Podcast

The Future of Cobotics

Mike Ouren, Cobot Guru and OEM Manager at Thermo Fisher Scientific and founder of Kevin Robot, joins Jeff Dance to discuss the potential of cobotics, current applications, advances in software and technology which make them safe to work alongside humans, as well as their potential impact on the future job market.

Mike Ouren, Cobot Guru and OEM Manager at Thermo Fisher Scientific

Mike Ouren: You can drive a robot today through any arbitrary trajectory in space, passively or actively, with this control technology, and it really kind of, its underutilized at this point. It’s kind of like a solution without a problem. But I think the problem can be brought to the solution in this case.

***

Jeff Dance: Welcome to The Future Of, a podcast by Fresh Consulting, where we discuss and learn about the future of different industries, markets, and technology verticals. Together, we’ll chat with leaders and experts in the field and discuss how we can shape the future human experience. I’m your host. Jeff Dance.

***

Jeff: In this episode of The Future Of, we’re exploring the future of cobotics. We’re joined by Mike Ouren, a cobot guru and expert, an OEM manager at Thermo Fisher, and also the founder of Kevin robot. We’re here to explore the future of cobotics.

It’s a pleasure to have you with me on this episode.

Mike: Pleasure to be here, Jeff.

Jeff: Good stuff. Hey, if we can start with telling the listeners a little bit more about your experience, particularly with cobotics.

Mike: Sure, absolutely. I was lucky to find my career before I went to college, so I went around a car factory as a tour, and here in Germany, in fact, that’s where I am now. And I had the opportunity to watch a bunch of robots putting cars together, and the rest is history. I was hooked all the way through college, went off and found a way to work for a robotics company immediately, right out of school. That was great. And along the way, you find yourself working with customers. If you’re working for a robotics company, I’d say one of the coolest parts of the job is the fact that you get to see so many different applications. A lot of things people never get to see. Laser welding or laser cutting or even applications that are just purely R&D using articulated robots. So I had that opportunity. Well, one of the applications I was called into was, at the time, a little startup called Restoration Robotics. People always confuse this with Restoration Hardware, but I’m hinting at the fact that I ended up working for a customer, which is exactly what I did. I left the manufacturer and went to a customer. But the idea is Restoration was a hair restoration robot using an actual industrial robot, not even designed to be a cobot, but it was adapted to an application as a medical device. It’s actually a totally different take on cobots, right. Where you’re actually using an industrial robot with no specific collaborative features. And because it’s in a medical device, you can hone in the exact use case of the robots to the point where it is a cobot at that point right. A doctor is working directly next to this thing in an operating environment. It’s an outpatient procedure. Hair transplant. It’s an outpatient procedure where you’re harvesting follicular units or hair follicles from the back of the head for transplant later to the front. And this thing is actually dissecting sub millimeter precision right there while the patient is breathing and moving a little bit and tracking all of the vision systems. So that was my first exposure to collaborative robots and from that point, laboratory automation. Great use case for collaborative robots. And I’ve just continued in that field from that point.

Jeff: Nice. For those that don’t know, Thermo Fisher, as I understand it, it’s a $240B company. It’s like one of the larger companies in the US. Do they have their own cobots as well?

Mike: Yes, we do. We have a series of collaborative robots that are used for primarily laboratory automation applications and combination of research as well as what I would call production applications. Anytime you’re doing batches, you can do batch processing of samples. They typically are done on Microtiter plates. And our robot is optimized for driving that process of moving the Microtiter plates from one station to the next or one instrument to the next.

Jeff: Got it. I noticed you’re also the founder of a co-founder of the Kevin robot. Tell us more about that.

Mike: Sure. Kevin robot was really a few of us that were involved in bringing Kevin robot to life. So I actually got to know the Fraunhofer Institute in Germany over the course of my career as I was working in the life science industry. And there was an opportunity to work together with them at a higher level with bringing to life this idea of combining technology. So Fraunhofer is a bit of a technology incubator, if you will, where they’ve had spin off companies in various areas of both mobile robots as well as mobile robot navigation and collaborative applications. So they have two spin offs that they’ve already spun out. They’ve become companies that are based in Stuttgart, one of which builds a mobile robot platform. So a base that can essentially use omnidirectional wheels to navigate in any direction. Right. The advantage of omnidirectional is you don’t have to necessarily steer like a car. You can just shift left, right, forward, back at any time. So you have that additional flexibility. That’s very handy to have in a tight space like you might have in a laboratory. That’s key number one. Number two is the navigation company. And those don’t necessarily grow on trees like software companies that have built out all the infrastructure to take all that sensory data and navigate into a room repeatably. That’s also another company that’s spun out of Fraunhofer, taking those technologies, combining them into a robot with a mobile robot with a robot arm attached to the top. Now we have the ultimate flexibility to move things around in the lab intelligently. And that is the nature of Kevin robot. The ability to set up an experiment on the fly with different instruments in your lab and you might just decide right before you go home for the night. I’m going to do a quick reader load out of this array of plates and move it over to the reader on the other side. It’ll be done by morning. I can even check it online to see what’s happening. That’s the whole basis for Kevin robot.

Jeff: Nice. Sounds really useful. Awesome. Well, let’s dive into the topic of cobotics. If we could start with kind of the one on one, a little bit of the present and then let’s talk about the future. Just for those that aren’t familiar with these types of robots. What is a cobot?

Mike: Well, the term robot we already know has been kind of used for a lot of different things, from movies to even a chatbot. Right? I’ve heard of chatbots. Right. It could be a software thing. Originally a robot was intended to be a mechanism that autonomous, right, in some way. And a cobot, that term is still quite specific. It’s a relatively new term utilized. I haven’t seen it kind of go off into a tangent area yet as far as what’s being referred to as a cobot. But a cobot is a machine that’s really approachable by a human being. It also has an element of being easy to use. There’s kind of two categories of cobot, right? You’ve got working with human beings or working alongside human beings. It may be fully automated and fully autonomous doing its own thing. But if a human needs to work or interact with it, cobots allow that to happen with less fewer steps involved, transition from autonomous to non-autonomous operation. And then the other aspect is the ease of use, I think. And that’s the one that kind of gets lost, right? What also came with cobots was better software to define an application with fewer steps and also hone in on exactly what the user wants to do more quickly, but also integrate different technologies like grippers and things like that into one system. Also less overhead, less technical approach.

Jeff: I think this notion of interaction is for a lot of people understand that’s like a new paradigm. Yes. In movies we see R2D2, C3PO, we see these like great examples of robots that sit in humans. But it really hasn’t been the history of robotics. Right. In business at least, there’s been a lot of consumer products that have failed 99% failure rate. But this is actually relatively new. Am I correct in saying that, like cobots and like the prevalence of them actually growing in our businesses and in our workplaces, it’s more of a newer trend if we think about testing things versus the reality of using things.

Mike: I would agree with that. I think cobots are inherently more difficult to find that quote, killer app. Right? In the original, the original robot space had two primary applications. It was packaging, high speed SCARA, cookies, you name it. Food packaging. It’s really been driven by the SCARA robot. It revolutionized packaging and batch processing applications. And then in the automotive space, the articulated six degree of freedom robot was dominant. But when you can put a box around something and say, inside that box, I can count on humans not being there, then you’ve got basically carte blanche to go do whatever you want to do inside that box. And you can do it fast, you can do it efficient. Cobots are a little bit more vague in the killer apps department, right? How do you apply that? You can’t go as fast generally because the laws of physics apply. If you have a fast robot, it weighs a lot, it’s going to hit somebody. Doesn’t matter how sensitive the joints are, its mass meets other mass. And then the other aspect is the organic nature of applications. People say, I just want to make a cup of coffee in the morning. Robot people like me go, do you realize what’s involved in that? Opening the door, putting the coffee, getting the grinder, all that stuff. I mean, there’s a lot involved in that. And that’s why I think it tends to kind of go into shoot high, don’t necessarily aim low, but be prepared to simplify. At the end of the day, that’s kind of the path the cobots have taken, in my opinion, so far.

Jeff: You mentioned life sciences. What are some other if you think about the present day, what are some other applications of cobots that we see in the workplace?

Mike: Well, I think bringing the right tools to the right place at the right time, if you’ve got resources that you need to bring around. This is really not even getting into having robotic arms attached to mobile bases, but I think mobile bases are getting better and better where you can put things on top, where you’ve got maybe a set of tools. The hospital industry has made huge use out of mobile transports inside of hospitals for prescription drugs that can be inherently dangerous just to even handle. So here you can put them in a locked away robot, but just delivery to the right place at the right time has been one of the huge application. And it’s not a cobot in the sense of a robotic arm, but mobile robots are cobots too, right? If somebody stands in front of one, they go around, right.

Jeff: They’re working with humans, right. They’re in and around humans.

Mike: Absolutely.

Jeff: They’re assisting, but they’re in and around humans versus locked away like a box.

Mike: Totally. Logistics. I see the logistics industry getting more and more intelligent and the application level pushing higher and higher because it’s a lot of the day to day work and a lot of tasks is just moving things around from one place to the other, whether it’s tools or raw materials used for things.

Jeff: Some stats and trends have been showing that as a percentage of robot sales, cobotics has been growing quite a bit in the last five years. What do you think is the key driver behind that trend. What are some things that are making that helping cobots grow?

Mike: It’s twofold. Demand is there, so you got to have that. And the market is now educated. They understand what you can do. When I first started in the space and really started following cobots, it was like, really? You’re going to do that? And people were rightly skeptical. It’s like, this is just an industrial robot that’s called a cobot instead of a robot. Right? That’s really what it felt like to me early on. And it’s slower yet it’s easier to program. Then people started figuring out, well, you don’t necessarily need full on speed for all applications. And in fact, there’s a whole subset of applications where people have just traditionally ignored it because they know they’re going to have to build out a whole infrastructure, a cage, sensors, safety technology, almost a two to one ratio in terms of costs. You buy the robot for 25,000, be prepared to spend another 25,000 just to put it in the box, and then you got to program it on top of that. So I think over time, people realize there’s a whole bunch of subset of applications that were skipped over just because of that. And now, like I said, these machine tending applications and also laboratory applications where you just need the robot, a gripper and maybe some storage, one fifth the cost of all this caging and safety technology. And you can put it right next to the instrument and run it. Whether it’s a machine tool or a laboratory instrument, it’s basically the same application. It’s machine tending. One might be 5kg, other one might be 500 grams. That, I think, is what opened the door big time. People saw that there was an awareness around it, and the doors were open to start doing it. It was always there from the beginning. Those applications were always there. They just got passed over because of all the requirements.

***

Jeff: Periodically, we bring in other experts and industry leaders to get their take on the future. They were not involved in today’s recording, but here’s what they shared.

John Houston: Hi, I’m John Houston, and I’m the lead robotics engineer here at Fresh. Cobotics is a really broad term, and it’s often visualized in arm-based scenarios. But it’s more than that. I think that cobotics is really going to take off in the realm of human augmentation. A vision system, for example, that keeps a heavy equipment operator from running over a person accidentally is just as much of a cobot as an arm helping to assemble cars on a factory floor. We don’t need to take the human out of the equation completely. Anything that makes the human more efficient, safer, et cetera, is a huge gain. And to be limited by this false barrier of full autonomy before anybody sees it right, is silly. And I think that it’s really limiting the development of a lot of great technology.

***

Jeff: Obviously a lot of the robotics technologies advance, but be able to work alongside humans and know that, hey, that could be safe. Now, what do you see as some of the key technology that advanced, that made that even easier or faster or cheaper?

Mike: Well, two things. One, the software has gotten way better. There’s something that revolutionized the mobile telephone industry. It’s called the software defined radio. And I would say in robotics we have the software defined robot controller, where you now have the ability with a robot controller to go in and control almost every major control parameter for the joints, how they’re controlled the force management, being able to mitigate collisions and things like that, even from ten years ago. Processing power is so good that you can do current loop, you can do voltage loop, all the dynamic control at very high rates in real time with something that costs not that much. And that is really revolutionized, like how you can bring this technology to market with a relatively reasonable price and the software to do it, to program the robot has come up so well that you don’t necessarily have to have a programming background to work with these machines. You still need to work through the process of what you want to do with the robot and some of the mechanics of it. But the training, the barrier to entry on training is way lower than it used to be. In that case, I think it’s one of the big factors that has a little bit to do with the hardware, but it’s really on the software side that I think has brought it to market in a bigger way.

Jeff: And they said the future of robots is probably software, right. As price comes down, as all the types of robots become more available, I’d see that continuing to advance. I guess you think about our smartphone, right? Once you have that platform, it’s like the proliferation of the apps made those things, oh, definitely just way more useful. It’s interesting that on our team, we’ve been working on where you are, this notion of where you are in an indoor space as well as where you are from outdoor space. And we would focus on small degrees, like precision, accuracy, but that’s actually been a hard problem to do well. And you had mentioned the institute in Germany that has some serious software about navigation, kind of where you are and that sort of thing, and then the usability software about programming things. What do you want it to do? But I think the mobility and navigation of where you are, I think we’ve been watching that advance for the last five years. And I think as we think about robots running around us and getting to where they need to go and then picking something up and going someplace else, right, that seems to be a major advance as well to support cobotics that aren’t in a single space right?

Mike: Definitely.

Jeff: You’re seeing that right in Life Sciences, right? Like a robot being able to go to another place accurately and grab something that’s come a long way in the last ten years. Right?

Mike: Yeah. That was really what got me so interested in the Kevin robot concept was for years I’d seen in academic papers I’ve been following the robot industry ever since the mid 90s. And the Robotics and Automation Journal. There are all these applications for mobile manipulation but the biggest kind of the breaker and the whole deal to get it commercialized was always around resolving the last few centimeters, essentially the last inch, let’s just say. And it’s like the last mile in cable and fiber optics. Same thing. It was the last inch in mobile robots that always basically made a robot arm attached to a mobile base less useful. So this idea of where we’ve got camera technology that’s in the thousands of dollars it’s embedded, I don’t need another processor for it or another computer. I can put it on board a mobile platform and really get full power out of the robot arm that’s sitting on top of that mobile platform. That’s a game changer. And doing that in real time from one station to the next without having to really train the user any differently on how to teach the robot they can run this process and not worry about it. So that’s definitely a huge game changer to take it to the next level.

Jeff: Who are some of the biggest kind of cobotic producers? I know Thermo Fisher does have some product lines around this that kind of focus on Life Sciences seems like. But who are some of the other big kind of cobot producers in the world that you think we should be watching?

Mike: There’s two categories there. There’s companies that were built on the foundation of being cobot companies to begin with. I think the number one is going to be Universal Robots out of Denmark. No doubt about it. They’ve grown and I’ve watched them grow in time. Another one is my former employer, used to be called Precise Automation, now owned by Brooks. Brooks Automation is famous for semiconductor robots. We were purely a collaborative robot company right when we got into that space, just because of the life science industry primarily. That company was actually founded by some of the original people in the robotic space who started an industrial robot company. So we saw both ends. Brian Carlisle and Bruce Shimano founded Precise and so that was really cool to work with them because they themselves learned a lot transitioning into collaborative only. And then you’ve got the industrial companies that are still primarily known for their industrial product lines KUKA, FANUC, ABB but they’ve created what they call cobot product lines. I still don’t see any of those actually being a real dominant factor. They all kind of share a little bit of pie, all adds up to a fairly large piece, but nothing like what I think U.R. has done. And a couple of other companies that, Techman is another one which works with depth technology. A lot of you also see a lot of similar robots coming out of China that don’t really have a major play in the North American or European market. But if you just look at Asia and APAC alone, they actually start representing a pretty significant quantity of robots that aren’t even exported.

Jeff: Let’s transition a little bit to the future as we think about different industries that might have some of the most rapid adoption. Do you have any perspectives on where we really see cobotics taking off?

Mike: Well, I see the food service as one that’s interesting for cobots. I’ve been watching that for 20 years, and it seems like finally now there’s enough interest from the companies to do that. I actually filled out the application to work for McDonald’s Engineering years ago. They actually have a full-on engineering group. Everything you see in a McDonald’s is purely engineered for McDonald’s. That’s it. Right? So I think that kind of attitude or that approach lends itself to doing more automation. It’s just taking a long time to kind of adopt it. So that’s one area that I think can finally grow a bit. The technology is there, and the interest is there. I also see for me, the laboratory is just in the beginning stages, right. It’s a low payload applications are some of the best applications for cobots. Right. If you’re moving 50 kilos, that’s probably not a good cobot application. But if you’re moving five or less, that works out well because you don’t have to build a mechanism that’s heavy. And you can use control methods that lend themselves to speed and still remain in collaborative at the same time. You can make anything collaborative. It just depends on how slow you want it to go. So I think the low payload electronics market is another one. Electronics is a lot like the lab. You do a lot of testing, you do a lot of batch processing. So electronics test is a great fit for collaborative robots. There’s already been several applications. You don’t see them as much. They’re not as, I think, visible, but they’re definitely happening there behind the scenes and growing. Yes. So there’s a couple of examples I see as future drivers for cobots.

***

Avinash Singh: I’m Avinash Singh. I’m a senior robotics engineer at Fresh Consulting. Farming is another interesting area with potential cobotic applications. We are seeing more and more cobotic applications on account of increased dexterity of the robots. With the rise in vertical farming and other botanical advancements, we should see an increase in cobotic applications in these areas.

***

Jeff: What do you think will be one of the big steps forward for cobotics? Are there some technologies still that obviously we’ve had this rapid confluence of technology in the last 15 years coming together to make things faster easier for robots in general, and cheaper. What else do you think is sort of on the verge that will help things really grow? Or do you think we already have the technology, it’s just sort of a matter of time?

Mike: Well, I equate it to the autonomous car industry right there’s sort of the last 1% of that, you know, we see, oh, 99% it’s fine, 1% not so much. So that 1% is going to take a lot of time. I hold fast on that. That opinion about the the whole awareness, the awareness around the the machine, whatever it is, whether it’s a car or robotic arm. I think awareness is another area where the next step from collaborative robots where you’ve got better tools, kind of like the tools that you have that help you drive a car better. The lane management, the thing that beeps at you when you’re closing rates too fast with the car in front of you or with the car behind you if you’re backing up either way. These awareness and mitigation technologies took years to get installed in the vehicles, but they’re tremendously helpful, I think with robots awareness technology and the ability to do more sophisticated auto teaching also basically giving the user the ability to present the part in a less precise way and still get a reliable outcome. This area is still very kind of top-down approach. Right. Your machine vision is it depends on a lot of things being right before everything else works out right. And I think getting that fewer things that you have to get exactly right speeds things up at the end of the day. Right. You can more fluidly go in and move to the next step in any given process. And I think it’s just a connection to the human if it’s a cobot right. You’ve got human interaction, it’s part of the process. The setup process is also part of the process. Even though you’re not doing it as often. That’s an area that still could use a lot of work right. Make it easier to use the robot, implement the robot.

Jeff: Interesting to think about, maybe not just the robot itself and advancing it’s, all that contextual stuff around you and how the robot interfaces with its context. Right.

***

Human interactability and usability is an often overlooked but crucial aspect of cobotics. We need to start looking past the simple buttons and screens to more of a human form of interaction. For example, if I were going to hand something to a person, they would see what I was doing and instinctively reach out to receive it from me. I think that by making these robot interactions intuitive, we open doors to more robotic applications.

***

Jeff: As we think about the future, like ten to 20 years from now. What do you envision if we jump forward the next 10 to 20 years?

Mike: Well, I think the electromechanical components used in robots, there’s definitely some improvements to be made there. We’ve been really working off of harmonic gearboxes and robot arms for a long time. They’re kind of, in a lot of ways, clunky, high reflected inertia and what I mean by reflected inertia where you’ve got essentially an inability to sense what’s happening on the end of the joint at the motor level. A lot of cobots that have any payload and sensing technology, unless it’s direct drive, unless there’s no gearbox at all, you have this isolation, and it just makes the control more clunky. Right. So direct drive is really going to the next level right now. There’s a few technologies that have come on the market where you’re able to get more power out of a direct drive, where you can eliminate the gearbox on major joints. And a robot seems like a small thing, but not having this 50:1 gear ratio is huge. We’ve seen it with electric cars just in the drive mechanism. The reliability of an electric drive, where you’ve got no gearbox, you’re getting amazing torque, amazing power, no gearbox, no transmission, all these valves and things. A gearbox is the same way. If a gearbox gets struck, if you get damaged from a collision, harmonic gearbox is toast. Right? It’s basically you got to pull it apart, replace it. And I think those are things that people think about, like what’s the maintenance cost of this machine? Right. I think for cobots, because the payloads are so low, you can eliminate gearboxes with the technology, but we’re already seeing it. There’s some robots on the market now that are using major joints that have direct drive. So I think that’s a big step. It will simplify the design of robots and give people a lot more flexibility and then also shrinking the controllers electronics. We’ve already seen it, but most of the cobots, the large players in the cobot still have, I would say, a small box, maybe a 5U rack-mount computer box that’s still attached to it with an umbilical cable, all that, that’s going to go to across the board, I think. I think people will be able to embed the controllers. There’s a few players that are doing it now, but from a cost standpoint, to be competitive, you’re not going to be able to sell a robot plus another box if the trend in the industry is moving away from that.

***

Avinash: Aided by the multifold increase in sensing and computing capabilities, we’ve seen a rising focus in areas of general purpose robotics. Research groups and companies are focusing on creating robots with capabilities to handle multiple tasks and scenarios by themselves. For example, a robot would know the difference between handling a delicate piece of kitchen equipment to opening a door versus picking a heavier object. There’s also focus on getting a robot to be able to navigate in a small space without the need for an extensive mapping mission or extensive technology or technical support. All this points to a robot that would have the capabilities to aid in general tasks. With Japan funding companies to create robots specifically for elder care, we can expect to see a rise in these types of cobots that help with daily household activities and aid the elderly with picking up, let’s say, heavier stuff and making their lives a bit better.

***

Jeff: Do you see cobots replacing any jobs in the future? Is it just helping us be more productive? So maybe one person doesn’t need the two hire as many people because they have more support, they’re more productive themselves, they need thoughts. A lot of people fear robots and they hear cobots. It’s like, oh, it’s maybe like a multiplier of fear. But if you think about dirty, dull, dangerous, it kind of makes sense. But as we think about the emergence of robots kind of working in and around us, any thoughts on the economy or jobs and how things can be affected?

Mike: No, I don’t really see any applications where it’s a true one-to-one replacement of people. I mean, in the case of the machine tending application, most companies have actually increased their headcount even though they’ve adopted several cobots. That’s a very big space for universal robots and other industrial, what I would call industrial cobot companies. That’s huge in that you’re able to allow more skilled people to leverage their skill across the company better than they would have if they came in as a machine or a machine tender or somebody who’s handling the machine tool. Now they can work with multiple systems all at the same time, thanks to the cobot. That’s not really replacement, that’s just growth through cobots. Same thing in the similar argument, but also different at the same time in the lab space where you’ve got processes that would be impossible, absolutely impossible to do with humans. So I was just looking at a system the other day, high throughput screening system, two robots fitting in the space of, let’s say, this room here, and they’re doing the work of 100 people that would take ten times as much time and have a whole bunch of errors and mistakes that the robots would never make to be able to do that’s. Life science. Life science. You can’t do this stuff with humans. As simple as that. The technology has enabled, thanks to the cobots.

Jeff: Interesting. So it’s possible that more and more cobots could also create jobs because you are more productive, you’re generating more output, so you can actually hire people to do more skilled work. Does that seem like a fair assumption?

Mike: Absolutely, yes. It’s a win-win, as far as I can tell.

Jeff: Nice. That’s fun to think about. As we think about this space growing.

***

John Houston: The less talked about side of cobotics is humans helping robots versus the other way around. If you have a solution to a robotics problem, but you’re stuck on the last few challenges of it, say for example, your robot navigates it carries out an inspection mission, but there’s one door along the path that it can’t open on its own, then have it ask for help. There are people out there focusing on full end to end autonomy and considered a failure whenever it doesn’t perform as such. They’re putting so much time in developing solutions that could have already been deployed in collecting data for future durations, stress testing their chassis, testing all the other functionality they have and improving it instead of grinding to a halt because they can’t open a door. Humans can help robots just as much as they help us.

***

Jeff: This has been a great conversation. I have a few more questions before we kind of wrap up. One is, what do you think would be one of the coolest experiences with cobots? What’s, like a really cool use case that you can think about?

Mike: I had the opportunity to use the world’s first pneumatic cobot, which is a machine. It’s built by a company in Essling in Germany. It’s called Festo. And it’s actually I always kind of wrote it off as pneumatics are so odd to think of a robot actually being powered by pneumatic energy and the ability to actually serve a pneumatically controlled or pneumatically powered joint. It’s got some work, they’re working on some things, but it’s incredibly smooth. And I thought this thing could be amazing for orthopedic recovery. Right. Anyone needs to go through a motion, because I’ve never felt a robotic arm so buttery smooth with all six degrees of freedom and this thing could replicate. Let’s say I’ve got a problem in my wrist. You could actually do a therapeutic movement where it’s guided through this very smooth range of motion, whereas with these harmonic driven robots would just be so clunky and jolty and you’d have variations, hysteresis, and the forces applied. So, to me, I thought that would be kind of to me, it would be a really next-level use case as well as giving a practical application for the pneumatics right out of the gate. If we could make that happen, I think that’d be cool.

Jeff: That’s cool. I think as we think about the problems in the future around aging and taking care of our elderly family or population in general, obviously, Japan is working on this a lot right now, but that seems to be a game changer type of technology that could play a big role in the future.

Mike: Yeah, it’s been interesting to watch what they’ve done in Japan with assistive robots. Toyota, Honda, they’ve all had projects around that, and some amazing R&D has come from that. I mean, what Boston Dynamics has done, the fact that the robot can be pushed over, get up on its own, and it doesn’t bother it at all. It just keeps on going. And you’ve got Spot walking around the streets navigating all by itself. That’s beyond novelty, I’ll have to say. Yes, the videos are fun to watch, but I mean, the R&D that’s coming from that is amazing. And I’d love to see more true human contact applications with robots like physiotherapy and things like that because that’s one that gets overlooked. Right. But the robot mechanisms, the software defined robot controller. You can drive a robot today through any arbitrary trajectory in space, passively or actively with this control technology. And it really kind of is underutilized at this point. It’s kind of like a solution without a problem, but I think the problem can be brought to the solution. In this case.

***

Avinash: We’ve mostly been focused on robots that look like robot arms and AMRs that go around places. I’d also like to bring up miniature scale robots. These are still in research phases, but there are a lot of research groups that have started using miniature scale robots as surgery aids that can be released in a human body and travel to a certain part and get imaging or attack a tumor and stuff like that. This would be an amazing area to see growth in and this would be an indispensable tool once it’s more usable to a surgeon. And the group of doctors that today have to perform an invasive surgery may not need to do that in the future.

***

Jeff: As someone who’s been in this space for almost 20 years and then you can kind of see the future with the trends as well, is there anything you haven’t told us that you think is important as we think about the future of cobotics?

Mike: I would just say as with any field that’s relatively young, watch out for the hype and kind of look for the practicality and things. I think a lot of things could get overhyped as far as their application. It tends to pull people out of the space where they can definitely as a business, make money, most importantly as an engineer, get the satisfaction of a successful application and not shoot too far because there’s a lot that gets missed when we shoot too far. And I’m not saying obviously shoot for the sky, right? The sky is the limit. I’m a fan of that. Go long. But also in a practicality sense, I think it’s also worth just stepping back and going, well, wait, there’s a whole business here we’re passing by because I’ve seen that myself with working with collaborative robots for most of my career where I’m just like there’s this whole area. I just haven’t been seen because I didn’t want to see it. I’ve been looking too far ahead.

Jeff: Yeah, that makes sense, I think. Any advice for those that want to get more involved in cobotics? This is an emerging space. It’s a fast growing space. What would you say maybe from a conference or books or podcasts? Like, any thoughts on how do we get more involved?

Mike: Yeah, there’s a few things I have in mind. I think one of the best is just going somewhere where you can kind of see how these things are used. Right. I mean, not even universities are getting better and better. In fact, my first job, we donated robots to universities around the area, which is cool because people got exposure right away. They remembered the robots that actually resulted in business later on. Right, so it’s good for business as well. But having a look at where these things are used, I would say one of my favorites is the laboratory automation conferences. And the smaller cobot focus conferences are great examples of seeing these things actually in use in a way that you never would have imagined, where you’re basically taking a robot and one robot is doing the work of 100 people simultaneously. Right. And you can get to know that. I think there’s a lot of ways to if you have a particular interest in a market. For me, the automotive industry drew me in. It turned out within a few years, I had nothing to do with the automotive industry as far as robots. So it just was a starting point and I realized there were all these other opportunities. But I cannot knock working for a robot company. You will see more things because the applications, especially the articulated robots, are nearly endless. Right. And that kind of exposure gives you, I think, the best perspective possible. And guess what? They’re hiring. That’s the space that’s growing. Electromechanical background is the ideal, but either one of those or even a programming background. Some of my most talented colleagues I’ve ever worked with actually came from pure software that work in the robot space. Right. So that speaks to that. Right?

Jeff: Yeah. I like this notion of picking something you’re passionate about. You said automotive and then thinking about the adjacent robotics field. If you’re interested in robotics because there is so many use case so much happening, there might just be the convergence of those two. Right?

Mike: For sure. That’s what happened for me. It happened not to be in the automotive space ultimately, but I would never would have discovered it had I not pursued that idea.

Jeff: Okay, nice. One last question, and that is as we think about designing the future with more intent, technology has a life of its own. It’s often outpacing humans and their ability to adapt. But as we think about this space, can you think of any principles that are important so that we design cobots with intent? If we’re going to be working a lot more with robots in the future and they’re designed to be worked with us, can you think of any principles that we want to get right so we don’t get it wrong? I think we constantly are trying to catch up and realize, whoa, like, hey, we’re actually checking these things 343 times a day on average. Is that a good thing? Stuff like that. How do we design the future of robotics with more intent?

Mike: Well, I think it requires the exact opposite of looking at something. So you’re talking about like a mobile phone, right? There’s so many different things your brain can be drawn to on that one screen. You’ve got the news, you’ve got the weather, you’ve got work, you’ve got emails, all those things, none of which all converge. I think thinking with convergence in mind is really the most powerful thing you can do in general, and it really applies to robotics. Having spent so much time in this industry, it’s so important to look at things from different angles, and that’s applying robots. It’s really a function of multiple disciplines, but also subjecting the concepts to the scrutiny from different perspectives. Right, so you’re looking at the pick and place application, even the simple pick and place application you’re looking at, okay, what am I handling? Does it have a sharp edge? Can I use a tapered lead in on my placement? Nest the gripper? We’re talking about slippery, will a vacuum work? Do we have a flat surface? That kind of thing where you’re just really fixated on the details of that and the success that comes from looking at the details in the robot space. I don’t care what dimension it is, whether it’s software controls, the mechanics, every single aspect of robotics and the application of robots benefits from this detailed thinking. And it sometimes gets lost when you’re kind of just, I’ve got that covered. Let’s move on to the next thing. Really pick it up, turn it around in your hand a bit before you say you’re settled on a particular approach.

Jeff: Thank you for that advice and all your experience and wisdom as we think about the future of cobots and cobotics. Again, thank you for being here as a specialist in the space, it’s been great to have you. Loved hearing your thoughts.

Mike: Absolute pleasure. Thank you for having me.

***

Jeff: The Future Of podcast is brought to you by Fresh Consulting. To find out more about how we pair design and technology together to shape the future, visit us at freshconsulting.com. Make sure to search for The Future Of on Apple Podcasts, Spotify, Google Podcasts, or anywhere else podcasts are found. Make sure to click subscribe so you don’t miss any of our future episodes. And on behalf of our team here at Fresh, thank you for listening.

As we continue to push the boundaries of technology and automation, it’s becoming increasingly important to ensure that robots can work seamlessly alongside humans. Just like cars have advanced with lane management and awareness technologies, cobotics is now taking a big step forward by making the setup process easier, more fluid, and more human-friendly. Let’s make the most of this opportunity to bring the power of automation to even more industries, helping to make tasks faster, safer, and more efficient for everyone. By working together, humans and robots can accomplish great things!