[00:00:00] Speaker A: Ease of use is instrumental in helping the industry adopt a new technology and run with it.
[00:00:08] Speaker B: So the ultimate goal is to eliminate the need for coding and minimize the integration efforts in automation processes.
[00:00:26] Speaker C: Hello everyone and welcome to the Robot Industry podcast. Thank you for subscribing and we're glad you're here. And I have two people actually on the podcast today from Mari, Maple Advanced Robotics Inc. Out of Toronto and out of Vancouver. I'd like to welcome Armin Katunabadi and Yi Lee to the podcast today. So we'll start with Yi. Can you tell the audience a little bit about you and a bit about marijuana?
[00:00:53] Speaker B: I'm the co founder and the EVP of Maple Advanced Robotics and with over 20 years of experience in robotics and automation production equipment systems, I've been deeply involved in developing innovative automation solutions and tailored for unstructured and uncertain scenarios. And Maple Advanced Robotics is a robotic software company based in Richmond Hill and we specialized in addressing the challenges of flexible production and freedom surfaces. And we deliver innovation solutions for high mix and low volume production.
[00:01:35] Speaker C: Thank you very much for that. Yi and Armin, your turn. Tell us a little bit about Armin and why you're here.
[00:01:41] Speaker A: Thank you very much, Jim. Very excited to be here today. So hello everyone, my name is Armin. I'm a serial entrepreneur. I have always been fascinated by high tech industry and high tech startups. I have created my own startup and I have been in the automation industry for the past almost eight years. You know, especially when it comes to robotic and robotic vision. I have introduced new technology to the industry. I have been through the ups and downs of, you know, difficulties to get the industry to adopt a new technology. I also have been working with NIST on the first ever standard for beam picking application which is going to be a new ASM and standard. I also sat on a strategic board of A three. Yeah. Now I'm here with Mari because I highly believe in what they're doing.
[00:02:44] Speaker C: And you've just joined, Armin, you've just joined Maple Advanced Robotics from Apara AI, correct?
[00:02:50] Speaker A: Yes, that's correct. So that's, that's actually the startup that I was referring to. So you know, I, I started that company late 2016 and I, I was with that company, you know, introducing a new revolutionary technology we call 4D vision for beam picking application, which is simply an AI engine that enables the robots to see, you know, the parts just like humans and then be able to manipulate and pick them. So I, and, and I gained a lot of valuable experiences, you know, dealing with A lot of large and, you know, small manufacturers and what their pain points are. And one thing that I learned, which I like to actually expand later maybe during this podcast, is that Ease of Use is instrumental in helping the industry adopt a new technology and run with it. And when I saw what he and his people have done in Maple Advanced Robotic, it's really got me excited because I think they have done a great job. So I left apparel back in May and I've been helping Mari for the past couple of months.
[00:04:02] Speaker C: Well, that's great. And e, this question's for you. Like, can you tell the audience how you got started? Like we know you're doing, your focus is on unstructured, uncertain robotic software, but can you tell us how you got started?
[00:04:14] Speaker B: Yeah, sure. Yeah. So in my previous professional endeavors and I dedicated to developing robot solutions for advanced manufacturing and through this journey I encountered extensive manual programming and integration efforts. So spending like countless hours aligning various robot hardware for automation projects and also dealing with the part variations, you know what I'm talking about. So motivated by those challenges and experience so with other two co founders and the team and we started creating this innovative robotics platform. So the ultimate goal is to eliminate the need for coding and minimize the integration efforts in automation processes. So our system simplified the deployment, operation and maintenance of robotic automation systems and make a groundbreaking advancement in user friendly AI software and plug and play hardware.
[00:05:20] Speaker C: That's a great story. Thank you so much. Now I'll go back to Armen for this question. Can you tell the audience that what's making the offering so different in the industry? Is it intentional ease of use?
[00:05:33] Speaker A: Oh yeah, actually thank you very much for that. Of course I will let Yi address the more technical aspects of it. But one thing that Jim, I actually learned is that all of the customers have some hierarchy of needs. So usually they require a certain performance and then that is met, requires certain reliability. And when that is met, they require ease of use. And finally comes the pricing. And one thing that I learned during my time at Appara is that this industry is a little bit different from the rest of the industry. So in order for you to be able to sell anything to any manufacturing company, you need to be enabled to provide the performance and reliability that's given. So if you cannot provide those, they are not going to even think about using your system. But I think one thing that really shines when it comes to Mari's product is ease of use. And I've seen a lot of products and what I can tell you is that the Way that they have designed the interface and how fast people can actually get the system up and running is something that really got me excited. So yeah, definitely one of the most important differentiating factors which is both done by designing a very intuitive graphic user interface and an advanced AI engine behind the scene that actually does the heavy lifting. So I, I will let actually Yi talk more about the technical side.
[00:07:03] Speaker C: Yi, I have a question for you too. Before you dig into the technical side. Is it really hard to make something seem really easy?
[00:07:12] Speaker B: Yes, that's true.
Especially in robotic world.
Yeah, like there's so many people when they look at the robot in action and they think the robot is doing great job and they're very precise and repeatedly. But the problem is behind the scene there's so many hours of engineers and robot programmers are fine tuning those trajectory and with the process in mind and let the robot to do those meaningful processes. And those are like significant efforts and hours which you cannot imagine how many how heavy that workload is and to make especially for those high volume production robots and they have to run as a clock in the factory for three shift and 24 7. That is a significant achievement if you think about the efforts.
[00:08:17] Speaker C: E what is Mari's scan and go workflow? Could you explain that to the audience?
[00:08:22] Speaker B: Yeah, sure. So this is the part which our software is developed for is what we call is autonomous adaptable robot system. Which means we want the robot to have the sense and sync and act functionality and to do that we need to combine with the computer vision and force control and compact surface trajectory planning capability to overcome the limitations of traditional teaching or offline programming. So the scanned goal is to a summary of this workflow which is that we let the 3D vision to scan the environment, including the object, the part and then the robot generates the trajectory and our software direct drive the robot to follow that trajectory and do the meaningful process. So for example, if you're using our software in the sanding application, so the Robot will using 3D vision to see where the part is and then generate the trajectory, hold the sander and to follow this shape and to back and forth and to get the desired finish. So that is kind of a summary of this procedure of our software, how.
[00:09:51] Speaker C: It works and what are some of the proprietary technologies. And either of you can answer this question.
[00:09:57] Speaker B: Our solution, the AI part is actually leveraged from the physical informed AI. So if we follow the Nvidia press release at the CES and you heard this word called the physical AI and yeah, that's true because for robotics we need the physics information to be involved as the input. So there is a significant of AI section which is related to the physical informed information.
So what we leverage, that is to get the robot to understand the environment. And then we can generate a robot trajectory from the freeform surface via cloud point directly. Which means we do not need CAD model and we do not need hand teaching of the waypoint. The trajectory is generated from the cloud point directly. And our algorithm is very fast. It actually can generate this trajectory less than a second. So with this performance and we are able to provide those process, which is we call scan the goal and placement go to enable the seamless online automation. Which means the system can automatically adapt to a highly variant workpiece and eliminate the need for manual teaching or any manual adjustments.
[00:11:30] Speaker C: Thank you for that. Armin, do you have anything to add?
[00:11:33] Speaker A: One thing that I need to mention is that, well, people may look at what our system does, which is simply surface finishing or sanding, and think that, okay, it comes up with a trajectory and then just follows the contour of the surface. But the fact is that it is much more complicated than that. When you are dealing, for example, with a task that requires you removing 20 micron of coating from a surface, then the deflection of that surface under the pressure of the tool of the robot also matters. So one thing that I think is very interesting with the patented AI, you know, that he and his team have developed, is that actually the AI can't that into the whole process. So if it is dealing, for example with plastic, it knows that the plastic part is going to go through this kind of deflection. So it actually adjusts the pressure accordingly. So I will tell you more about that in in a case study that we have. But I think this is by itself amazing that the system can actually consider the material that it is working with and then provide great quality and consistent quality.
[00:12:47] Speaker C: Okay, so don't hold back now. Tell me a little bit about this plastic part, because this is fantastic that the system knows what it's dealing with. So maybe give us an example of some that application and maybe if there's some other ones out there too.
[00:12:59] Speaker A: One of the applications that I think actually is this ability really shines at is something that Murray did for me, Magna. So as you know, Magna is a Tier 1 automotive manufacturer and they build almost everything that is used in a car. So one of the many parts that they build are the class A plastic parts that they build for the body or interior, you know, interior of the car. So Sometimes the class A coating doesn't have the quality that they need and it needs to be reworked, meaning that they need to remove the coating of that part and go through the coating process again. And because every part is different and they do a lot of these kind of parts, they used to do it manually and there were people who actually did this highly, what should I say, dusty and tiresome work of removing the coating from, you know, the part that actually caused fatigue and probably the health risk. And they were looking for a solution and they actually went to other, you know, system integrators to find the solution. And they also looked into Mari's technology and Mari was able to provide them a solution for reviving the coating which is only 20 microns. And they were able to deploy the system in less than a month. And I think at 50% the cost of otherwise they had to pay to the system integrator. And it has been a great success.
You know, the rework of the parts has been fully automated. Now the scrap, scrap rate has been reduced and I think they have achieved 70% saving in labor and at the same time have been able to increase the productivity by 20% which I think is phenomenal.
[00:15:03] Speaker C: Well, that's exciting. Certainly those numbers for MAGNA are working out great. So what are some of the other trends e that you see in the industry?
[00:15:14] Speaker B: Yeah, so the trends I just read a report like from IFR 2 days ago and every year they will provide like some predictions about about the global robotics trend for each year. So this year and for 2025, the first topic is AI. So they said the trend towards AI in robotics is growing. That's of course because what we can see is by using various AI technology and robots actually can perform a wider range of tasks and more efficiently. So in our case, as you can see by combining with 3D vision and physically informed AI, it truly help the robots to understand and react in their external environments, which is quite useful in high mix, low volume production and also in the public environment as well. So for example, we mentioned about sanding on the vehicular parts as Armin described that also can be used for the car repair shops because that is a truly high mixed scenario.
Unlike magnet environment they already know what the part number is and what the type of the part. But in the car repair shop you can have a Toyota coming in today and the BMW tomorrow. So there's no way you can do offline programming with CAD model and those scenarios. But with the help about automatically path generation from the 3D vision and the cloud point directly and we can make it happen and to be automated in even in the car repair shop. So my prediction about the robotic trend is autonomous past Generation is generative AI for robots and it will create ChatGPT moment and for robotic and automation industry.
[00:17:27] Speaker C: Thank you for that.
Armin, we you had talked or we had talked a little bit earlier about an award that you guys are up for. Can you tell the audience a little bit about what's happening there? And we're taping in end of January. So I just wanted to give our audience a little bit about where, where we are right now in time.
[00:17:45] Speaker A: We know that we are among the three finalists in an upcoming event in Germany. Hannover Mess Robotic award is usually given to the best technologies of that year. And we have almost competition from almost all over the world because it's a very well known event and a lot of companies attend this competition. We actually got the exciting news that we are among the three finalists there. So we have one competitor among those three from Germany and another one which is actually an international company from us but also has a subsidiary I think in Belgium. So yeah, I mean fingers crossed. We are really excited and we will see who becomes the first. But we are really hopeful to bring that price on.
[00:18:36] Speaker C: Well, you know, it's I think very exciting when you're in the top three. Like it's exciting just to be nominated but when you're kind of in the top three, I think that that tells a lot about you guys and you're, you're still a bit of a startup, right? Like how long have you been in business now?
[00:18:51] Speaker B: Yeah, sure, yeah. We are a startup with experienced engineers and so Mario is about 4 years old and but our many of our engineers and programmers are in this industry for over 10 years and some of them and over 20 we're like sticking around in this industry. And for example for myself after my graduation from U of T in robotics and I joined the R and D Center which is have this opportunity to experience those challenges of robotics, this uncertain environment. And so those challenges just always like keep inspiring us and to think about the solution and how we can make robot more adaptable. And so that is kind of give us kind of always those challenges around us to find the technology breakthrough moment we really like at this moment. We'd like to appreciate the support from Engine which is the Next Generation Manufacturing Canada Fund and that is also support Mary at the very beginning and to initiate this breakthrough technology development from four years ago. And we are a spin off company and pharma engine and now we were actually making a product and the impact to the industry.
[00:20:34] Speaker C: That's great. Thank you very much for that. E. Armin, you kind of mentioned earlier in the podcast that ease of use is so instrumental. And of course you've got experience in this. Did you want to add on to anything there?
[00:20:45] Speaker A: We need to consider that no matter what we do, robot is not going to replace the humans. It's actually a tool. So it is a tool that is going to make a human worker more productive. So just like any other tool, that human worker should be able to easily use that system. So if that human worker thinks that, okay, this is too complicated for me and I cannot do what I want with it, then he probably is not going to use that tool. So it is like many tools that people buy and then it just sits in their shop or production line and nobody uses. So what I really learned is that you can build something. And I like to use the example of an iPhone that you can give to a person who is not computer guru and that person can easily start using it right away. Then you have done a great job. I mean, we have examples of use cases for Mari where people that have never seen or used the robot before are using it with their technology. For example, we have cabinet manufacturers that are actually using this product, which tells a lot about how easy it is to use. We are not talking about Magna, that has an army of roboticists. We are talking about a shop with less than 50 people who is actually using this, which I think is phenomenal.
[00:22:13] Speaker C: That's very exciting. And of course, I'm just about to pick up my car from a body shop because somebody backed into me and I'm kind of thinking, wow, I wonder if they had a MARI system in their body shop for the painting. I wonder if this might have been something to give me my car back a little bit earlier.
[00:22:28] Speaker A: And that's interesting that you say that, Jim, because, you know, we actually are currently talking to two of the largest chain, you know, car shops in Canada. They are excited. So we probably are going to start testing a pilot test with them soon. And what they told us was that what they are excited about is consistency of the quality because, you know, they say that, yes, we have these prep people and then we have these painter people. And, you know, depending on the mood of that person, the quality of the work that day differs. Now, you don't have that problem with the robots, right? I mean, that person, whether he's not feeling well or is, you know, thinking about something Else just can move this system because it is also mobile. You can, you can move it next to the car and just instruct it that hey, you know, remove the paint from this side or just finish it smoothly and go and do something else while the robot is still doing it. That is not to mention the shortage of labor, which is by the way very serious in North America. So yeah, I agree that when they hopefully adopt the system, you can get your car back way faster.
[00:23:44] Speaker C: Hey gentlemen, thank you very much for joining me today. Have we forgotten to talk about anything?
[00:23:48] Speaker A: I just wanted to maybe give the audience some statistics about how severe the labor shortage is. And adoption of these new technologies is an obligation regardless of whether we like it or not. So back in April 2023, the manufacturing openings, job openings was 668,000. It is anticipated that by year 2030 there will be 2.1 million open jobs. That is not going to be filled. That is huge. Today almost 77% of manufacturers are complaining about ongoing jobs worker shortage. Now, considering that 70% of the manufacturing workforce are working for high mix and low volume manufacturing, then we are talking about 1.4 million people or worker shortage in the industry by 2013. And technologies like what Mari has developed is going to help to bridge that gap.
[00:24:58] Speaker C: Those are some pretty frightening statistics. Thanks very much. It is how can our listeners get in touch with you?
[00:25:04] Speaker B: Yeah. So you can Visit our website maplerobotics.com and also follow up from the LinkedIn and we have a channel in LinkedIn if you search Maple, Advanced Robotics and also the YouTube so you can have a look about our scan, the goal and the place and goal in live demo. And also you can send us the emails from the inforaprobotics.com thank you so.
[00:25:34] Speaker C: Much for that and I'll put those in the show notes as well. Armin E. Thanks for your time today.
[00:25:41] Speaker A: Thank you very much.
[00:25:41] Speaker B: Thank you. Thank you, Jim. Thank you very much.
[00:25:44] Speaker C: Our sponsor for this episode is Earhart Automation Systems. Earhart 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. And Earhart's hard to spell. It's E H R H A R D T and their
[email protected] and I'd like to acknowledge a three the association for Advancing Automation. They are the leading trade association for robotics, vision and imaging, motion control and motors and the industrial artificial intelligence technologies. Visit automate.org to learn more. And if you like to get in touch with us at the Robot Industry podcast, you can find me Jim Beretta on LinkedIn. Today's podcast was produced by Customer Attraction Industrial Marketing, and I'd like to thank my team, Chris Gray for the music, Jeffrey Bremner for audio, my business partner Janet, and our sponsor, Earhart Automation Systems.
