Automate Your Manufacturing Processes with 3D Machine Vision Scanners [Podcast]

27 min 51 sec

One of Hermary’s long-time integration partners, Concept Systems, invited Terry Hermary to its Podcast to talk about how 3D machine vision advances industrial automation.

Concept Systems was founded in Albany, Oregon in 1999. Concept Systems has over 500 years of combined experience in manufacturing automation. Having worked with companies that make everything from jetliners to french fries, Concept Systems has helped many end-users increase efficiency and yield, reduce waste, accelerate production time, and improve plant safety through automation.​

Visit conceptsystemsinc.com to learn more about their capabilities and services they offer.​

About the Speakers

Tyler Kern

Podcast Host

Tyler is an experienced B2B host with a background in radio and sports broadcast. Those years behind the mic and in the booth give him a unique perspective on both the technical side of hosting and on running a well-organized, professional, and engaging program within any time constraint. He works to create a conversational and light atmosphere during interviews, drawing out the best of guests’ knowledge from right under their nose. His guests often end recordings with “Wow, that was easier than I expected!”

Doug Taylor

Principal Engineer of Concept Systems

The principal control engineer at Concept Systems, Doug is an expert in 3D machine vision and motion control. He is also vastly knowledgeable in working with industrial robots. Doug’s integration experience spans almost 30 years, 21 of which have been with Concept Systems. Doug is a pioneer in the automation forefront, engineering many systems that solve challenging and unique industrial problems.

Terry Hermary

CEO and Co-founder of Hermary

The CEO and Co-founder of Hermary. As a licensed Professional Engineer (P. Eng), Terry is analytical and quality-centered, making him the perfect liaison between customers and his R&D team. Anyone who works or comes into contact with him knows that Terry believes in seeking out win-win-win opportunities. This has also become the foundation of Hermary’s guiding philosophies.

Podcast Transcript

Tyler Kern: Today we’re discussing how 3D vision has changed manufacturing. And joining me for this conversation is Terry Hermary. He’s the co-founder and CEO of Hermary. Terry, welcome to the podcast. Thanks so much for joining us.

Terry Hermary: A pleasure to be here.

Tyler Kern: Absolutely. We are thrilled to have you all today and also joining the podcast today is Doug Taylor. He’s the Principal Engineer at Concept Systems. Doug, welcome to the podcast. Thanks for joining us.

Doug Taylor: How are you doing.

The Relationship between a Component Supplier and a System Integrator

Tyler Kern: How you doing. Well, It’s a pleasure to have both of you on the show today. Now, Doug, I wanted you to kick us off today. Just by describing the partnership between Hermary and Concept Systems. Can you break that down for us just to get us started?

Doug Taylor: Well, this partnership is really a valued thing at Concept. I tend to get into situations where people need very unique solutions. And that means I need some very unique suppliers. And one of my most trusted, longest supplier is Hermary, and Terry, in specific. We go back a long ways. The first time I met him, I wanted to buy his stuff. And he told me no. And he wouldn’t sell it to me. And that’s continued quite a while. He’s a very, very specific person. He produces probably one of the best lasers out there if not the best. And it’s just nice to be able to bring him some off-the-wall ideas and say, what are you thinking, and you know, maybe get a demo out there and give it a try and take the data and collaborate. It’s just really nice. It’s almost like having an engineer in your back pocket.

Tyler Kern: Absolutely! And Terry, I would love to get your perspective as well, when it comes to working with people like Concept Systems and Doug on the various projects that he brings to your table. Just tell us a little bit more about that.

Terry Hermary: From my observations, if somebody has a unique and challenging automation project, they could think of a lot of different places but probably one of the better places to take it to would be Concept Systems, and they do have some unique requirements. For whatever reason, I’ve always been quite interested in finding new product introductions, developing new applications, innovative solutions to things and it’s been very rewarding and a pleasure to work with Doug at Concept over the years. To do a number of those projects together that we can look back on and see that was, you know, successful. We get something that nobody’s done before. And in some cases, we see you know, it’s kind of blazing the trail in industries.

3D Machine Vision in a Nutshell

Tyler Kern: Right, that is really cool to be able to look back and say that. We’ll discuss some examples in this area here in just a moment. But Terry, give us an introduction to 3D machine vision. How does this work and what does it do? Tell us a little bit more about that?

Terry Hermary: Yeah, 3D vision is not that well understood by a large portion of the manufacturing industry as well as the systems integrators/automation companies. It’s getting better, but it’s still a technology that is, like I say, not that well understood. What 3D machine vision can do is it can augment the capabilities of automation systems and make them much more powerful and adaptable to the situation, amongst a number of other things.

One of my missions in my job is to do what I can to help educate people that want to learn more about 3D machine vision and where it’s good to be used where it’s not the right thing to be used. Because we do have 2D vision out there and that’s relatively well known, and, I would have to say that, Cognex has done a wonderful job of educating the industry on that. But when it comes to the 3D side of things, its day still hasn’t come yet, but its day will come very soon.

Tyler Kern: And Doug, is there anything you’d like to add to Terry’s comments?

Doug Taylor: Yes, 3D vision, if I was to say something as an analogy you know, people like me, we like analogies. But if 2D vision is a car, 3D vision is a helicopter. So there’s a lot more requirements put on the engineer when you’re doing 3d vision. You know, coordinate transforms. You know, 2D, you have a pretty straightforward coordinate system, XY.

Whereas 3D, you have you know, a lot of different things. You have different planes. You have different math formulas like that products between vectors and things like that. There’s a lot more complexity, but once you dive through the complexity, there’s a lot of treasure out there to mine. It really is. There’s a lot of data, you get a lot of data, you get it very quickly. But if you have the right techniques, you can actually get what you’re looking for, in a very usable manner.

Video you might be interested in

Capturing 3D Data with Geometric Measurement Techniques

3D vision data can be captured using many different ways. This video breaks down different geometric measurement techniques into simple concepts and illustrates them using real-world applications.

Advantages of Employing 3D Vision

Tyler Kern: Right, so with that being the case, let’s dive into some advantages. This provides for end-users who employ 3D vision, what are some of the benefits? Can you break that down for us?

Doug Taylor: So the main benefit of 3D vision is that you can interact with your data that is not plane. So a lot of people, they have a conveyor, there’s, there’s whatever they’re looking for is on the conveyor and you kind of interact with it on that plane of the conveyor. Whereas in 3D vision that gives you the ability to interact with the whole environment. The nice thing about 3D vision is it’s so flexible, because it allows you to do things.

But with every flexibility comes complexity. The benefits of it is that you can do things you wouldn’t have any other way of doing. The detriment of it is that you can do things that you really shouldn’t do and so it’s really, you know, walking through the forest and finding just the thing you want to find not getting lost along the way. That’s really it. But once you do find what you’re looking for, it’s amazing work in there really is.

Terry Hermary: I love that analogy. One of the things that I have come across in different conversations and conferences and things like is that people that have tried to use 2D technology to solve a 3D problem, simply because they were not aware of 3D. They were very much aware of 2D and the 2D is great for an awful lot of applications. Don’t get me wrong. I’m not slighting 2D at all here. What I’m seeing is that 3D could be much, much more deterministic than 2D and that’s kind of, I guess, the best I can say on that one. If you’ve got to be right 99.999% of the time, you really want to consider 3D as your technology.

Doug Taylor: Just expand upon that. You know, back in the day, the success ratio on something for a normal system might have been 99% and everything’s fine. But the way the industry is working right now, getting more accurate, and more accurate than that is really part of the ROI statement. You know, it’s harder to get to the ROI. Everything is harder now. So that means you need more refined tools.

3D Machine Vision in Industrial Applications

Tyler Kern: Now, Terry, earlier, we talked about examples. I’m curious, from your perspective, some of the examples where this has been deployed. What are some of the facilities where you’ve implemented 3D vision and what impact has it had on their operations?

“I see so much opportunity out there for this kind of technology to reduce waste, to improve throughput, to improve quality.”

Terry Hermary

Terry Hermary: Yeah, Hermary grew up in the processing industry, sawmills, the veneer mills, that sort of thing. And actually the wood processing industry segment is the second biggest user, by industry segment, of 3D machine vision and optimization. The biggest one is semiconductors. Now, semiconductors are using a different technology. It’s the interferometry and things like that.

So the sawmills have gone from being very labor-intensive and dangerous environments to being able to scan all the products and make decisions and have all the cutting tools adjusted automatically based on the scan data that’s been received. So you know, it’s a little bit strange, when you think about it. If you want to go to a sawmill, you go all the way out to the bush, down some pothole-y road, and into some dirt driveway and things like that. But inside today’s sawmills you’ll find some very, very high-tech [tools].

Other industries where this technology has been used so far is one of the big ones, that I’m proud to say, is within the meat processing industry, where there was an application where was initially motivated and justified by safety. But like many, many times when we see this technology going into applications, there’s unintended benefits that nobody had expected. And it would be very hard for anyone to have guessed or predicted that they were going to happen, where they actually get much better throughput. Maybe they meet the safety requirements hands down, but they get much higher throughput on a bottleneck in the processing line. And also they get a higher yield from a very valuable portion of the materials that they’re processing. So that’s one.

Another one is in the packaging industry, where there’s some fairly straightforward measurements that can be taken that can solve challenges that the industry has been facing for years and years, and the industry just had to live with it. And now, they’re embracing 3D machine vision. Again, it’s a very simple kind of application that they are using it for, but it’s solving a problem the industry has been living with for a long, long time.

And I guess another one is the industrial bakeries, you know, we’re hearing more and more from different industry segments that it’s harder and harder for these factories to get the people to do some of the jobs that people have done in the past but are not willing to do them anymore because they’re, quite frankly, jobs that could easily be automated. And so it’s these are jobs which should be automated, because having a human doing something that, you know, using that human as a machine, isn’t a very nice thing to do. And then of course, in the steel industry, we have some great fun there, with Doug and Concept Systems, and I guess I’ll turn it over to Doug, because Doug probably has some applications he wants to put on the table.

Doug Taylor: We’ve done some fun things with your toys. You know, it’s always amazing. All the jobs actually don’t sell. You know, those are some of the jobs that didn’t sell that we tried. There was some real neat stuff that you know – anymore – you really can’t talk about everything because of all the nondisclosures. But there were some really amazing solutions we were able to come up with that the ROI didn’t work out. But they were gonna be pretty cool. But the jobs that you did talk about you know, specifically, you know, working in the steel mills, that one was, you know, something they had to do because, you know, there was a lot of safety concerns. And it worked out pretty well. It really does. Actually kind of strangely amazing, you know, you think that I’ve seen a few things and I wouldn’t be amazed by this stuff, but it’s amazing what Terry’s scanners actually can do. It’ll scan red hot metal, and with a red laser. It’s easy. Outside. It’s a crazy thing.

Terry Hermary: I’ve been following on that one I can’t resist. It’s a little bit of a frustration for me, but an opportunity for industry. I see so much opportunity out there for this kind of technology to reduce waste, to improve throughput to improve quality. Yeah, there’s all kinds of opportunities and a number of different industries that I’m aware of, a whole bunch of industries that I’m sure I’m not aware of, but I was having a chance to get a tour of a plant or something like that I could pump them up pretty quickly.

3D Vision’s Role in Industry Growth

Tyler Kern: So, Terry, what are some of the industries that you think are poised for growth with 3D vision? Are there any that come to mind immediately?

Terry Hermary: Well, let’s start with who was Hermary. Hermary designs and manufacturers machine vision sensors, okay. We do not use system-level solutions. You know, we look to companies like Concept to be doing that, or machine builders where they’re building a machine and they can augment their machines’ capabilities by putting in some machine vision.

Rather than maybe even talking about the industry where I see opportunities, I would maybe look a little bit more at what are the impediments to these because there’s just so many industries. The biggest impediment, I think, is the ability for systems integrators that have the other skillsets, the control side of things and analysis of data and things like that, to bring into their toolbox, 3D scanning capabilities, yeah?

And Doug, and Concept, I’m talking about, you know, competition for you guys, but you guys are already way at the top in terms of the most challenging stuff. I’m talking about some of the more simpler things to do that are already done in for example, in the wood industry and things like that. You look at the wood industry, and you look at how the product flows and what has to be done to make sure everything is right, and you look at a steel plant that’s making billets or rolling out H beams and things like that. There’s a lot of similarities there, and there’s a lot that can be learned, you know, taking in the fight from the wood products industry into that industry, the steel industry, for example. And that’s just one of them.

Like I say, there’s an awful lot and really what’s needed is, you know, a broader base of systems integrators, that are 3D scanning-savvy or capable and again, that’s what I see one of the things that I see is my role at Hermary is to seek out those kinds of companies and help them to get geared up with this kind of technology. And we do that, you know, whenever we get the opportunity to, and for the most part, we’ve been successful and but not always, so.

Using 3D Data for Process Optimization

Tyler Kern: Now we know that data is an important aspect of 3D vision as well. So how can data that is collected be used to optimize processes and things of that nature?

Terry Hermary

Terry Hermary: Okay, I’m thinking on the micro level, you talk about data and one of the first questions people ask is what this 3D data look like. And that’s pretty straightforward, but it’s a bit of, sort of a science class, a very simple one.

But the bigger picture, the big-data picture kind of thing is, for example, let’s go back to the beef industry, where they’re using the scanner to acquire certain portion-specific information that’s used to derive a solution which is a non-trivial solution that’s requiring a path/pattern to be followed as material moves along the conveyor. But additional information they can get from that is, they can get throughput information, they get some idea about, not just throughput health, but also throughput portions, they can tell when there’s gaps in the data, meaning, you know, how’s the whole factory running, you kind of have like, like a stethoscope on your factory that’s all that information is available. So there’s one example of how data can be used in other aspects.

You know, another thing about using 3D data in a factory application is that you could be measuring things in real-time. Whereas in the past, if you wanted to measure stuff that was important enough to measure, you might have to shut down your production line in order to get access to the materials to the products in order to measure them. Whereas with 3D, you have some scanners mounted in the factory and you don’t have to have human to get involved in getting that information, basically immediately as a product to be produced.

Tyler Kern: Absolutely, Terry. Doug, is there anything you’d like to add? On the topic of data?

Doug Taylor: Well, after listening to Terry, there’s three types of data that we get, and one is the actual 3D data. And then the next type of data that we get is the solution data. Meaning what are we going to do with it? And then the third type we get is, I told it to do this, did it actually do this sort of data? So there’s always an echo of value out there, you know, beyond the initial just ‘Can I actually measure it at all?’

What’s funny is the quantity of data you get out of a 3D scanner is unimaginable to most people. Like it’s not uncommon with this scanner to you know, to get 600 points 200 times a second. And now what are you gonna do with it? You have this, basically a firehose of measured points coming at you. And you kind of gotta have your act together. Like connecting to that firehose or else you know, it’s almost just gonna hose you down.

So a lot of the problems with using 3D data are structural, meaning it’s all been done by people and, you know, there’s no good place to put the scanner, yada yada yada. But once you finally break through those problems, it is amazing. You know, the information you can gather beyond what you’re actually looking.

Recommended for you

FAQ on Point Cloud for Industrial Automation

Point Cloud data is indispensable in enabling process optimization, 3D inspection, and dimensional monitoring. A series of short videos using real-world data to showcase Point Cloud representation.

Tyler Kern: That is excellent stuff. And, you know, I think that your example of trying to drink out of a firehose, when it comes to data is a really good one.

Terry Hermary: On that topic, if I may, is the ability to predict the future with the data, you know, statistical process control, for example, when you’re measuring, getting all this data, and you can predict that your process is drifting and you can take the alert, some higher form of intelligence or take whatever kind of corrective measures so that you are continuing to make product that’s in spec and you take the corrective measures before you’re making out-of-spec product or before it gets being scrapped at the end of the day. So, so there’s a lot of being able to avoid problems, you know, before they actually happen, available to use of 3D scan data as well.

The Outlook on Digital Transformation

Tyler Kern: Now, one thing I would love to hear from both of you on is whether or not you’ve noticed that attitudes have changed towards digital transformation throughout the pandemic in terms of what it can do for a facility and the benefits of it and that sort of thing.

Terry Hermary: My observations on that is that when a new technology is going into industry, it takes very confident people, people that are not scared of losing their position or their job or whatever, because you’re taking a risk. And that’s sort of like the first thing that has to happen. The first major breakthrough. What I would say, I have seen is, I’ve seen, not the first adopters, but the early adopters. The early adopters seem to be more likely to be adopting now a little bit less resistant or hesitation, maybe than they were in the past because they do see their fields now – it’s challenging particularly in getting employees into the business – but they can see now this technology as being a way of not needing so many employees to do these very mundane, repetitive tasks. But in terms of the breakthrough type of things, I haven’t really seen much of a change in sort of call it the uptake, if you like, of new technology.

Tyler Kern: And Dough, what do you think about that, especially that pandemic the impact that the pandemic has had on the acceptance and adoption of new technologies? Is there anything you’d like to add?

Doug Taylor: One thing I’ve seen is, first, the robots out there right now are a lot better than the robots were a little bit ago. And the collaborative robots are basically an unfulfilled promise that’s getting fulfilled. I mean, a lot of processes that would benefit from robot assistance. But in general, you know, it started as just, oh, it’s this high-tech thing or whatever.

The problem with using a robot is it can do a lot of things but it’s hard for it to sense its world. And that’s where 3D scanning really helps. It’s one thing to know that, let’s say, is painting something or it’s doing something, but it doesn’t know what to do. And it’s hard for it to figure out what is to do. You can just train it, you know, by rote and it’ll do that.

But the nice thing about the Hermary scanner is its size for a human. That SL-1880 has an 18-inch standoff to 84-inch standoff, which is a bit more than a human but not that much more. So it allows you to see things and do things with the robot. If you’re willing to go through the pain and suffering that coordinate math is all about. And that that’s usually the first hurdle, is you got to say, okay, here’s the coordinate origin of the scanner, and here’s the coordinate origin of the robot. It is different. How do I make it the same? That’s the first problem and then you get to the second one is here’s a coordinate origin of the thing I want to play with. And I want it to rotate about this origin. So there’s a series of hurdles you got to go through but at the end of the day, the robots are now good enough to do it. And for a long time, the scanners have been definitely good enough to do it.

And generally, you know, unfortunately, we’ve lost people during this COVID thing and people have changed careers and moved on and there’s a lot of different turmoil in the market. I think it’s really a good time for people to be looking at, do I really need a human working? Do I need something that is a little bit different? It’s not so much of a safety risk for it to do repetitive motion or something like that? Because it is a machine.

So I see a lot of people really trying now that I’ve never seen trying, you know, developing 3D, developing robotics and developing, ‘How do I’, you know, ‘start to address these big problems that we’ve always wanted to fix.’ But you know, there’s been a good reason why we didn’t do it, you know, whether it’s just inertia or the people or even just getting people to maintain it once it’s built. There’s a shift going on. And fortunately for us, we have companies like Hermary, you know, you take a look at Fanuc’s new, you know, line of robots, things are now coming into alignment, and I think, you know, it may be spaceman sort of stuff, you know, to have robots everywhere, but I see there’s gonna be a lot of robots in people’s future.

3D Machine Vision as an Engineer’s Tool

Tyler Kern: You have both been fantastic guests here today. And I’ve loved this conversation. Before we wrap up. I’d love to just get any final thoughts, conclusions or anything that you want the audience to walk away with here today after hearing our conversation so Terry, let me kick it over to you first for any final thoughts.

Terry Hermary:  My message would be that 3D vision’s very empowering. Yeah, there’s some hurdles you have to get across. But the thing is, once you’ve crossed those hurdles, now once you know what your coordinate remapping matrix looks like, generally that doesn’t change, and it’s fixed and so it’s not an ongoing struggle or challenge. It’s kinda getting over that peak onto the other side of the hill. And it’s a beautiful valley on the other side where you can realize the capabilities of 3D machine vision and really have a much more powerful automation system that’s able to adapt to its environment and to the particular materials that have been processed.

So yeah, I would encourage anyone to who’s interested in hearing more about this kind of stuff to feel free to give me a call because that’s what I do here. I sit here and wait for the phone to ring and look forward to talking to people about the kind of challenges they have and things like that. And if it’s an end-user calling, a lot of times, we’ll get Doug and Concept Systems involved and things like that. So it’s about all I would have to say, thanks.

Tyler Kern: And Doug, wrap us up with your thoughts here today. You get to tie a bow on this thing. So, Doug, the floor is yours.

Doug Taylor: Yeah, two things. One, it is fun to do this work. It’s really rewarding. It’s really challenging. And there’s a lot of pain involved with it. But if you’re the type of person that wants to do this work, you know, either work with us or work for Concept Systems or anything like that. We want to talk to you, you know, there’s a lot of work out there. And it’s one of those things us dinosaurs – there’s kind of a class of engineers I call dinosaurs you know – that’s old people. And then there’s these young guys and the young guys and girls, you know, are really well trained to do this. They just need an opportunity but they need to know there’s an opportunity to be had and there is an opportunity. So Concept, we want some help.

Second thing, if you have a really challenging problem, and I mean really challenging, contact Concept Systems about it. We like the hard stuff. It’d be fine if all you wanted to do is cut circles out of wood. We can do that all day. But if you want to do some involute spline thing you know while it’s rotating at 50 hertz, that’s the sort of stuff we want to do. We like that stuff. And we like people that want to like that stuff, too. So that’s kind of where we’re at. That’s why we partner with really smart people like Terry Hermary and the Hermary scanners and we have other partners too, but we value every one of them because when somebody finally gives us something to sink our teeth in, then we got to go to team that’ll give them a solution, you know, or at least be part of the solution. And in every case working with Terry, he’s always made me look at it. His stuff works the first time, I want a little change, I get a little change. It’s really a pleasure to work with Terry. It’s a pleasure to work with a lot of the top-level people in this world. But Terry and we give each other a lot of pain and a lot of suffering and that’s just kind of our way. And it’s kind of fun to work with people like Terry.

Tyler Kern: Excellent stuff, Terry Hermary, the co-founder and CEO of Hermary and Doug Taylor, the Principal Engineer at Concept Systems, Terry and Doug, it’s been a pleasure having you on the podcast today. Thank you so much for joining me.

Terry Hermary: It’s a pleasure and just want to say to Doug, thank you very much for not asking about our new scanning platform, the Amadeus platform today, thanks.

Doug Taylor: I expect some scanner in my mail soon. Thank you.

Tyler Kern: Absolutely and everyone out there thank you for joining us for another episode of Beyond the Concept, the podcast from Concept Systems. We’ve loved having you along with us here today. To stay up to date with the latest from Concept Systems, make sure to go subscribe to the podcast on Apple Podcasts or Spotify. Or of course, go visit the Concept Systems website as well. All of those are excellent resources to stay up to date with the latest in thought leadership from Concept Systems. And of course, stay tuned to upcoming episodes of the podcast. We’ll be back shortly with more episodes of the show. But for this one for my guests today, Terry and Doug. I have been your host, Tyler. Thanks for joining us.

Free eBook Download

The evolution of the sawmill industry and how it became one of the world’s most digitally advanced businesses.

Industry experts share best practices on realizing hidden savings in automation projects.