virtual presence and (2) eliminating the immediate need to purchase special equipment for motion capturing.

KW - Graphical user interfaces

KW - Joints

KW - MATLAB

KW - Robot kinematics

KW - Robot sensing systems

KW - XML

UR - http://www.scopus.com/inward/record.url?scp=84949979273&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949979273&partnerID=8YFLogxK

U2 - 10.1109/TePRA.2015.7219691

DO - 10.1109/TePRA.2015.7219691

M3 - Conference contribution

AN - SCOPUS:84949979273

VL - 2015-August

BT - 2015 IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015

PB - IEEE Computer Society

T2 - IEEE International Conference on Technologies for Practical Robot Applications, TePRA 2015

Y2 - 11 May 2015 through 12 May 2015

ER -

A virtual demonstrator environment for robot imitation learning

Abstract

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this