The statistical gauge in geometrical verification

Abstract

International audienceCurrently, most measurement and geometrical verification processes are based on the characterization of distances or angles between geometrical elements. In a previous paper, a method was developed for the verification of ISO specifications using the statistical information included in a set of points. This method is based on the development of measurement sequences and hence a sequence of geometrical constructions. The main problem in this verification process is to propagate uncertainties. With a correct expression of a virtual gauge, it is possible to avoid this propagation; indeed by using virtual gauge manipulation, initial best-fitted surfaces are never touched. The aim of this paper is to present a new approach using this statistical information to describe the measured part and develop a virtual gauge system for verification. This will be done in two parts. The first part will focus on the statistical modelling of the measured surface; the second part will describe a verification method based on a virtual, i.e. numerical, gauge

A task specific uncertainty analysis method for least-squares-based form characterization of ultra-precision freeform surfaces

“…Current research on the analysis of the uncertainty in data processing in geometric measurement is still focused on simple geometry such as circle, sphere and cylinder [6,17,18], and relatively little research work has been conducted on freeform surfaces. Some researchers attempt to represent the freeform surfaces by assembling simple geometry such as sphere and cylinder so that the problem is converted to analyze each assembled simple geometry [19].…”