Loading [a11y]/accessibility-menu.js
Neural Process Enhanced Machining Error Evaluation for Coordinate Measuring Machines | IEEE Journals & Magazine | IEEE Xplore

Neural Process Enhanced Machining Error Evaluation for Coordinate Measuring Machines


Abstract:

Coordinate measuring machines (CMMs) that provide ultraprecise results are widely used in precision manufacturing. However, the traditional point-by-point measuring appro...Show More

Abstract:

Coordinate measuring machines (CMMs) that provide ultraprecise results are widely used in precision manufacturing. However, the traditional point-by-point measuring approaches are time-consuming, limiting their applications in modern manufacturing, especially for large-scale surface measurements and online measurements. In this article, we propose an efficient machining error evaluation approach for complex surfaces based on the neural process. Without sacrificing accuracy, the proposed method improves the efficiency of measurement by sparse sampling and reconstruction. Using the neural process, the reconstruction algorithm models the machining error accurately and robustly. Besides, by estimating the reconstruction uncertainty, we further improve the sampling efficiency by introducing adaptive sampling. Intensive experiments show that the proposed method can achieve accurate reconstruction with high efficiency compared with the existing approaches.
Article Sequence Number: 2514512
Date of Publication: 25 July 2022

ISSN Information:

Funding Agency:


References

References is not available for this document.