Kriging-based surface error measurement method with human-factor resilience using articulated arm coordinate measuring machine

Abstract

Articulated arm coordinate measuring machine (AACMM) has been widely utilized in complex surface measurements due to its flexibility and portability. However, the handheld operation of AACMM makes it highly susceptible to human factors, particularly in the measurement of thin-walled surfaces deformation, often resulting in significant measurement errors. This paper proposes a Kriging-based surface error measurement method to address the impact of human factors on the measurement accuracy of AACMM. By employing a surface profile deviation calculation method based on maximum deviation, the deformation of thin-walled surfaces is quantitatively evaluated. The Hammersley method is used to generate a small number of sample points to construct a Kriging model, which predicts deformation errors and their uncertainty distribution. Two sampling rules are designed: Rule 1 selects points with the highest uncertainty in predicted deformation errors to avoid local convergence; Rule 2 targets extreme points of maximum uncertainty in the predicted error distribution. This approach enables focused sampling in areas with larger deformation errors, thereby improving both measurement accuracy and efficiency. Experimental results demonstrate that the proposed method significantly enhances the reliability of thin-walled surfaces deformation measurements and exhibits strong resilience to coordinate deviations caused by handheld operations, providing an effective solution for practical applications.