iTP-LfD: Improved Task Parametrised Learning from Demonstration for Adaptive Path Generation of Cobot

Shirine El Zaatari; Yuqi Wang; Weidong Li; Yiqun Peng

doi:10.1016/j.rcim.2020.102109

iTP-LfD: Improved Task Parametrised Learning from Demonstration for Adaptive Path Generation of Cobot

Shirine El Zaatari, Yuqi Wang, Weidong Li, Yiqun Peng

Wuhan University of Technology

Research output: Contribution to journal › Article › peer-review

24 Citations (Scopus)

357 Downloads (Pure)

Abstract

An approach of Task-Parameterised Learning from Demonstration (TP-LfD) aims at automatically adapting the movements of collaborative robots (cobots) to new settings using knowledge learnt from demonstrated paths. The approach is suitable for encoding complex relations between a cobot and its surrounding, i.e., task-relevant objects. However, further efforts are still required to enhance the intelligence and adaptability of TP-LfD for dynamic tasks. With this aim, this paper presents an improved TP-LfD (iTP-LfD) approach to program cobots adaptively for a variety of industrial tasks. iTP-LfD comprises of three main improvements over other developed TP-LfD approaches: 1) detecting generic visual features for frames of reference (frames) in demonstrations for path reproduction in new settings without using complex computer vision algorithms, 2) minimising redundant frames that belong to the same object in demonstrations using a statistical algorithm, and 3) designing a reinforcement learning algorithm to eliminate irrelevant frames. The distinguishing characteristic of the iTP-LfD approach is that optimal frames are identified from demonstrations by simplifying computational complexity, overcoming occlusions in new settings, and boosting the overall performance. Case studies for a variety of industrial tasks involving different objects and scenarios highlight the adaptability and robustness of the iTP-LfD approach.

Original language	English
Article number	102109
Number of pages	18
Journal	Robotics and Computer-Integrated Manufacturing
Volume	69
Early online date	17 Dec 2020
DOIs	https://doi.org/10.1016/j.rcim.2020.102109
Publication status	Published - Jun 2021

Bibliographical note

NOTICE: this is the author’s version of a work that was accepted for publication in Robotics and Computer-Integrated Manufacturing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Robotics and Computer-Integrated Manufacturing, 69, (2021) DOI: 10.1016/j.rcim.2020.102109

© 2020, Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/10.1016/j.rcim.2020.102109

Funder

This work is funded by the PhD studentships from the Coventry University , the Unipart Powertrain Application Ltd. (U.K.) , the Institute of Digital Engineering, U.K. , and a research project sponsored by the National Natural Science Foundation of China (Project No. 51975444 ).

Funding

This work is funded by the PhD studentships from the Coventry University , the Unipart Powertrain Application Ltd. (U.K.) , the Institute of Digital Engineering, U.K. , and a research project sponsored by the National Natural Science Foundation of China (Project No. 51975444 ).

Keywords

Collaborative robots (Cobots)
Intuitive programming
Learning from demonstration
Reinforcement learning

ASJC Scopus subject areas

Control and Systems Engineering
Software
General Mathematics
Computer Science Applications
Industrial and Manufacturing Engineering

Access to Document

10.1016/j.rcim.2020.102109

Post-PrintAccepted author manuscript, 2.38 MBLicence: CC BY-NC-ND

Cite this

@article{13d7b326fd7543e3877dc20a6affe175,

title = "iTP-LfD: Improved Task Parametrised Learning from Demonstration for Adaptive Path Generation of Cobot",

abstract = "An approach of Task-Parameterised Learning from Demonstration (TP-LfD) aims at automatically adapting the movements of collaborative robots (cobots) to new settings using knowledge learnt from demonstrated paths. The approach is suitable for encoding complex relations between a cobot and its surrounding, i.e., task-relevant objects. However, further efforts are still required to enhance the intelligence and adaptability of TP-LfD for dynamic tasks. With this aim, this paper presents an improved TP-LfD (iTP-LfD) approach to program cobots adaptively for a variety of industrial tasks. iTP-LfD comprises of three main improvements over other developed TP-LfD approaches: 1) detecting generic visual features for frames of reference (frames) in demonstrations for path reproduction in new settings without using complex computer vision algorithms, 2) minimising redundant frames that belong to the same object in demonstrations using a statistical algorithm, and 3) designing a reinforcement learning algorithm to eliminate irrelevant frames. The distinguishing characteristic of the iTP-LfD approach is that optimal frames are identified from demonstrations by simplifying computational complexity, overcoming occlusions in new settings, and boosting the overall performance. Case studies for a variety of industrial tasks involving different objects and scenarios highlight the adaptability and robustness of the iTP-LfD approach.",

keywords = "Collaborative robots (Cobots), Intuitive programming, Learning from demonstration, Reinforcement learning",

author = "\{El Zaatari\}, Shirine and Yuqi Wang and Weidong Li and Yiqun Peng",

note = "NOTICE: this is the author{\textquoteright}s version of a work that was accepted for publication in Robotics and Computer-Integrated Manufacturing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Robotics and Computer-Integrated Manufacturing, 69, (2021) DOI: 10.1016/j.rcim.2020.102109 {\textcopyright} 2020, Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/10.1016/j.rcim.2020.102109 ",

year = "2021",

month = jun,

doi = "10.1016/j.rcim.2020.102109",

language = "English",

volume = "69",

journal = "Robotics and Computer-Integrated Manufacturing",

issn = "0736-5845",

publisher = "Elsevier",

}

TY - JOUR

T1 - iTP-LfD: Improved Task Parametrised Learning from Demonstration for Adaptive Path Generation of Cobot

AU - El Zaatari, Shirine

AU - Wang, Yuqi

AU - Li, Weidong

AU - Peng, Yiqun

N1 - NOTICE: this is the author’s version of a work that was accepted for publication in Robotics and Computer-Integrated Manufacturing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Robotics and Computer-Integrated Manufacturing, 69, (2021) DOI: 10.1016/j.rcim.2020.102109 © 2020, Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/10.1016/j.rcim.2020.102109

PY - 2021/6

Y1 - 2021/6

N2 - An approach of Task-Parameterised Learning from Demonstration (TP-LfD) aims at automatically adapting the movements of collaborative robots (cobots) to new settings using knowledge learnt from demonstrated paths. The approach is suitable for encoding complex relations between a cobot and its surrounding, i.e., task-relevant objects. However, further efforts are still required to enhance the intelligence and adaptability of TP-LfD for dynamic tasks. With this aim, this paper presents an improved TP-LfD (iTP-LfD) approach to program cobots adaptively for a variety of industrial tasks. iTP-LfD comprises of three main improvements over other developed TP-LfD approaches: 1) detecting generic visual features for frames of reference (frames) in demonstrations for path reproduction in new settings without using complex computer vision algorithms, 2) minimising redundant frames that belong to the same object in demonstrations using a statistical algorithm, and 3) designing a reinforcement learning algorithm to eliminate irrelevant frames. The distinguishing characteristic of the iTP-LfD approach is that optimal frames are identified from demonstrations by simplifying computational complexity, overcoming occlusions in new settings, and boosting the overall performance. Case studies for a variety of industrial tasks involving different objects and scenarios highlight the adaptability and robustness of the iTP-LfD approach.

AB - An approach of Task-Parameterised Learning from Demonstration (TP-LfD) aims at automatically adapting the movements of collaborative robots (cobots) to new settings using knowledge learnt from demonstrated paths. The approach is suitable for encoding complex relations between a cobot and its surrounding, i.e., task-relevant objects. However, further efforts are still required to enhance the intelligence and adaptability of TP-LfD for dynamic tasks. With this aim, this paper presents an improved TP-LfD (iTP-LfD) approach to program cobots adaptively for a variety of industrial tasks. iTP-LfD comprises of three main improvements over other developed TP-LfD approaches: 1) detecting generic visual features for frames of reference (frames) in demonstrations for path reproduction in new settings without using complex computer vision algorithms, 2) minimising redundant frames that belong to the same object in demonstrations using a statistical algorithm, and 3) designing a reinforcement learning algorithm to eliminate irrelevant frames. The distinguishing characteristic of the iTP-LfD approach is that optimal frames are identified from demonstrations by simplifying computational complexity, overcoming occlusions in new settings, and boosting the overall performance. Case studies for a variety of industrial tasks involving different objects and scenarios highlight the adaptability and robustness of the iTP-LfD approach.

KW - Collaborative robots (Cobots)

KW - Intuitive programming

KW - Learning from demonstration

KW - Reinforcement learning

UR - https://www.scopus.com/pages/publications/85097882614

U2 - 10.1016/j.rcim.2020.102109

DO - 10.1016/j.rcim.2020.102109

M3 - Article

SN - 0736-5845

VL - 69

JO - Robotics and Computer-Integrated Manufacturing

JF - Robotics and Computer-Integrated Manufacturing

M1 - 102109

ER -

iTP-LfD: Improved Task Parametrised Learning from Demonstration for Adaptive Path Generation of Cobot

Abstract

Bibliographical note

Funder

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this