Uncertainty Analysis for Coordinate Estimation Using Discrete Measurement Data

With increasing emphasis on the accuracy of dimensional inspection in coordinate metrology, it becomes necessary to perform uncertainty analysis for coordinate estimation using discrete measurement data. This project is designed to study the uncertainty of coordinate estimation using discrete measurement data. In this study, the geometric error on a machined surface is characterized, and techniques for decomposing the different geometric error components from measurement data are developed. Based on the results of geometric error decomposition, the uncertainty of coordinate estimation is classified into the Type I and the Type II problems. In the analysis of the Type I problem, quantitative measures are derived for the prediction of coordinate variation. In the analysis of the Type II problem, the uncertainty of coordinate transformation obtained from two different coordinate estimation approaches are investigated and compared.