Systems integrators can become victims of their own success. In these cases, remote support becomes a much more viable and cost-efficient option.

Some systems integrators can become victims of their own success. When integrators deploy many systems and the crew that built the cells don’t have the bandwidth to travel and see all these customers and can’t staff up massive service departments – it can become an issue. Companies looking for a more cost-effective means of keeping customers happy while also building new cells now have options, according to Fredrik Rydén, CEO of Olis Robotics. In this episode of Manufacturing Matters, Rydén joined TECH B2B Marketing’s Jimmy Carroll to discuss the company’s unique service within the industrial automation space, which is a remote monitoring system that runs next to a robot system, not only monitoring it 24/7 and catching, recording, and logging faults, but also enabling remote control and teleoperation of the system.

Jimmy Carroll: [00:00:06] Hi everybody. My name is Jimmy Carroll. I’m the vice president of operations at Tech B2B Marketing, and I’m here at the A3 Business Forum 2025 in Orlando. I have the pleasure of being joined by Fredrik Rydén of Olis Robotics. Fredrik, thank you so much for taking the time. I really appreciate it.

 

Fredrik Rydén: [00:00:19] Thank you.

 

Jimmy Carroll: [00:00:21] Of course. So I like to start off by asking everybody this. For those who don’t know, please tell us about Olis Robotics and what you guys do there.

 

Fredrik Rydén: [00:00:28] So let me tell you about myself first. So I’ve been working on remote controlling robots for over 15 years now. I started out in the offshore oil and gas sector, ended up working, doing big R&D projects with NASA and a bunch of the defense primes. And about three or four years ago, we started doing software for the industrial automation market we have here. And that’s basically software that you install and you run it on a device next to the robot, and that basically watches it 24/7 and catches any faults and records them and enables you to share those diagnostics very easily and also enables remote control and teleoperation and all those things.

 

Jimmy Carroll: [00:01:05] So that’s interesting. Part of my background is in machine vision — not as an integrator or anything, not an engineer, but covering it. So when I think of cameras in a robot space, I’m thinking vision guidance or inspection, but that’s obviously not what you’re doing here. So it’s remote monitoring. Do you also do the installation or do you do the systems integration or are you primarily software?

 

Fredrik Rydén: [00:01:27] So we partner with the people that are expert at installing robots. I think in theory if there was a simple install, Olis could do it. But we choose not to because the reality is to be good at installing a robot, you actually need to have a tremendous amount of domain expertise or application expertise, right? And that’s what we have in our system integrator networks that deploy our solutions.

 

Jimmy Carroll: [00:01:47] And I guess being here at the A3 Business Forum is a perfect example of the importance of partnerships. For you, talk a little bit about the importance of partnerships and what that means for you to be able to serve the global robotics community.

 

Fredrik Rydén: [00:02:00] Well, I think that’s a good point. If I didn’t have the system integrator network, I would have to staff up tremendously to be able to serve North America and beyond, right? It just wouldn’t be sustainable. And I wouldn’t be able to reach every little town, have manufacturing operations all over North America. So those partnerships are absolutely essential for us to be able to do what we do. And that lets us truly be the experts on the software, making it plug-and-play and something that the integrators can just use without really having to worry about how does it work or is it going to work? They just kind of plug it in and use it just like you do with your phone. You don’t think too much about it.

 

Jimmy Carroll: [00:02:43] So from that perspective, if somebody was a larger company or maybe they had an internal systems integration person, is it hardware agnostic or do you have a certain setup where you have to have X amount of cameras? A certain process or things like that?

 

Fredrik Rydén: [00:02:59] So we support the major robot brands and controller brands that people use primarily here in North America. So it’s agnostic in that sense. Then you can have surveillance cameras or even webcams would technically work. But most of the time people have these PoE, power over Ethernet, IP cameras.

 

Jimmy Carroll: [00:03:19] Machine vision cameras. So it seems like such a natural, obvious application, but I don’t know a ton about it I guess. So what are some of the downfalls of not having remote monitoring, and what are some of the benefits of it? And again, I know some of those are obvious, but maybe you have some specific examples or anecdotes of why this is important.

 

Fredrik Rydén: [00:03:41] Well, I think there are a couple of reasons. So we know that people that have their first automation cell installed can sometimes struggle. They need that handholding, but they can’t necessarily — they’re too small to justify hiring a dedicated robotics expert to sit around all day and just wait for problems to occur, right? So remote support becomes a much more viable and cost-efficient option in that case. We’re also seeing integrators that frankly have become victims of their own success. They’ve frankly deployed too many systems, and there are sometimes thousands of them out there. And the crew that built the cells, they don’t necessarily have the bandwidth to travel and see all these customers, and they don’t want to staff up massive service departments to serve that. So they’re looking for a more cost-efficient way to keep their existing customers happy while they build those new cells.

 

Jimmy Carroll: [00:04:37] Well, that just makes a lot of sense when you frame it that way. One thing that I like about going to these shows in general, A3 events, you get to talk to a lot of different folks and see different talks. And in the case of Automate, see these demonstrations in person. How, if at all, have advancements in things like different cameras or AI or end-of-arm tooling, how have those changed the requirements of the needs for remote monitoring, and has it increased the need? And have you had to do anything on the software side to be able to accommodate these new advancements?

 

Fredrik Rydén: [00:05:14] Yeah, I think as the complexity of the automation goes up, so does the need for remote monitoring and diagnostics. What I find so fascinating about this industry is that AI or machine learning or computer vision, whatever you want to call it, has been quite widely deployed since the ’80s in some cases. So there’s actually a fair amount of experience with what we call AI today in this field. And sure, it’s gone 10 or 100 times more robust, easier to use, than all those things. But it’s actually not a new technology in the sector, which many sectors can’t really say.

 

Jimmy Carroll: [00:05:54] Yeah, that’s interesting. And it’s almost like every year this show grows, and every other event tangentially related to A3 or industrial manufacturing grows and it evolves. But something that caught me off-guard, and I don’t know if you heard it earlier, but Rodney Brooks said that only 20% of warehouses are using automation. And that just blows my mind because I think of warehouses and I think of warehouses being a really good example of different technologies that are advancing and converging to come together. So you’ve got AMRs, you’ve got cobots, you’ve got traditional robots, you have AI, like you’re saying, high-speed barcode reading, general quality inspection. And just to think that there’s 80% of these that are unautomated — I don’t know if that’s a word or not — but either way, that just goes to show the opportunity that folks like you and everyone here has. I think that’s very interesting.

 

Fredrik Rydén: [00:06:49] And I think a lot of that comes from the fact that the automation still isn’t easy enough to use. Success is not guaranteed when you’re buying automation. And until we can guarantee success, until it’s just as simple as, like, I buy a car and I know that I’m going to be able to commute to work every day. Like, that’s sort of guaranteed. Unless you buy a fancy sports car and it breaks down. But you get my point, right? If you buy a Honda, you know it’s going to take you to work and back. And I don’t think we’re quite there yet with automation, I think some of the applications are becoming so commonplace that we understand them very well and they work really well. But I think there’s still concerns with, like, okay, we can integrate that into a new facility, but how do we retrofit an old facility? A lot of the complexity comes into play. I hope the software that we can bring, the infrastructure that we can bring, can provide the level of support to those end users to ensure that they’re successful the first time and every time they deploy automation. And ultimately I think taking the risk factor off the table for the end users or the buyers of the automation, that’s ultimately what’s going to drive adoption more than AI ever will.

 

Jimmy Carroll: [00:08:05] Okay. Interesting. And you touched on it in the beginning of your response. And I like that because I wanted to ask you about it: ease of use, usability, things like that get put out there a lot as a term. And maybe it’s an obvious term because people want things to be easier to use. But what are some ways both within what you guys do and in the general industrial automation space that folks are helping to lower the barrier of adoption to robotics and automation systems.

 

Fredrik Rydén: [00:08:29] I think people are transitioning from selling technology to selling products. And the way I think about that is: a product is something that I can buy and use without knowing how it works. It just works. You know, like my cell phone. I actually don’t know all the details of how everything is connected and how it all works or Wi-Fi or cellular connectivity. I have a vague idea how it works, but I’m still an expert user of it. And I think that’s the place we need to get for industrial automation, where people have a rough idea. Yeah, the robot picks here and there’s a camera, but they don’t actually need to know: Oh, there’s an I/O signal coming on this I/O going out to this PLC making the decision for this.

 

Jimmy Carroll: [00:09:12] What’s an encoder?

 

Fredrik Rydén: [00:09:13] Yeah, exactly. When people have to be such deep experts, it’s just not going to scale.

 

Jimmy Carroll: [00:09:20] That’s interesting. And I love that because I was talking about it. I did a podcast a couple of days ago, and one of the things we talked a lot about was the rise in availability of these application-specific purpose-built systems that just do sanding or grinding or bin picking or depalletizing. And that’s exactly what you’re saying here, I think. A product that’s sort of ready to go. You may need an engineer to do the initial setup, but from there on out, the person that’s been working on that hardware or the factory floor can turn it on and operate it and ideally troubleshoot it. And then with technologies like yours, be able to remotely monitor and troubleshoot. So let’s see what else? I did want to ask you a little bit more about AI, and we touched on it. In your experience, and I know that it’s not specifically what you guys do, but I’m very curious of your opinion. What are the short-term and long-term benefits of AI? It’s been subject to overhype, as we saw earlier, and everybody knows that. But it’s adding real value in a number of different ways. What are some key or maybe some notable ways that you’ve seen AI adding value recently?

 

Fredrik Rydén: [00:10:30] Well, I think we’ve seen the explosion in the picking space, of actually getting dynamic, unstructured environment, picking of items that maybe the robot has never even seen before. And we’re starting to see more widespread adoption of those. And it works really great. But it works because you have very experienced robotics technicians carefully setting it up and calibrating everything and putting everything into place. And I think what I’m most excited about for AI is getting it to a place where it’s not just for those very special, purpose-built installations, but it’s more of a general thing, like it’s as easy to use as starting an app on your phone, and you kind of get AI on your robot somehow. Like, that’s kind of the point we need to get to, because the AI is out there. It’s available in the cloud, the cloud services. You can send it pictures, you can send it log files, and it can send you analysis back. I think what is missing is just that last mile of how do we give the AI the data it needs to analyze? And how do we get the analysis back to the people that need to read it, which is probably robot technicians, operators, plant managers, so forth. I actually don’t think this industry needs to focus on building our own AI. It’s more about making the tools available to the people that need them in a way that’s sort of digestible.

 

Jimmy Carroll: [00:11:57] Yeah.

 

Fredrik Rydén: [00:11:58] I used to really drive home the point: we cannot have controls engineers having to become fully fledged programmers in order to leverage AI. It needs to be as simple as modifying some ladder code or something like that.

 

Jimmy Carroll: [00:12:16] And I have a thought. It’s kind of a simple thought, but I like to run it by people that are smarter than me and see what you think. Today, I guess they still exist, robot systems that are blind. They don’t have vision guidance. There’s still tons of applications today where that’s fully acceptable. Parts or products or whatever are presented in a structured manner, and those things don’t change. They remain static. When you add vision to those things, that helps add new layers of flexibility to help guide the robot, to compensate for parts moving, and things like that. When you add AI to the robot and the vision, it adds another layer of flexibility for things like what you’re saying, like bin picking or picking from clutter, from random bins or cartons, or picking organic fruit or organic matter, a chicken breast or whatever it might be. And it gives you a new layer of flexibility. Well, let me just stop there. Am I on the right track? Do you agree with that? And anything you’d add to it?

 

Fredrik Rydén: [00:13:28] I think it’s important to keep in mind that introducing AI in a robot application shouldn’t change the whole architecture of everything. You’re still going to have a PLC, probably in charge of the overarching structure and safety of the cell. You’re still going to have robots probably executing traditional trajectories, but a module of those trajectories are probably going to be generated by AI. And I think it was John Battelle that said it on the panel too. Actually, it was probably Rodney Brooks who said: It works well, but you have to sandbox it. AI doesn’t really work that well when you let it take over everything.  And so when you integrate AI, you need to have a pretty good understanding of industrial automation and software architecture, where it fits in, where it doesn’t belong. Hint: the safety systems. That’s where it doesn’t belong. But it can add tremendous value in: What’s in front of me and how do I pick it? So I forgot what your exact question was. But I think it’s really important to understand where to put AI and where to not put it.

 

Jimmy Carroll: [00:14:41] Yeah, well, I ask long questions, so that’s totally on me. But you did answer it. And that’s a really good point about understanding where to put AI, because like Rodney was saying, it’s not for everything. You know, some people just say, I think we need AI. Well, you have to do thorough application analysis to determine like, this could be a good fit. And anyway, enough about AI. I wanted to go back to a question about usability and ease of use and the idea of some of these more application-specific or even the idea that you’re driving towards, what are some ways beyond cost coming down that we might be able to see increased adoption in small to medium-sized enterprises?

 

Fredrik Rydén: [00:15:21] I think predictability. It’s kind of back to my point on risk. The more we can make automation a predictable investment with a very predictable return, I think that’s really what’s going to scale. I think if you drive the price down tremendously, maybe you’re sacrificing quality of the robot arm or something like that. Yeah, price is going down, but you’re also increasing the amount of risk you’re taking on by buying it. So I personally don’t think it’s the answer. Yeah, it’s always good if we can streamline our supply chains and get prices to come down and economies of scale and all that stuff. But ultimately I think what’s going to bring that scale is, again, that risk going down, predictability up.

 

Jimmy Carroll: [00:16:08] Well, that makes plenty of sense. What, in general, based on what you’ve seen on stage or in conversations you’ve had or whatever, what are you most excited about this week? What’s most interesting to you?

 

Fredrik Rydén: [00:16:22] Well, I’m excited to hear that a lot of our partners are starting to see orders come in, and they’re getting busier. They’re deploying more automation than they did last year at the same time. So I think that is extremely encouraging news for the whole industry. And I think it’s important to keep the focus on that, like this is not about building new technology, building cool robot applications for the sake of building them. But we need to be able to get them out into the world. And the way that happens is those orders and that money coming in. That’s when we know that it’s working.

 

Jimmy Carroll: [00:16:55] That’s great to hear. I’m very glad to hear that obviously. We’ve talked about a lot, but if you have any predictions for maybe the short term and the long term within industrial automation, I’d love to hear them. And if not, I totally understand. But it’s always something that I do like to ask.

 

Fredrik Rydén: [00:17:12] It’s hard to say. It’s sort of related to what Rod Brooks said about humanoids. I think he’s right in a lot of those points. I don’t think we’re going to see widespread adoption of humanoid robots bought for ROI. They might be bought for entertainment and other purposes. But I don’t think we’re anywhere close to seeing general-purpose robot manipulation in factories or in homes for that matter.

 

Jimmy Carroll: [00:17:47] Yeah, I now would agree. My opinion doesn’t go as far as yours in this, but I do agree. These things aren’t going to be affordable for quite some time. And I don’t know if they’ll be safe, at least in the short term. And the other thing is: Will they be able to do anything that a cobot arm on an AMR can’t do now? I’m not sure. But to your point about entertainment, they are very cool. They’re fun to look at. They’re interesting. What else that we haven’t covered that you might want to make mention of or you’re excited about?

 

Fredrik Rydén: [00:18:21] I think we’re good. I think it’s been a great two days here so far. Like I said, there’s a lot of energy. There’s a lot of people thinking about making their automation offerings simpler, more streamlined, easier for the end user, and all that stuff. And so I’m excited to see that we’re aligned with the rest of the community here. I feel like most of us are actually swimming in the same direction, which is incredibly encouraging.

 

Jimmy Carroll: [00:18:47] Yeah, absolutely. So Fredrik, thank you so much for taking the time. If folks want to learn more about Olis, is it olisrobotics.com?

 

Fredrik Rydén: [00:18:56] You can go to Fredrik Rydén on LinkedIn. That’s a good start. I post a lot of content about remote control, remote monitoring, videos. And sometimes it’s about our products, sometimes it’s about other stuff. But then from there you can often find Olis Robotics LinkedIn page. That’s a very good resource as well.

 

Jimmy Carroll: [00:19:13] Okay. Well thanks very much. If anybody has any questions or comments for Fredrik, I’d be happy to pass those along. Just check us out at Manufacturing-Matters.com. And thanks for tuning in.

 

Fredrik Rydén: [00:19:22] Thank you.