Learning from demonstrations with partially observable task parameters

Tohid Alizadeh; Sylvain Calinon; Darwin G. Caldwell

doi:10.1109/ICRA.2014.6907335

Learning from demonstrations with partially observable task parameters

Tohid Alizadeh, Sylvain Calinon, Darwin G. Caldwell

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

27 Citations (Scopus)

Abstract

Robot learning from demonstrations requires the robot to learn and adapt movements to new situations, often characterized by position and orientation of objects or landmarks in the robot's environment. In the task-parameterized Gaussian mixture model framework, the movements are considered to be modulated with respect to a set of candidate frames of reference (coordinate systems) attached to a set of objects in the robot workspace. Following a similar approach, this paper addresses the problem of having missing candidate frames during the demonstrations and reproductions, which can happen in various situations such as visual occlusion, sensor unavailability, or tasks with a variable number of descriptive features. We study this problem with a dust sweeping task in which the robot requires to consider a variable amount of dust areas to clean for each reproduction trial.

Original language	English
Title of host publication	Proceedings - IEEE International Conference on Robotics and Automation
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3309-3314
Number of pages	6
DOIs	https://doi.org/10.1109/ICRA.2014.6907335
Publication status	Published - Sep 22 2014
Externally published	Yes
Event	2014 IEEE International Conference on Robotics and Automation, ICRA 2014 - Hong Kong, China Duration: May 31 2014 → Jun 7 2014

Other

Other	2014 IEEE International Conference on Robotics and Automation, ICRA 2014
Country	China
City	Hong Kong
Period	5/31/14 → 6/7/14

ASJC Scopus subject areas

Software
Artificial Intelligence
Control and Systems Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/ICRA.2014.6907335

Fingerprint Dive into the research topics of 'Learning from demonstrations with partially observable task parameters'. Together they form a unique fingerprint.

Cite this

Learning from demonstrations with partially observable task parameters. / Alizadeh, Tohid; Calinon, Sylvain; Caldwell, Darwin G.

Proceedings - IEEE International Conference on Robotics and Automation. Institute of Electrical and Electronics Engineers Inc., 2014. p. 3309-3314 6907335.

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

Alizadeh, T, Calinon, S & Caldwell, DG 2014, Learning from demonstrations with partially observable task parameters. in Proceedings - IEEE International Conference on Robotics and Automation., 6907335, Institute of Electrical and Electronics Engineers Inc., pp. 3309-3314, 2014 IEEE International Conference on Robotics and Automation, ICRA 2014, Hong Kong, China, 5/31/14. https://doi.org/10.1109/ICRA.2014.6907335

@inproceedings{39f5e36091fb47dd9912f445423c44b5,

title = "Learning from demonstrations with partially observable task parameters",

abstract = "Robot learning from demonstrations requires the robot to learn and adapt movements to new situations, often characterized by position and orientation of objects or landmarks in the robot's environment. In the task-parameterized Gaussian mixture model framework, the movements are considered to be modulated with respect to a set of candidate frames of reference (coordinate systems) attached to a set of objects in the robot workspace. Following a similar approach, this paper addresses the problem of having missing candidate frames during the demonstrations and reproductions, which can happen in various situations such as visual occlusion, sensor unavailability, or tasks with a variable number of descriptive features. We study this problem with a dust sweeping task in which the robot requires to consider a variable amount of dust areas to clean for each reproduction trial.",

author = "Tohid Alizadeh and Sylvain Calinon and Caldwell, {Darwin G.}",

year = "2014",

month = sep,

day = "22",

doi = "10.1109/ICRA.2014.6907335",

language = "English",

pages = "3309--3314",

booktitle = "Proceedings - IEEE International Conference on Robotics and Automation",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "2014 IEEE International Conference on Robotics and Automation, ICRA 2014 ; Conference date: 31-05-2014 Through 07-06-2014",

}

TY - GEN

T1 - Learning from demonstrations with partially observable task parameters

AU - Alizadeh, Tohid

AU - Calinon, Sylvain

AU - Caldwell, Darwin G.

PY - 2014/9/22

Y1 - 2014/9/22

N2 - Robot learning from demonstrations requires the robot to learn and adapt movements to new situations, often characterized by position and orientation of objects or landmarks in the robot's environment. In the task-parameterized Gaussian mixture model framework, the movements are considered to be modulated with respect to a set of candidate frames of reference (coordinate systems) attached to a set of objects in the robot workspace. Following a similar approach, this paper addresses the problem of having missing candidate frames during the demonstrations and reproductions, which can happen in various situations such as visual occlusion, sensor unavailability, or tasks with a variable number of descriptive features. We study this problem with a dust sweeping task in which the robot requires to consider a variable amount of dust areas to clean for each reproduction trial.

AB - Robot learning from demonstrations requires the robot to learn and adapt movements to new situations, often characterized by position and orientation of objects or landmarks in the robot's environment. In the task-parameterized Gaussian mixture model framework, the movements are considered to be modulated with respect to a set of candidate frames of reference (coordinate systems) attached to a set of objects in the robot workspace. Following a similar approach, this paper addresses the problem of having missing candidate frames during the demonstrations and reproductions, which can happen in various situations such as visual occlusion, sensor unavailability, or tasks with a variable number of descriptive features. We study this problem with a dust sweeping task in which the robot requires to consider a variable amount of dust areas to clean for each reproduction trial.

UR - http://www.scopus.com/inward/record.url?scp=84929191868&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929191868&partnerID=8YFLogxK

U2 - 10.1109/ICRA.2014.6907335

DO - 10.1109/ICRA.2014.6907335

M3 - Conference contribution

AN - SCOPUS:84929191868

SP - 3309

EP - 3314

BT - Proceedings - IEEE International Conference on Robotics and Automation

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 IEEE International Conference on Robotics and Automation, ICRA 2014

Y2 - 31 May 2014 through 7 June 2014

ER -

Learning from demonstrations with partially observable task parameters

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Cite this