Data-driven process characterization and adaptive control in robotic arc welding

Peng Wang; Joseph Kershaw; Matthew Russell; Jianjing Zhang; Yuming Zhang; Robert X. Gao

doi:10.1016/j.cirp.2022.04.046

Data-driven process characterization and adaptive control in robotic arc welding

Peng Wang, Joseph Kershaw, Matthew Russell, Jianjing Zhang, Yuming Zhang, Robert X. Gao

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

4 Scopus citations

Abstract

Robotic arc welding (RAW) has been an essential process in various assembly systems, such as automotive manufacturing. However, its implementations lack adaptivity to compensate for process variations. This paper presents a data-driven process characterization and online adaptive control framework for RAW to automatically and efficiently achieve desired weld pool condition, given any initial conditions. Based on optical imaging, pool width is characterized through a pixel-level image segmentation network and then used for determining the parameter adjustment for robotic execution through a gradient-based controller. Experiments demonstrate quick process convergence within 7 adjustment periods and an error band within 10.9%.

Original language	English
Pages (from-to)	45-48
Number of pages	4
Journal	CIRP Annals
Volume	71
Issue number	1
DOIs	https://doi.org/10.1016/j.cirp.2022.04.046
State	Published - Jan 2022

Bibliographical note

Publisher Copyright:
© 2022 CIRP

Funding

This work is supported by the National Science Foundation under grants CMMI-2024614 and CMMI-1830295 .

Funders	Funder number
National Science Foundation Arctic Social Science Program	CMMI-2024614, CMMI-1830295

Keywords

Adaptive control
Robot
Welding

ASJC Scopus subject areas

Mechanical Engineering
Industrial and Manufacturing Engineering

Access to Document

10.1016/j.cirp.2022.04.046

Cite this

@article{a5812dc6b493454e8c10504275133d43,

title = "Data-driven process characterization and adaptive control in robotic arc welding",

abstract = "Robotic arc welding (RAW) has been an essential process in various assembly systems, such as automotive manufacturing. However, its implementations lack adaptivity to compensate for process variations. This paper presents a data-driven process characterization and online adaptive control framework for RAW to automatically and efficiently achieve desired weld pool condition, given any initial conditions. Based on optical imaging, pool width is characterized through a pixel-level image segmentation network and then used for determining the parameter adjustment for robotic execution through a gradient-based controller. Experiments demonstrate quick process convergence within 7 adjustment periods and an error band within 10.9%.",

keywords = "Adaptive control, Robot, Welding",

author = "Peng Wang and Joseph Kershaw and Matthew Russell and Jianjing Zhang and Yuming Zhang and Gao, {Robert X.}",

note = "Publisher Copyright: {\textcopyright} 2022 CIRP",

year = "2022",

month = jan,

doi = "10.1016/j.cirp.2022.04.046",

language = "English",

volume = "71",

pages = "45--48",

number = "1",

}

TY - JOUR

T1 - Data-driven process characterization and adaptive control in robotic arc welding

AU - Wang, Peng

AU - Kershaw, Joseph

AU - Russell, Matthew

AU - Zhang, Jianjing

AU - Zhang, Yuming

AU - Gao, Robert X.

PY - 2022/1

Y1 - 2022/1

N2 - Robotic arc welding (RAW) has been an essential process in various assembly systems, such as automotive manufacturing. However, its implementations lack adaptivity to compensate for process variations. This paper presents a data-driven process characterization and online adaptive control framework for RAW to automatically and efficiently achieve desired weld pool condition, given any initial conditions. Based on optical imaging, pool width is characterized through a pixel-level image segmentation network and then used for determining the parameter adjustment for robotic execution through a gradient-based controller. Experiments demonstrate quick process convergence within 7 adjustment periods and an error band within 10.9%.

AB - Robotic arc welding (RAW) has been an essential process in various assembly systems, such as automotive manufacturing. However, its implementations lack adaptivity to compensate for process variations. This paper presents a data-driven process characterization and online adaptive control framework for RAW to automatically and efficiently achieve desired weld pool condition, given any initial conditions. Based on optical imaging, pool width is characterized through a pixel-level image segmentation network and then used for determining the parameter adjustment for robotic execution through a gradient-based controller. Experiments demonstrate quick process convergence within 7 adjustment periods and an error band within 10.9%.

KW - Adaptive control

KW - Robot

KW - Welding

UR - http://www.scopus.com/inward/record.url?scp=85130463742&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85130463742&partnerID=8YFLogxK

U2 - 10.1016/j.cirp.2022.04.046

DO - 10.1016/j.cirp.2022.04.046

M3 - Article

AN - SCOPUS:85130463742

SN - 0007-8506

VL - 71

SP - 45

EP - 48

JO - CIRP Annals

JF - CIRP Annals

IS - 1

ER -

Data-driven process characterization and adaptive control in robotic arc welding

Abstract

Bibliographical note

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this