Least-squares fit of measured points for square line-profiles

J. Davidson, S. Savaliya, Jami J. Shah

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

2 Scopus citations

Abstract

The pseudoinverse of a rectangular matrix is used to compute the least-squares fit of a set of points that have been measured along a line-profile. Tolerances on line profiles are used to control cross-sectional shapes of parts, such as turbine blades. The specified profile is treated as a moving platform of a hypothetical, redundant, and planar in-parallel-actuated robot, and all the measured points are presumed to be fixed in it. The locations of the linear actuators are represented with screw (torsor) coordinates, and these are arranged in a matrix equation that relates the three small displacements of the platform to the corresponding deviations (treated as small displacements) of the measured points. The Moore-Penrose (pseudoinverse) solution uniquely produces displacements of the platform which correspond to the least-squares minimum for the deviations at all of the measured points.

Original languageEnglish (US)
Title of host publicationProcedia CIRP
Pages203-210
Number of pages8
Volume10
DOIs
StatePublished - 2013
Event12th CIRP Conference on Computer Aided Tolerancing, CAT 2012 - Huddersfield, United Kingdom
Duration: Apr 18 2012Apr 19 2012

Other

Other12th CIRP Conference on Computer Aided Tolerancing, CAT 2012
Country/TerritoryUnited Kingdom
CityHuddersfield
Period4/18/124/19/12

Keywords

  • CMM data reduction
  • Fit
  • Least-squares
  • Point-cloud
  • Profile
  • Pseudoinverse
  • Regression

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Control and Systems Engineering

Fingerprint

Dive into the research topics of 'Least-squares fit of measured points for square line-profiles'. Together they form a unique fingerprint.

Cite this