Precise Detection of Pose and Assembly Error Based on Partial Scan Data for Fabrication of Tubular Joints

Yan Zhi Tan, Qiang Fang, Chee Khiang Pang, Abdullah Al-Mamun, Fook Seng Wong, Chee Meng Chew

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Automated and accurate determination of the pose of a roughly placed workpiece is essential for fabrication tasks such as robotic welding. In this paper, an algorithm based on cylinder fitting and Iterative Closest Point (ICP) alignment is proposed for detecting both the pose and assembly error of tubular joints in a T-, K-, or Y-shaped configuration using partial scan data. A typical TKY joint consists of a brace and a chord component. In order to also compute the assembly error, the poses of the individual joint components are determined separately. The pose information is obtained by aligning scan data points with ideal CAD models using cylinder fitting and ICP matching techniques. Due to uniformity of the chord surface, assembly error due to a translation along, or a rotation about the chord longitudinal axis are not determined. The assembly error obtained can be used to determine whether realignment is required, or combined with the pose information to re-plan paths for subsequent tasks. The effectiveness of the proposed algorithm is verified with simulation studies and application to an actual robotic welding system to locate a tubular workpiece.

Original languageEnglish
Title of host publicationICARM 2018 - 2018 3rd International Conference on Advanced Robotics and Mechatronics
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages799-804
Number of pages6
ISBN (Electronic)9781538670668
DOIs
Publication statusPublished - Jan 11 2019
Externally publishedYes
Event3rd IEEE International Conference on Advanced Robotics and Mechatronics, ICARM 2018 - Singapore, Singapore
Duration: Jul 18 2018Jul 20 2018

Publication series

NameICARM 2018 - 2018 3rd International Conference on Advanced Robotics and Mechatronics

Conference

Conference3rd IEEE International Conference on Advanced Robotics and Mechatronics, ICARM 2018
Country/TerritorySingapore
CitySingapore
Period7/18/187/20/18

ASJC Scopus subject areas

  • Artificial Intelligence
  • Mechanical Engineering
  • Control and Optimization

Fingerprint

Dive into the research topics of 'Precise Detection of Pose and Assembly Error Based on Partial Scan Data for Fabrication of Tubular Joints'. Together they form a unique fingerprint.

Cite this