A Library for Constraint Consistent Learning

Jeevan Manavalan; Yuchen Zhao; Prabhakar Ray; Hsiu Chin Lin; Matthew Howard

doi:10.1080/01691864.2020.1786723

A Library for Constraint Consistent Learning

Jeevan Manavalan, Yuchen Zhao, Prabhakar Ray, Hsiu Chin Lin, Matthew Howard

McGill University

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

8033 Downloads (Pure)

Abstract

This paper introduces the first, open source software library for Constraint Consistent Learning (CCL). It implements a family of data-driven methods that are capable of (i) learning state-independent and -dependent constraints, (ii) decomposing the behaviour of redundant systems into task- and null-space parts, and (iii) uncovering the underlying null space control policy. It is a tool to analyse and decompose many everyday tasks, such as wiping, reaching and drawing. The library also includes several tutorials that demonstrate its use with both simulated and real world data in a systematic way. This paper documents the methods contained within the library, including the implementations of said methods in tutorials and associated helper methods. The software is made freely available to the community, to enable code reuse and allow users to gain in-depth experience in statistical learning in this area.

Original language	English
Pages (from-to)	1-13
Number of pages	13
Journal	Advanced Robotics
Early online date	7 Jul 2020
DOIs	https://doi.org/10.1080/01691864.2020.1786723
Publication status	E-pub ahead of print - 7 Jul 2020

Keywords

Software Library
Constraints
Learning from Demonstration
Model Learning
learning from demonstration
Software library
constraints
model learning

Access to Document

10.1080/01691864.2020.1786723

A Library for Constraint_MANAVALAN_Accepted8June2020_GOLD AAMAccepted author manuscript, 1.59 MB

Cite this

@article{8ca851818e0e4f8bade385d27ac36743,

title = "A Library for Constraint Consistent Learning",

abstract = "This paper introduces the first, open source software library for Constraint Consistent Learning (CCL). It implements a family of data-driven methods that are capable of (i) learning state-independent and -dependent constraints, (ii) decomposing the behaviour of redundant systems into task- and null-space parts, and (iii) uncovering the underlying null space control policy. It is a tool to analyse and decompose many everyday tasks, such as wiping, reaching and drawing. The library also includes several tutorials that demonstrate its use with both simulated and real world data in a systematic way. This paper documents the methods contained within the library, including the implementations of said methods in tutorials and associated helper methods. The software is made freely available to the community, to enable code reuse and allow users to gain in-depth experience in statistical learning in this area.",

keywords = "Software Library, Constraints, Learning from Demonstration, Model Learning, learning from demonstration, Software library, constraints, model learning",

author = "Jeevan Manavalan and Yuchen Zhao and Prabhakar Ray and Lin, {Hsiu Chin} and Matthew Howard",

year = "2020",

month = jul,

day = "7",

doi = "10.1080/01691864.2020.1786723",

language = "English",

pages = "1--13",

journal = "Advanced Robotics",

issn = "0169-1864",

publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - A Library for Constraint Consistent Learning

AU - Manavalan, Jeevan

AU - Zhao, Yuchen

AU - Ray, Prabhakar

AU - Lin, Hsiu Chin

AU - Howard, Matthew

PY - 2020/7/7

Y1 - 2020/7/7

N2 - This paper introduces the first, open source software library for Constraint Consistent Learning (CCL). It implements a family of data-driven methods that are capable of (i) learning state-independent and -dependent constraints, (ii) decomposing the behaviour of redundant systems into task- and null-space parts, and (iii) uncovering the underlying null space control policy. It is a tool to analyse and decompose many everyday tasks, such as wiping, reaching and drawing. The library also includes several tutorials that demonstrate its use with both simulated and real world data in a systematic way. This paper documents the methods contained within the library, including the implementations of said methods in tutorials and associated helper methods. The software is made freely available to the community, to enable code reuse and allow users to gain in-depth experience in statistical learning in this area.

AB - This paper introduces the first, open source software library for Constraint Consistent Learning (CCL). It implements a family of data-driven methods that are capable of (i) learning state-independent and -dependent constraints, (ii) decomposing the behaviour of redundant systems into task- and null-space parts, and (iii) uncovering the underlying null space control policy. It is a tool to analyse and decompose many everyday tasks, such as wiping, reaching and drawing. The library also includes several tutorials that demonstrate its use with both simulated and real world data in a systematic way. This paper documents the methods contained within the library, including the implementations of said methods in tutorials and associated helper methods. The software is made freely available to the community, to enable code reuse and allow users to gain in-depth experience in statistical learning in this area.

KW - Software Library

KW - Constraints

KW - Learning from Demonstration

KW - Model Learning

KW - learning from demonstration

KW - Software library

KW - constraints

KW - model learning

UR - http://www.scopus.com/inward/record.url?scp=85087664568&partnerID=8YFLogxK

U2 - 10.1080/01691864.2020.1786723

DO - 10.1080/01691864.2020.1786723

M3 - Article

SN - 0169-1864

SP - 1

EP - 13

JO - Advanced Robotics

JF - Advanced Robotics

ER -

A Library for Constraint Consistent Learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this