Precision engineering has been, and continues to be, one of the disciplines needed to enable future technological progress, especially in the area of manufacturing. Demands for increasing precision have also evolved in almost all other industrial sectors; ranging from planes, trains and automobiles, to printers (printing electronics and optical surfaces), scientific and analytical instruments (microscopes and telescopes to particle accelerators), medical and surgical tools, and traditional and renewable power generation. At the root of all of these technologies are increasingly advanced machines and controls, and in manufacturing, advanced products. Metrology plays a key role in precision engineering, allowing the degree of precision and accuracy to be quantified and used in later processes. Modern digital manufacturing (a.k.a. Industry 4.0) requires the creation, manipulation and sharing of large amounts of data. But, the adoption of digital technology in manufacturing processes is currently hindered by lack of efficiency and confidence in data that is captured within those processes. Confidence in data is the key enabler for adoption of the Industry 4.0 methodologies. Through traceability, metrology is one of the pillars for demonstrating confidence in data, without which, industry suffers from unnecessary waste, inefficient processes and increased costs for quality. A significant issue with the adoption of digital manufacturing is the vast amount of data that can be produced with new measurement technologies. But, this data-rich issue can be an opportunity if advanced data handling, analysis and learning methods can be developed and employed. These issues are ideal for machine learning (ML), which is only now being utilised for measurement applications, to enhance the capability and performance of instruments, e.g. understand surface orientations, automatically segment 3D point clouds, infer surface information from missing data using a priori information and automatically segment objects, especially for machine vision applications.
Many of the speed and data challenges that plague metrology for digital manufacturing could be overcome by combining ML approaches with precision engineering/metrology. The special issue will investigate where machine learning can be applied in the following areas:
- Coordinate metrology
- ML in optical metrology
- ML for quality control and in-process metrology
- ML for surface metrology
- ML for X-ray computed tomography
- ML for critical dimension and overlay
- Sampling and interpolation in surface measurement
- Ultra-fast surface defect and feature recognition
- Uncertainty methods with ML measurement models
- Monte Carlo simulation using ML methods
Guest Editors:
Prof. Richard Leach
University of Nottingham
Prof. Richard Leach
University of Nottingham
Dr. Samanta Piano
University of Nottingham
University of Nottingham
Dr. Benjamin Haefner
Karlsruhe Institute of Technology
Karlsruhe Institute of Technology
Prof. Robert Schmitt
RWTH Aachen and Fraunhofer IPT
RWTH Aachen and Fraunhofer IPT
Prof . Bianca Colosimo
Polytechnico Milano
Polytechnico Milano
Submission Deadline: February 2021
Acceptance Deadline: April 2021
Acceptance Deadline: April 2021
Please submit your paper here -
No comments:
Post a Comment