Episode Transcript
Speaker 1 00:00:12 Humanizing robots to make them more human so they can be easily introduced into manufacturing plants regardless of their field.
Speaker 2 00:00:21 Hello everyone. And welcome to the robot industry podcast. My name is Jim burrata and I am your host. And thank you for joining us today. My guest is Michael Muldoon. He is from avian R and he's the director of sales and product development. They're based out of Montreal and I've been to their facility before. So I'm really excited to have Michael on the program today. Michael, welcome to the podcast.
Speaker 1 00:00:44 Thank you, Jim. Really appreciate the invite. Hey, let's
Speaker 2 00:00:47 Talk about AB and R and can you give our listeners a brief about what you do? Okay.
Speaker 1 00:00:54 Yeah. So just by being on the podcast, we are in robotics. Um, but there's some very specific fields that we, we specialize in. So one of the technology verticals is robotic finishing and then we have, uh, surface inspection as well. So there's, um, and primarily we focus on some of the higher end industries in terms of regulations, such as aerospace and medical device. Um, you'll see if you follow us on LinkedIn or Twitter, we're also doing a, a push to, to take our solutions and bring them into more general manufacturing as well, to
Speaker 2 00:01:31 Very nice AB and are started in precision inspection and grinding and polishing of aerospace components. Um, when did you start using things that you learned in aerospace and apply them to medical device?
Speaker 1 00:01:43 Yeah, well, the history of ENR is a little bit more nuanced than that. We started in robotics and machine vision and, you know, we were in all types of industries and typically, um, you know, we'd start close to home. So our first aviation experience was at GE in Bromont Quebec, and that's where we started to cut our teeth on debriefing applications. Um, you know, when any time you, you get into many different things or be a generalist, typically you'd have to do the really tough stuff. So we, we had a lot of fun with, and we were really good at w robots and machine vision. And we started to think about, well, what if we specialized in, uh, technology, uh, a specific field and a market and just became really good at that. And what type of opportunities would that open for us? And, you know, that's when the focus in aerospace started to happen.
Speaker 1 00:02:43 So that would have been maybe around 2005 to 2007. And, um, we started to look at what were the needs in that industry. And, you know, back then robots for aerospace was not a commonplace. Let's say, you know, it was, um, an industry that there was lots of needs. There was lots of, uh, manual labor put into these parts. And there was a very talented group of people, you know, the, the operators on the shop floor, they knew their, their parts and their processes inside out. Um, but it was, you know, adopting robots was important for them because they had a cap or a ceiling on their growth and transferring that knowledge. And so we recognized that and we thought it was an opportunity and we started to try and figure out, you know, how can we frame up an offering and, and, you know, um, uh, systematically knock off some of those hand finishing and inspection processes that were so common there to, to allow those customers to unlock the potential of automation like, uh, the automotive industry had done.
Speaker 2 00:03:51 And so then you transitioned to medical device.
Speaker 1 00:03:55 Yeah. So that's pretty recent actually. So the, those efforts, um, so as we did our specialization in taro space, we, we knew one in one sense that allowed us to compete internationally. So, um, we became known in the aerospace industry in particular blades, veins, rotating parts as the go-to player for any robotic finishing and inspection. Um, so as we came, became better at that, all our customer names, you know, we were both 90% aerospace and we knew one day that would come back and bite us in the butt. And that's when we, we started looking around and said, okay, what is similar that has, you know, um, similar challenges, um, high value products that, you know, we can contribute and help, and the medical implant industry in specific specifically orthopedic implants. Um, so we're talking knees hips, um, ankle implants, shoulder, those type of, uh, trauma plates, that industry there was, um, turned out to be really interesting. And so maybe let's say three months before the pandemic started, we started our sales effort. The development efforts started maybe about eight months before that, where we started to, you know, reach out to customers or look for industrial partners that would be willing to, you know, throw some parts at us and see what we could do with them.
Speaker 2 00:05:25 I've mentioned that the, at the preamble there that I was in Montreal and I was lucky enough to get a, uh, a walkthrough a year near downtown plant. And when I walked through, it was, it was a lot of encapsulated, uh, robot tools. Uh, we usually with conveyance, with vision, with lighting. And so these were small modules of automation. Uh, is that typically of what, uh, what type of machine that you might deliver to a client today?
Speaker 1 00:05:50 Yeah, so when we started going internationally, did the, the, the effort to make things easy to install that was big on our mind, you know, so we just, uh, uh, we have, uh, someone going out on the road to South Korea, for example, next week. And so when we, we don't want to send an army of people necessarily, and we want the customer to be able to do the majority of works. Then when we show up, we're doing the value added things. So typically, yes, everything is kind of like a island automation or within a box. And there's a couple of reasons, one to make it easy. Uh, also when you start grinding or polishing some of these materials, they, the dust, you have to collect it. And so there's all the dust collection that goes into that and the environmental controls, um, and all the safety around that as well, too.
Speaker 2 00:06:41 No, there's some cause some good points there. Um, is there, I wanted to ask you a little bit about the tech behind some of the very difficult things that you're doing. And I wanted to talk a little bit about the difference between accuracy and repeatability in robotics. And can you explain the difference for our audience?
Speaker 1 00:06:58 Sure. Yeah. So repeatability that's where robots really, uh, grew from. Uh, and that's just, if you want to go from a single point doing that repeatedly every time, that's what the robot is, is very good at. Now, the path that takes to get there, there can be variations and most robot companies won't put an actual number on how accurately it follows a path they'll estimate for you, but in the literature you won't, you won't see it. So that's the difference where you, you asked something to do to a point how, how accurately does it hit that point? That's the measure of repeatability and then does it follow the same path each time? That's how accurate that's the way we think about it.
Speaker 2 00:07:41 Great. And how do you overcome the issue with respect to vision inspection and Paul?
Speaker 1 00:07:47 Okay. Two different kind of realms and two different impacts. Um, for visual inspection, you know, our, our, our flavor of inspection is looking for niche, Dene scratches on a complex surface. So in a critical surface, so you imagine a knee implant, it's a mirror, a near mirror finish, and you can't have any blemishes on that knee. So when the surgeon picks it up, if he sees one of those, it's going back to the manufacturer. Um, now looking for those defects, we don't have to be accurate. Uh, we just need to be repeatable. So when we pick up the part, we're holding it in front of the camera and we're taking a picture and we're controlling the angle of the lights and the cameras and our algorithms take care of the rest now to make a measurement. That's where accuracy comes into place. You know, every time a robot picks it up, if you want to do any form of metrology then and relate it, potentially some, um, w one angle relative to a datum that's maybe blind on the other side that the robot's hating onto.
Speaker 1 00:08:51 That's where the accuracy of the robot comes in. And to be honest with you, for the types of processes we deal with, we try and avoid the metrology. If, if we're trying to measure something, it's like a go-no-go gauge. So in process, the customer wants to know, okay, am I going in the right direction? Is everything okay? And that's usually good enough if they want some higher end stuff. There's a lot of really cool tools that they're, non-contact CMMS, where they get into that art, right? And typically you don't see a robot inside of there unless it's loading and unloading it. Okay. Um, in the finishing realm, um, when, when you're polishing a surface or when you're delivering an edge, you know, you typically, you want to follow the, that edge or that surface and, you know, parts, even in the aerospace industry where they say they're super accurate, there's variations that happen.
Speaker 1 00:09:46 So the trick with the robot is to add some form of touch or some form of calibration to accommodate for those variations and allow the process to absorb any, uh, not imperfections or any variances in the, in the, in the part. So you, you don't worry about accuracy. You, you worry about dealing with variations. Primarily. Now we've got one process that, um, that this year, uh, we finally, it was on our hit list to do it's called a tri-blend right? So you've done an airfoil and you've got a leading edge of an airflow, comes down into the platform and you've got three surfaces coming together. So that's why they're the tri-blend comes in and you've got to blend all those surfaces together and you have to create a shape. You have to create a radius. And that's where the accuracy becomes very important. So it's, it's not only the robot, it's also the tools that you use.
Speaker 1 00:10:41 So we do a lot of dressing of tools. We do a lot of, um, uh, tracking of real diameters. And then, uh, sometimes we'll do an inspection before and after to make sure we're going in the right direction. Um, and, um, the way you teach your robot, you know, in the number of control points that helps you to, um, if you're teaching more points, you're going to have a little bit finer control on dealing with the inaccuracy of the robot cause your, your path is going to be tighter. So definitely there's lots of little tricks that, of working around and, and using a robot for what it's good for
Speaker 2 00:11:22 The, um, I just kind of imagining, uh, uh, a lot of your vision professionals, uh, with taking pictures or, or, you know, even just connecting lighting to glass objects, sorry, like a mirror, like surface object, right. That to be a big, you must spend a lot of time, uh, attracting a vision pros to your, uh, to your team.
Speaker 1 00:11:43 Yeah. Machine vision, people that are creative bunch, and they're, they're not easy to replicate. So for sure, you, you know, usually, uh, you can kind of, you can kind of smell the, the person and then do they have the right kind of creativity to get into it? Cause you're right. It's optics, it's this mechanical. And then it's, it's the software to the algorithms that go into it and understanding, you know, the parameters and how to play with them so that you can deal with certain variations. And then the other variations they have to be dealt with mechanically. If you don't get a good image, then you won't get a good result of the backend.
Speaker 2 00:12:21 Yeah, absolutely. No, it's a very interesting part of the automation puzzle. I did want to talk a little bit about, um, uh, one thing I noticed when I was in Montreal and I think it's the same thing now, uh, you have a big partnership with FANUC and it's a bit unusual to be tied to one robot brand. Can you expand a little bit on that for our,
Speaker 1 00:12:39 Yeah. That's in the early days, you know, we turned a lot of heads by doing that as an integrator and, you know, we, we had dealt with many different arms and different colors. So when we decided to focus, it was really, um, you know, every time we did an application, it was almost rewriting the book, right. So, and when you rewrite a piece of robot code or any code there's bugs that can, that will be in there, right. There's always bugs in software. Um, so by standardizing on a platform that allowed us to focus on the tools to make our job easier and hence make the customer's job easier. So we've got two software platforms now, uh, they're built off the same kernel, but one's focused on robotic finishing. We call it brainwave and the other is focused on visual inspection and we call that owl vision.
Speaker 1 00:13:32 And those grew, um, in large, well, part of the reason we could grow those so well was by keeping the robot platform stable. Right? So we focused our attention on, um, you know, in the robotic finishing, what makes brainwave really interesting as you've got, you know, you've got your spindles, you've got your force F four sensor, you've got your compliance devices. And you've got a lot of different pieces of the puzzle that have to come together and create a process. So the brainwave software centralizes all that and allows us to do, you know, interesting trips once it's all on the same page, you know, then you can start talking adaptive processes, you can start doing more complex things much faster and iterate much easier.
Speaker 2 00:14:18 So let's change the conversation again. And let's talk a little bit about data and is this something that's important to your customers now?
Speaker 1 00:14:26 Yeah. Yeah. So that's been going on for a little while where, you know, no longer is a robotic system, something that moves apart from a to B or does a task. It's also an information system. People want to know, okay. When it did it, what, what were the parameters or what happened while it was doing it? Um, we had an example, so we were supporting a customer in UK and all of a sudden, out of the backend of our machine, it was a polishing application. They were getting a batch of parts that had, uh, problems. And so, you know, of course we're like, oh my gosh, maybe the calibrations out of, uh, out of sync or maybe they put the wrong abrasives. Um, but we, we have a lot of, uh, we built in a database into our system. And so we S we store a lot of the parameters that happen during what we call a recipe.
Speaker 1 00:15:20 A recipe is the process parameters that happen during a, let's say finishing process. And as we started to rewind and look at, you know, let's say the force, uh, the forced schedule that happened on each of the parts we noticed in a certain spot, the forced schedule is out of whack compared to the previous batch. And then as we dug deeper, we said, okay, take a look at those parts and check in this, this region. And sure enough, there was a core shift in their task part. And so they were able to use, use the information to troubleshoot that earlier in the process, they had a core shift that you couldn't see by eye. And so these parts were going down the line. Everyone thought it was great, but then all of a sudden an error happens, right? So it gives the troubleshooting tools, um, and, um, you know, a way for our customers now to, to improve their operations.
Speaker 2 00:16:14 I thought that the question would be more about your customers using data, but you're using data for your customers. So that's very cool.
Speaker 1 00:16:22 Yeah. Well, it's the chicken or the egg when it comes to data and even the topic of AI that's hot right now, and you need to have data in order to do anything in AI. So as you learn about AI, you'd go shoot, okay. We need to, we need to have data. So, you know, making the recipes easier to teach and having some algorithms or some AI that, that helps that in tune to an, a robot. Um, but we need to have a dataset of a robot that ran really well. And also a data set of robots that were not, were not optimal. So that's, that's one of the driving forces behind it. Um, I have to say, I think our customers, they're pretty smart and they were pulling on data before we knew how important it was. Yeah.
Speaker 2 00:17:06 And are you one of your customers talking about AI and the application from that data?
Speaker 1 00:17:11 Yeah, everybody well, okay. So when we look at the, let's go to the Aero engine, OEMs, everyone has a team that's revolving around it and trying to figure out how to, how to deal with it. The big challenge in aerospace, um, is making AI explainable, um, rolls Royce. They've got a really interesting initiative. They, they open-sourced, um, that, um, where they, they do this checklist is AI, is it ethical? Is it explainable? And they have this reader, when they approach an application, will they be using AI in the right light? And they made that available to everybody. So it's, it's very important. Uh, but you know, with customers in regulated industries, it's a little bit of a conundrum because it is so powerful. They want to use it, but they have to be able to explain when you run data through an, a network and makes a decision, why did it make that decision? So, you know, there's, there's still some, some work to be done to, to use it everywhere.
Speaker 2 00:18:18 I remember that podcast, actually, it's your podcast that I listened to about having rolls Royce, your experts from rolls Royce on it. That's the one pager that they just said, we're just going to make this available to everyone because it's good for the world.
Speaker 1 00:18:30 That's right. And I always remember, I forget the name of it, but it's the only Thea framework. And it came back to me and, and, uh, it's, it's really interesting, you know, the, um, uh, what the, what the team has done over there and, and how they're willing to go out and, and share that because aerospace industry, you know, everybody is, has some partnerships, but nobody likes to talk about it. And so there's a lot of, uh, information silos and, and stuff like that. So it's, it's nice to see a company that feels that a topic is important enough. They're willing to share.
Speaker 2 00:19:03 Yeah. And I'll put that what a link to your podcast and the show notes, uh, from this, uh, from this episode, I wanted to ask a question about the pandemic, and hopefully we're kind of on the edge of the end of the pandemic, but has the pandemic affected the business of, and,
Speaker 1 00:19:19 Well, certainly it slowed things down for us when our installed team never stopped. Um, they, they continued to go to installations. We were considered essential service, uh, the sales team, we couldn't travel. So that certainly made it a very interesting time. We all became internet personalities and, you know, we got ourselves a set of hollow lens and we're giving virtual tours, the interest never stopped. Um, the money available. It certainly did slow down, uh, especially when, um, the first lockdown happened. And then nobody was sure where this thing was going. Right. You know, would a vaccine come out, uh, how long would we have to make cash last? So it certainly slowed things down this, this summer, it picked up, uh, and it's super busy. Uh, right now I'd say June was probably the turning point. You know, my feeling is I don't see how that's in Canada. We're pretty lucky. You know, everybody is above 80% vaccination rate. Um, so I don't see reasons why they would, would close us down again. And everybody wants to get back to business and, and back to life here and fly again. Absolutely. Yeah, no, absolutely. Okay.
Speaker 2 00:20:37 Um, what are the questions I'd love to ask? Some of my experts is, is how do you attract personnel to, and AR because you've got some very interesting and unusual skillsets robotics for sure, but, uh, vision and sales and all the rest.
Speaker 1 00:20:52 Yeah. So we have a great team here and, um, our, um, HR and marketing team, you know, that's one of their priorities and we've, we have this strategic project, we call it that, you know, to make us an employer of choice. So next time you come to, to Avion our we're actually on the south shore now, but the office is set up. We've got a, it's a really great vibe at the office. Um, there's fresh fruits every day. There's, uh, um, uh, Sancta sets a it's the five to seven where we get together. Uh, uh, the management during the pandemic, the, the level of communication was great. So we work really hard at making Adrian are a great place to work and, you know, podcasts like this, you know, having the chance to chat with you, Jim, and, and talk about what we do and, and, you know, be open about what, what it is that we're into and share some of the information, uh, that's important because it's, it is a fantastic place to work. I started here in 2007 and, um, the, the different applications and the different things you get into everyday is really interesting. And when you tackle that with a good group of people, it's, it's a lot of fun.
Speaker 2 00:22:02 Yeah. It's such an interesting, interesting industry that I it's kind of like glue, right. It's hard to get hold of that for sure. And I did want to give a plug to your podcast. It's called a robotics for all podcast by a V and R and you're on all the, uh, on the apple and Google and all this.
Speaker 1 00:22:21 Yeah. Thanks for that, Jim. Absolutely. That's a part of the pandemic and becoming those internet personalities. And I took to, I was a fan of your podcast and, uh, we said, you know what, let's give it a try here.
Speaker 2 00:22:34 So where do you see industry going from here? I'm kind of assuming that as people start flying again, and we are the we're hopefully on the end of the pandemic, uh, where, where do you see industry going?
Speaker 1 00:22:45 Okay, well, things slowed down. Uh, I don't see things back to 2019 levels yet in terms of levels of production, but due to the slowdown, there was, uh, did people let go from a lot of organizations? So there is a need to hire people back. We have many customers that are saying, okay, we need to get people back. They have trouble finding the same people. So there's a skill shortage. And so they're saying, okay, well, maybe this is that opportunity to really look at technology and robots to help me get back up to production levels where I'm the demand that I'm facing. Um, in some areas, um, you know, the, the electric car business, for example, here in Quebec, uh, we've got a great aluminum industry. And so some of the secondary aluminum markets they've, they've benefited from things like the electric cars and third, the order books are full, but there's not enough people to do all of this. So that I see that as a driving force, um, how long that will last. I'm not sure. Cause you know, there was the Serb up here in Canada and a lot of the government, uh, um, help for people which has ended now. So I think a lot of people would get back to work. Um, but I, I see there's the demand for things is, is, uh, is greatly picking up in some areas they just haven't stopped or slowed down.
Speaker 2 00:24:17 So the ROI discussion to kind of it's, it's complicated, right? It's always like, Hey, this back in two years, but now with this, with this labor gap, I'm sure you're finding the same. Yes.
Speaker 1 00:24:27 Yeah, yeah. It is. It did change. Um, the way customers look at ROI sometimes just to get product out the door, they need to invest in build, um, a modern infrastructure. Um, overall I think that that's required. I think if we're going to have the silver lining in the pandemic, you know, those who survived got better and you know, this is a real example of why some of this technology will help us. Um, you know, it's not about replacing jobs, it's about, you know, securing production and growing your, your facility with the good people that you have in so securing jobs and, and growing things with a smart use of technology.
Speaker 2 00:25:14 Thank you for that. Um, Mike, I'd like to ask you a personal question too. Like what do you like to do when you're not selling robot solutions?
Speaker 1 00:25:21 Oh, well I was, I guess I'm one of those guys that was driving, uh, some of the demand for sporting goods products. So, um, I've got three kids, they're all into sports. So I'm a coach. Uh, I met coaches for soccer. Uh, I've got an 11 year old boy that, uh, only wants to think about soccer and he says, so that keeps me quite busy. And then I've got my, my nine-year-old, he's a hockey goalie. So I've had to learn a little bit about that and I get on the ice and work them out. And, um, and then my daughter she's into dance and I know nothing about dance and I promise I'll never try and learn anything about dance. I've I've got the robot down pat and that's it. Um, but she's, she between all of those, I'm a great chauffer. And then when I have my time I'm into cycling so that, uh, you find me on the road at five 30 in the morning, uh, cycling and trying to keep, uh, keep my stamina up so I can keep up with all these kids. So
Speaker 2 00:26:19 You can do the driving.
Speaker 1 00:26:20 Absolutely. Yeah.
Speaker 2 00:26:21 That's great. Well, thanks again for coming on the podcast. How can people get ahold of you?
Speaker 1 00:26:25 Yeah, absolutely. So you can check out our website, avian R avr-global.com. Um, search my name on LinkedIn. Um, I mapped them, uh, on LinkedIn. So it's the whole company. So you, you find a lot of, uh, good people to ask questions and, um, um, we're up here in Montreal. So, you know, if you want to come and have a visit, if, uh, uh, we love having people come through and, and take a look at what we do
Speaker 2 00:26:52 And Montreal is a great city to visit, uh, more restaurants per capita than New York city. I've heard. That's, I'm not sure if it's true, but I'd like to try it out more and I look forward to it. Yeah. Yeah.
Speaker 1 00:27:03 I heard that stat too. Jim, in that we've got some great restaurants.
Speaker 2 00:27:07 Absolutely. Thanks again, our sponsor for this episode is Earhart automation systems or hurt, builds, and commissions turnkey solutions for their worldwide clients. With over 80 years of precision manufacturing, they understand the complex world of robotics, automated manufacturing and project management, delivering world-class custom automation on time and on budget contact one of their sales engineers to see what Earhart can build for you. I'd like to thank and acknowledge our partner. 83, the association for advancing automation. They're the leading trade association in the world for robotics, vision and imaging motion control and motors, and the artificial intelligence technologies visit automate.org to learn more. And I'd like to thank our partner painted robot. He did robot builds and integrates digital solutions. They're a web development firm that offers SEO and digital social marketing and can set up and connect CRM and other ERP tools to unify marketing sales and operations. So there are painted robot.com and if you'd like to get in touch with us at the robot industry podcast, you can find me Jim Beretta on LinkedIn. We'll see you next time. Thanks for listening. Be safe out there. Today's podcast was produced by customer attraction, industrial marketing, and I'd like to thank my nephew, Chris gray for the music, Chris Coleman for audio production, my partner, Janet, our partner's eight three painted robot and our sponsor Earhart automation systems.
