By David Tillett
While it has a long history supplying products to manufacturers, Zebra Technologies Corp. is new to the field of fixed scanning and industrial machine vision – but it’s made a bold entrance.
In May the Lincolnshire, Ill., company announced its arrival in the market with a portfolio of fixed industrial scanners (FIS), machine vision smart cameras and a software platform that can run both.
“Traditionally … machine vision has got this sort of like mystic, magic feeling about it like it’s something very complex [that] only very few experts can manage,” said Donato Montanari, VP of machine vision solutions and GM at Zebra. He said that simplicity is one of the main advantages his company’s products offer its customers.
“If you go back … 10, 15 years ago, providers of machine vision would not provide FIS devices and vice versa. … We’re using the same camera, the same software … so for us coming in as a new player has been easier, because we didn’t have any legacy to defend,” he said. The majority of its scanner and camera hardware can be used interchangeably for scanning and machine vision tasks, and its Aurora software platform has an intuitive, user-friendly interface.
He said plastics processors — especially injection molders — would be a good example of potential users of these technologies. “If you look into injection molding … the level of accuracy, and the complexity of these molds— it’s insane. And therefore, on the one hand there is a very strong need for quality control. And on the other hand — especially in the medical field, there is also [a] very strong need for track-and-trace. So, you need to know from which lot it comes, where it goes, supply chain and all of that. And again, historically, these two needs are very different, and they’re tackled by different technologies and different software packages.”
The ability to use Zebra’s Aurora software for both scanning and machine vision tasks makes it easier for manufacturers to start using either technology, and to expand their use over time. As an example, customers might start off using a fixed industrial scanner that reads barcodes on products that pass by on a conveyor belt. This task would allow them to get acquainted with Zebra’s Aurora software. As their needs develop, they could add simple inspection tasks — for example, the presence/absence of parts — using the same hardware and simply upgrading the software license.
As the complexity of the inspection tasks increases, users can add or upgrade the sophistication of the camera and/or lighting, using the same software interface — at every stage, only incremental changes would be needed to add capabilities to a line they’ve become familiar with. As an added bonus, the flexibility of the sensors and cameras means fewer spare parts are needed to keep a line running.
“So, we’re trying really to make it very much accessible,” he said. “A little bit like the Apple/iOS paradigm, where you say, ‘OK, I’ve tried to try to make it as simple as possible, and I hide all the complexity behind it. And I can do that, because I control both the hardware and the software.’”
“The idea is that, 70, 80 percent of the times you use it, it’s going to be very, very fast to make it do what you need to do to deploy an application. … So, the whole idea is that you can get to a satisfying, accurate program without really being a machine vision expert,” Montanari said.
Montanari said that while injection molding could benefit from automation for quality control processes, he and his colleagues have seen processors who have vision systems on the floor that aren’t being used because they are unable to maintain or update them, and quality control tasks are being done manually.
“And this is a huge problem, because first of all, it’s very hard to find labor,” Montanari said. “Second of all … because of the pandemic it’s even harder to put labor next to each other. So, the question is, why are these companies still choosing to do human inspection, instead of automatic inspection? And one of … the general trends that we’ve seen in the past is that a lot of these companies consider machine vision something very difficult to deploy, very expensive to deploy, that you cannot do internally … you need integrators and partners.”
As cloud-based technologies are becoming more and more prevalent, Montanari said Zebra has taken steps to build in data security. Working on the cloud presents the classic double-edged sword for manufacturers: On the one hand, it offers the promise of great advantages, and on the other hand, cyberattacks are becoming a more widespread concern. He said Zebra has taken a unique step to protect some of the higher-end models of its cameras by providing them with two physically separated ethernet IP ports, with a different chip controlling each port.
He said this arrangement uses one port to connect to a machine’s PLC, so pass/fail information is conveyed that way, and the other port can be used to move images to the cloud independently for machine learning AI techniques, etc. He said that from an outside perspective, the dual-port camera looks like it has only a single port, and the port connected to the machine’s PLC is not discoverable from outside.
Looking to the future, Montanari said that automation will continue to grow in importance. “Automation, in general, is going to be the key here. Because the labor part of the equation is never going to go back to where it was — especially in the Western countries. So, really, the way to grow the business is by automating, and the more you can automate your machines, the more you bring in IoT, robotics — and with robotics, there’s a lot of vision there.”
He said that increased life expectancy will also bring growing demand for medical devices, “… and many of these medical devices require high-end injection molding. One very specific example is hearing aid devices, which, 20 years ago, was like, something very clinical. And [today], a lot of older folks around the world, especially in the Western countries — I’m not saying they’re a commodity, but they’re … very much used. … All of that is injection molded, it’s very hard to make. … you need a lot of a lot of quality control.”
“And many companies still do this visual inspection manually. And so that affects their throughput. And this is a classic example where machine vision comes in.”
David Tillett, associate editor
Contact:
Zebra Technologies Corp., Lincolnshire, Ill., 877-208-7756, www.zebra.com
