FEATURE:AEROSPACE >
configuration that facilitates satellite rotating measurement. Specific metrology applications in place
across aerospace include identification and classification of airfoil blades, blisks (bladed disks) and other blades; chatter measurement for possible excess vibration in those blades; and inspection of cooling holes.
Quality assurance For quality assurance, Lockheed Martin UK uses OCR from High QA to digitally interpret previously vectorised part drawings. Measurement data calculated by the FARO articulated arm is generated into a report, which can then be distributed to the rest of the engineering team. Garry Ellingham, project chief engineer
at Lockheed Martin UK, Ampthill, says: “From a metallics perspective, we have a radiography capability here for small scale phased array ultrasonic testing, in the lab. But we do work with some third party suppliers around computed tomography. “This provides us with an
understanding of 3D representations, displaying how these structures look internally – porosity being the main aspect to examine, for the metallic parts.” Working with third parties, a full matrix
capture featuring multiple carbon-fibre reinforced polymer (CFRP) elements helps to optimise non-destructive testing. Acquired machine vision data determines where these materials would fit within composite parts. When it comes to working with low- quantity, high-value aerospace products, Ellingham explained the need to refrain from lifting AI projects off for the sake of hype.
He says: “This can be a pitfall where
you want to apply a new scanning technology, and replace something you think isn’t doing your job – but it is. It goes back to that quality of inspection; we have focused on investing time in properly understanding the requirements of these imaging technologies.” Will Cullen, head of quality at Lockheed
Martin UK, Ampthill, says: “Across all business areas and functions, from the MD through and down the root and branch, people aim to work together. We’re not the old school quality police; it’s all about collaboration.”
Future trends Looking to the future, computer vision and imaging AI is set to become ever more customisable, with fewer steps being required to complete specific functions. Deeper insights will become more accessible as a result of fusing data
FARO articulated arms contain a point probe and scanners that measure down to 50 microns, to ensure measurement compliance of all aerospace vehicle parts
assets from multiple sources including 3D scanners, thermal, multi-spectral, and X-ray imaging. Going forward at Lockheed Martin UK, for example, Windchill model-based system engineering (MBSE), provided by PTC, is being tried out on screens, to design and review parts virtually with automation, before physical manufacturing begins. This helps to minimise errors in production prior to distribution to customers. It is predicted that this will be rolled out across the Lockheed Martin UK business by the end of the year. Lopez says: “I think we’ll see more
frequent and affordable 3D inspection use cases. We’re looking at ways to reduce
30 IMAGING AND MACHINE VISION EUROPE AUGUST/SEPTEMBER 2024
complexity of 3D development toolsets and physical hardware, by leveraging mass production as we scale.” From here, AI is predicted by Lopez to become unified with 2D and 3D inspection capabilities in one environment, allowing engineers and quality assurance professionals to “compare rules-based and AI techniques and approaches, to see which is getting the best results”. According to Zohar Kantor, who is
the Chief Revenue Officer at AI quality inspection provider QualiSense, “the next frontier is using AI to automatically identify recurring issues or patterns in inspection data. “This necessitates a holistic, strategic
view of the entire production process, not just defect detection at individual stages. A single anomaly might hold less meaning than a pattern across time, especially if linked to data from other manufacturing sensors.” Tom Allen, founder and CEO of The AI Journal, believes that “we’ll likely see more embedded machine vision capabilities built directly into manufacturing equipment, robots, drones and autonomous systems to enable real- time inspection and monitoring directly at the source.” Allen concludes: “There’s a massive
opportunity for contracts with businesses in the arena of airports and engine manufacturers because of how many parts go into it from health and safety, throughput, production time, and less recalls due to better quality control.” i
Lockheed Martin UK, Ampthill
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39
