The University of Sheffield's Advanced Manufacturing Research Centre (AMRC) has been busy working on projects in the fight against the coronavirus pandemic. As a member of the UK High Value Manufacturing Catapult, the AMRC is part of the industrial consortium that is helping deliver more than 10,000 ventilators to NHS doctors and nurses. In addition, its engineers are 3D printing 1,000 face masks a week for NHS staff at local hospitals. Greg Blackman speaks to Kieran Edge, technical lead for machine vision at AMRC's Integrated Manufacturing Group, about new vision projects and the presentation he is to give for UKIVA's vision technology hub, to be broadcast on 14 May.
What is AMRC working on during Covid-19 lockdown?
A lot of our partners – AMRC has 125 industrial partners including aerospace giants Boeing, Rolls-Royce, BAE Systems and Airbus – have eased up on the production side of their businesses, and are focusing more on R&D at the moment. We've had a lot of enquiries recently, more than we normally get. We're undertaking projects that we can either do remotely or that we can start remotely – software development, for example.
There are areas where we're seeing more interest in machine vision. A lot of firms are looking at how they will start up production again, by using streamlined working methods, and with fewer staff in any location at a given time. Some processes can be automated using machine vision to help with social distancing restrictions.
There are lots of parts of the manufacturing sector that haven't been badly hit yet by the lockdown, and they are still very much interested in how to reduce overheads using automation. Aerospace is still looking at machine vision for inspection. We're working with one of our partners to investigate inspection of sheet metal in aerospace to improve quality and reduce the cycle time for production.
We're also getting a lot of interest, in many different applications, because operators are taking this time to reflect on their current processes – it's the ideal time for them to think about this.
AMRC has got a lot on with the Ventilator Challenge, but is not using machine vision. It has used augmented reality to give engineers instructions, overlaid via AR headsets, on how to adapt and optimise production lines for making ventilators. AMRC's Design and Prototyping Group are making face masks, coordinating its efforts with Sheffield Teaching Hospitals NHS Foundation Trust. The group worked in shifts over the Easter weekend, with all the staff, from receptionists, composite engineers and senior project managers, volunteering to help with assembly.
What will you discuss in your UKIVA hub presentation?
I'll be talking about the state of machine vision in high-value manufacturing. One of the things I'll discuss is global supply chains and the impact machine vision could have on these. Global supply chains in aerospace are really quite long – from Europe to India and China – and there's interest at the moment in how to shorten them. When they're that long, the chain is only as strong as its weakest link.
AMRC is looking at ways to use machine vision to automate and reduce the cost of parts, so they can be manufactured in fewer locations and keep supply chains compact. Airbus send parts all over Europe; wings, for example, are assembled in North Wales and shipped to France.
AMRC is also working on identifying technologies that apply across sectors. One area is maintenance, which is big in aerospace, but it is also big in the rail industry. Vision systems play a major role in the maintenance of aircraft and trains. We're looking at different techniques, whether it's artificial intelligence for inspecting similar features on different components – bolts look the same whether they're on a train or a fighter jet, for instance. There's also CAD-based analysis, overlaying CAD files onto 3D images. All this work is to develop adaptive vision systems for different types of maintenance activities.
Polarisation imaging will also be part of your talk. How are you using it?
Polarisation imaging can give rapid information for inspecting components that have been difficult to inspect in the past, such as carbon fibre sheets and medical joints, as well as shiny surfaces and specular curve reflective surfaces. We think polarisation imaging will have a lot of applications in the future.
The two main areas where we're using polarisation imaging at the moment are defect detection on medical components and on composites such as carbon fibre weave.
We're making big strides in this area; we've currently undertaken a package of work at the AMRC between the integrated manufacturing group, which focuses on digital inspection and AI, and the composites group. We'll be working with the composites group to look at how to integrate polarisation inspection directly into a carbon fibre weaving loom.
Polarisation imaging can inspect carbon fibre weave in real time at 30fps. We can make the inspection in-line with production of the fibre as it is being woven – the technique isn't restricted to inspecting the finished product.
When unpolarised light falls on carbon fibre it becomes polarised. Therefore, a vision system can pick out individual strands in the weave based on the angle of polarisation of the light reflecting from the strands. That allows the system to inspect the strands of the carbon fibre weave directly.
The system designed for inspecting medical components measures the intensity of polarisation reflected from the surface. That's using different light sources. The intensity of polarised light gives information about surface smoothness – the more polarised light reflected, the smoother the surface; it's a good indicator of surface defects.
The carbon fibre inspection work is more mature than medical component inspection – we've got a development route for composite inspection, to integrate the technique into a system; we're currently looking for partners and funding opportunities for the medical component inspection.
UKIVA's Technology Presentation Hub, including Kieran Edge's talk, will go live on 14 May. It's free to register.
Further reading
The sky’s the limit - Greg Blackman visits the University of Sheffield’s Factory 2050 where Rolls-Royce, McLaren and Siemens, among others, are investing in research on digital manufacturing
Get in touch
If you have experience working on some of the topics Kieran covers - aerospace inspection or polarisation imaging - please get in touch to ask about writing for us: greg.blackman@europascience.com
