Speaker 0 00:00:00 Are you squared strives to make the world a safer place using robotics, simply put we're developing intelligent mobile manipulation systems that empower humans to do their jobs faster and more safely.
Speaker 1 00:00:19 Hello everyone. And welcome to the robot industry podcast. My name is Jim Beretta and I'm your host. It's my pleasure to introduce you to Yorgen Peterson. You're going Peterson is president and CEO and founded our East squared robotics, a leading developer of human-like robotic arms in July, 2001. He is responsible for the strategic direction of the company, developing partnerships and alliances and overseeing the day to day operations. His mission is to develop intelligent mobile manipulation systems to empower humans, to do their jobs faster and safer. He's a leader in the unmanned systems community. He has served as the chairman of the robotics division of the national defense industrial association, the board of trustees for the <inaudible> and the board of directors for the national advanced mobility consortium. Mr. Peterson also is an advocate for robotics and manufacturing growth in Western Pennsylvania. He currently serves on the board of directors for catalyst connection and the technical advisory committee for the advanced robotics for manufacturing Institute, with the goal of fostering manufacturing growth in the Pittsburgh region. He also serves on the board of directors for the Pittsburgh robotics network. He has contributed to the growth of robotics market for the past 25 years, beginning his career as one of the founders of Carnegie Mellon's national robotics engineering center, the N R E C. He holds an ms degree in robotics and a BS degree in electrical and computer engineering from Carnegie Mellon university. Welcome to the webinar. We're glad to have you.
Speaker 0 00:01:54 Well, thanks for having me. I look forward to it.
Speaker 1 00:01:57 Can you explain how intelligent mobile manipulation differs from traditional robotic solutions?
Speaker 0 00:02:04 Sure. Most people, when they think about a robotic arm, they think of what are traditional robotic arms. And they think about it in two ways, both as the physical and the logical from the physical perspective, these robotic arms are excellent at what they do. They're very, they're large, they're bulky, they're power hungry, but they're very fast. And they're very repeatable, meaning that when you get to the logical side, it's very easy to program them and they can perform a task over and over again, very reliably, uh, and very accurately. And more recently you have the what's called the cobalt, which are smaller versions of that. Uh, and there, they also can be programmed, but they have limited capability as are. They typically operate around humans, but neither the traditional robotic arms nor more recent cobots are really, truly appropriate for mobile applications. And that's where we come in, where we are introducing this concept of intelligent mobile manipulation.
Speaker 0 00:03:11 So on the, on the physical side, what that means is that these arms are very lightweight. They're compact, they're power efficient. They can deal with shock and vibration, and they still retain all the qualities of the, um, the larger variants where they're still fast. They're very accurate, but they were designed from the ground up for mobile applications and I've already squared. We differentiate ourselves one step further by even allowing these arms to go outside and deal with the elements. Then from a logical standpoint, rather than having to program these arms, we are approaching it from the standpoint of, we want these arms to be able to perceive the world, understand what they're seeing and translate, uh, that reasoning into action and be able to deal with more, uh, dynamic or unstructured environments versus simply programming an arm to do the same action over and over again. So that's really the different view of what is a traditional robotic arm versus our, uh, intelligent mobile manipulation approach.
Speaker 1 00:04:20 Yeah, that's a great differentiation. Thank you. How did, how did you get started in intelligent mobile manipulation?
Speaker 0 00:04:27 Yeah, in the early years of the company, we were, uh, really contract engineering, but as time went on, uh, in, uh, around the 2006 timeframe, we were exposed to the, uh, small business innovation research grant program. And we were able to secure a grant that was focused on small, lightweight mobile, modular manipulators. And this was during the, the height of the Iraq and Afghanistan Wars. And they, these robotic arms were put on small track robots to go out and do dangerous tasks, such as explosive ordinance disposal or EOD. Uh, so we saw that those arms were doing a great job. They were, they were saving lives, but they were very primitive. And we knew that we could introduce more humanlike qualities into, uh, the physical as well as add more intuitive interfaces. And that's what we did in those early years. And since 2006, we've evolved and we have, uh, transformed these arms into these more intelligent, uh, and even more capable, uh, robotic arms that, uh, exist today, which really exude humanlike characteristics in terms of the way that they move in terms of their physical attributes. And in terms of the way that they think
Speaker 1 00:05:55 I'm just imagining the user interfaces that you were doing in 2006 and what here in 2020, that must be quite a different thing.
Speaker 0 00:06:03 Yeah, it's evolved significantly over the years. Uh, back in the early days, it was buttons and switches. Uh, then it evolved into like gaming controllers. Uh, today we still use, uh, joystick like controllers, but, uh, they're more like puppet arms, if you will, like scaled models of the arms that you're controlling, such that they provide a very intuitive interface. However you move your arms is exactly how the remote arms work. So it feels like an extension of yourself. And the way that we measure intuitive is in two ways, one is training time, which is usually under one minute of training. And then two is time to proficiency, which is generally one to two hours of stick time, if you will. And you're performing very complicated tasks using a highly dexterous two arm system, uh, as if it were an extension of yourself,
Speaker 1 00:06:58 That's a great way to put it. I really liked that that tack time to be using a robot. So what industry sectors does Ari squared operate in?
Speaker 0 00:07:06 Uh, we support multiple industry sectors, but, uh, there's three in particular that represent the lion's share of our focus. Uh, first we, as you, uh, I mentioned, we got our start with the EOD application helping to diffuse IDs and roadside bombs using manipulators. So that was in the defense department and we still serve the defense department, uh, both the ground and subsea on the ground side. We're, we're still do that bomb disposal mission, but now we're into refueling aircraft. We are into helping to repair a runway. We are performing complicated tasks, such as atonomous the identifying casualties on the battlefield and helping to extract them into, to safety. But in more recent years around the 2018 timeframe, we started getting involved in subsea applications. In particular, we, uh, have developed a robotic system that can autonomously neutralize mines on the aviation side. There's an analog to some of the defense work.
Speaker 0 00:08:16 You know, I mentioned how we're refueling aircraft on the defense side, uh, in the aviation space, we are also doing similar type tasks related to aircraft maintenance and servicing. Uh, I can't get into too many details, uh, given, uh, the sensitivity with some of our partners. Um, and then the third market that's, uh, really ramped up this year is in the medical space. Similarly, I can't get into a lot of specifics there, but what I can say is that we're developing surgical robotic arms for a major medical device provider. Uh, and it's gonna provide a new capability that hasn't been before. We're going to be producing robotic arms that are lighter, uh, at a more intelligent and are able to perform tasks more effectively than, uh, anything that's currently on the market.
Speaker 1 00:09:07 Well, that's very exciting and it's really interesting to be in all these new markets. And there's something kind of nice about creating like a quasi medical device or a surgical device or whatever that can save lives.
Speaker 0 00:09:20 Yeah, that's right. And that's truly, you're getting to the essence of RA squared. We are a company that's founded on making a positive impact on the world. That's one of our guiding principles. Everything we do is truly about either saving lives or improving quality of life. That's the theme. And that's the, the driver of who Ari squared is, uh, you know, we are looking to make the world a safer place. And, you know, we're doing that by developing these intelligent mobile manipulation systems that can both empower humans to do their jobs better, but allows them to also remain safe.
Speaker 1 00:09:57 Thank you. You're good many applications that you're developing require specialized end of arm tools. So what's your approach to that?
Speaker 0 00:10:06 The lion's share of applications require a traditional gripper, which is a two finger gripper and that, that candles usually 90% of the applications. And that's what you see primarily in the, in the marketplace, whether it be in intelligent mobile manipulation space or in the more traditional industrial automation market. And the reason is that your thumb and your index finger represent the lion's share of your capability as a human when you grasp. And then you add more capability in terms of conformal grasping and secure grasping as you move from your index finger out to your pinky, but the Pinky's really not providing a significant impact, uh, on, uh, manipulation tasks. So what we have done is said that in many cases, we'll directly integrate a two finger gripper into our robotic arms, but along the way, we took the approach of looking at it from a modular standpoint, that we can essentially take the hand and remove it, uh, or put on a different type of hand, whether it's go from a two finger gripper to a three finger gripper or go to a specialized tool by using this quick release mechanism and an industrial automation space, you see usually a new Maddix style, quick releases for changing out tools, but because we're in the mobile manipulation space, we are all electromechanical.
Speaker 0 00:11:37 We re, we typically aren't carrying around an air compressor, uh, to tool change. So we've developed a lot of IP over the past decade and a half related to how to change out tools or end effectors at the end of an arm. And we can do that either through a mechanical slash electrical power takeoff, or we have variants that have an all electric power takeoff. And additionally, you can have those interfaces be manually changed out, or you can have a, a tray of tools if you will, and carry that around with you like an electrician moving through the world, right? You have your tool belt, and then you can reach into that tool belt at any time, grab the tool that's optimal for the task at hand and put it in your hand, whether it's a hand grabbing a tool, or whether you're just, you know, changing out your hand for a drill or another type of tool. So this, uh, quick release mechanism, uh, is utilized across our complete portfolio of robotic arms. Whenever you need to be able to change out the functionality and have an optimum solution, uh, for the task at hand,
Speaker 1 00:12:56 It's like your own technical tool belt, right? For grippers. That's exciting.
Speaker 0 00:13:00 Yeah. It's a, the Swiss army knife of a robotic arm, uh, and of tools.
Speaker 1 00:13:05 Well, you, you've done some really exciting and interesting projects. What's the most difficult autonomy project that you've worked on so far.
Speaker 0 00:13:13 Yeah. So we've worked on a lot. We push ourselves and we do things too that people haven't done before. I think our most challenging one is one of our most recent. And I talked about it earlier, uh, in the, in the podcast is our ability to go underwater and operate and manipulate objects underwater. So we have a program with the office of Naval research ONR, and this program is about locating mines, underwater, going up to the mine and then neutralizing that mine. And what makes this autonomy a problem? And we successfully demonstrated that, uh, just a few months ago, the ability to autonomously neutralize mine, what makes it so challenging is that one you're underwater, right? So you could have objects floating around you. You could be interpreted murky waters where you can barely see, but then you have the added complexity of you're on a underwater vehicle.
Speaker 0 00:14:22 That's constantly moving. So you have direction every direction. Uh, so you can have waves that affect your positioning. You could have currents, uh, that affect, uh, your positioning and you have to constantly adapt and close the loop on what you're seeing. But what allows us to tackle this problem is we have two core technologies that we have in our library that we can pull out to help us with. One is our RA squared detect, which is the computer vision algorithms, uh, that is a library of computer vision algorithms that we can, uh, use in tandem to improve the reliability and quality of what we believe we're seeing in the world. And when you use multiple algorithms, instead of just using one algorithm, you get a much more robust, more reliable output. Then after we've perceived the world, and we have an understanding of what we're seeing.
Speaker 0 00:15:21 We have RD squared intellect, which is the other technology that we pull out. And what that does is it overlays artificial intelligence to help us reason about what we're seeing, and then translate that reasoning into actions that our arm has to perform. But as it's doing that, we have to go right back to our Ari to detect because the world is constantly moving. So we need to make sure that our gripper or hand or whatever the tool is at the end of the arm is achieving what we expect to achieve in real time and close that loop constantly because the arm is moving and the vehicle is constantly moving in this, uh, underwater environment. Uh, but we successfully overcame that autonomy challenge, uh, just a few months ago and are excited to push the envelope of what's possible now that we have that amazing baseline.
Speaker 1 00:16:18 Wow, that's so exciting. What you're building so many robots and different creative applications like underwater and nuclear working in the rain and snow. How do you retain your knowledge base?
Speaker 0 00:16:29 Well, our knowledge base is our people, right? The, the strength of a company, uh, are the people, uh, that is what we focus on. And we're lucky that the people that we have, our true robotics, people that have a breadth of knowledge where they're not just spit, you know, specialists in one area, they to be in robotics, you have to have an understanding cross-discipline mechanical electrical software to understand how all of those disciplines relate. And then once we have the people in house to do it, we gotta make sure we keep them right. Uh, and we do that by creating an environment where, uh, we, uh, have our core values and we treat, uh, we have trust, respect, integrity, and positivity. We have guiding principles where we value our people. We empower our team. We innovate through robotics and all for the common goal of making a positive impact on the world. And that really binds, uh, our group together and focuses our group on, uh, achieving that vision of making the world a safer place. And that that's how we retain that knowledge bases by just retaining the people and keeping them focused on really hard, challenging problems in a very healthy environment.
Speaker 1 00:17:58 Well, I was going to ask you about attracting or retaining talent, but I think it's all basically coming from top down, right? Here's what we stand for. And we do what we say
Speaker 0 00:18:08 That's right. I, I truly believe that if you focus on people, everything else works itself out. If you focus on creating an environment where people love going into work, every day magic happens. I've learned over the past 19 years of running R squared. And especially in more recent years, that if you focus on your core values and your guiding principles, that that's how you build a strong culture. And if you focus on people, you'll have tactical success, you'll have innovation, you'll have growth, you'll have profitability because a company is it's people, right? And so that's the approach I've taken. And at first, uh, you know, I thought, well, maybe that won't scale as we continue to grow, but I haven't seen that. I've seen that as we went from 50 people in the beginning of the year to over 75 people, uh, today that the culture is the same that the people love coming into work. They love to be here. And I, and I think that's what attracts people from the outside is we're solving some of the world's hardest problems related to manipulation and computer vision and autonomy. But at the same time, we have a culture that rewards comradery that welcomes that feeling of belonging and that there, there are no dumb questions and that you can truly do whatever you want here. And that's what creates our success.
Speaker 1 00:19:38 And also I was going to ask you again, how do you create that culture of innovation, but it's kind of like it creates itself, right? You take on really, really difficult challenges and in some of the worst environments possible. Yes.
Speaker 0 00:19:53 So it's hard, right? From one perspective in that, Hey, we have specific obligations and contracts and deliverables, right? So you want to just put your blinders on from one perspective and just focus on that and don't be distracted by anything else. Right. But I think it's falling too old to always have those blinders on you. If you're going along and you find a new way of approaching a problem, even if it's out of scope of that project, talk about it, you know, bring it up, raise it up through, through the company. And if you pursue it, that might be your, your next differentiator in your, your next year's product. So we really encourage people to communicate and to collaborate and to be creative. Uh, we have programs that are, you know, we're throwing stuff at the wall and seeing what sticks. And we have some programs where we're just turning the crank and shipping out product.
Speaker 0 00:20:56 Right? You have to have that spectrum. If you go all one way or the other, you know, if you go all creativity, you're never going to get anything to the market. If you go all product, you're going to become irrelevant before you know it. So you have to maintain that spectrum. And then that's exactly the environment that we've created, where we have these geniuses who are coming up with brilliant ideas, and aren't afraid to communicate it and introduce it. It doesn't mean we can pursue everything, but having a culture where you encourage it, uh, will create, you know, success down the road.
Speaker 1 00:21:34 And what's on your kind of short-term radar. You know, I keep hearing machine vision, and I'm a big believer that having this great machine vision experience and creativity is probably one of the strongest assets that you have. Is that true?
Speaker 0 00:21:50 Yeah. So we're, we're, uh, excuse my nerdy term, but we're approaching the asymptote of what's possible from the physical side right there. The, the level of innovation that you can achieve there is, is diminishing just simply because we're still using rare earth metals and motors and gearboxes, right. Until there's some new breakthrough there, we're in an optimization path, right. And we're on the leading, uh, we're the pointy end of those, the pointy end of the spear there, for sure. But the, the, the level of impact that you can have, there is not going to be as significant as a level of impact that you can have by introducing computer vision and AI. On top of that, they're all extremely important disciplines and are important to the growth of the business. But what are I'm seeing is now taking the core IP that we have, which is this incredible power, dense, manipulation capability that can work anywhere on the planet, right.
Speaker 0 00:22:49 It's actually what we have now and making it think more like a human. I think it's going to be a really long time before we have true autonomy, but between between now and then really the world that we're going to live in is this supervised autonomy where there's a human still in the loop and how involved in the loop that person is, is to be determined. Uh, you can have a person that's, uh, checking in and, uh, verifying actions that a robot takes on every step, along the way, you know, one step up from directly controlling a robot, or you can have a human in the loop that is only there for those rare edge cases, uh, or errors that come up where the robot doesn't know how to resolve it on itself. And there's, that's the spectrum that we're going to live in for a long time, uh, until we get to that full autonomy, which is many years out, but there's significant benefit to realized between now and then in the supervised autonomy world.
Speaker 1 00:23:50 Wow. That's really exciting. And, uh, thank you very much for this. You're going. I'd like to thank you for taking time out of your day to chat with our audience. How can people get in touch with you
Speaker 0 00:24:00 And the best way to reach out to us as go to our website, uh, or go visit us on LinkedIn? Our website is www R E squared.com. So it's R E S Q U a R E d.com. You can follow us on Twitter at RA to robotics, or you can just type in RA to space robotics into LinkedIn, and you can find us, and we'll be happy to talk to you, especially if people are looking for a job where it seems that we're in a constant hiring mode these days. So, uh, we we're, we're constantly looking for new talent that, uh, aligns with our values and are ready to take on some really hard technical challenges.
Speaker 1 00:24:42 Well, I can't imagine that you've probably got one of the most exciting sandboxes, uh, on the planet for when it comes to robotics and automation.
Speaker 0 00:24:50 Yeah. It really is exciting to be on the cutting edge of what's possible in terms of intelligent manipulation. Uh, but that's exactly where we'd like to reside. And, uh, we welcome anyone who wants to join us to go pioneer. What's possible.
Speaker 1 00:25:05 Thank you. I would like to thank and acknowledge our partner, a three, the association for advancing automation, 83 is the umbrella association for the RIA AIA MCMA and eighty-three Mexico. And these four associations combined represent almost 1300 automation manufacturers, component suppliers, systems integrators, and users, research groups, and consulting firms throughout the world that are driving automation forward. I'd also like to thank and recognize our partner painted robot painted robot builds and integrates digital solutions. There are 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. If you'd like to get in touch with us at the robot industry podcast, our email is the robot industry
[email protected]. Or you can find me Jim Beretta on LinkedIn. We'll see you next time. Thanks for listening and 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 Colvin for audio production, my partner, Janet, and our sponsors a three and painted robot.
