Making Industrial Work Safer With Ascend Robotics

[00:00:00] Your agents aren't producing accurate answers because they don't have a complete semantic understanding of your data. And Denodo is solving this and solving it through semantic consistency. Through semantic consistency, your agents can start making accurate predictions in real time. So see what else Denodo can do by visiting denodo.com to learn more. But now let me introduce you to today's guest.

[00:00:32] Welcome back. I'm recording this episode from the show floor at IFS Unleashed 2026 in Orlando, where I've been exploring how AI and robotics are moving beyond impressive demos and into the industrial environments where the real work happens. And today's conversation will offer some fascinating examples from robots painting military ships

[00:00:57] to machines working near 12,000 volt electrical cables in places where, if we're honest, sending people would be incredibly dangerous. And David will also share how his robots have helped complete ship painting projects three to four times faster, reduced material waste by 25%, and even allowed experienced workers to move from hazardous environments and into much safer operating roles.

[00:01:25] We're also going to be talking about the realities of deploying robotics, why machines must adapt to unpredictable environments, and how AI is making development faster. And perhaps most importantly, we'll explore why the future of robotics could be about keeping experienced people in the workforce rather than replacing them. But enough for me. Let me beam your ears to the show floor now, and you can join me and David in conversation.

[00:01:56] So thank you for joining me on the show floor here at IFS Unleashed. Can you tell everyone listening a little about who you are and what you do? Okay, so we're Ascend Robotics. We do construction and maintenance robots that reach up high and go in dangerous, dirty, hot spaces that people are reluctant to go. And we're recording this live from your booth. What kind of questions are people asking you? Is there a general curiosity there? It is general curiosity. They want to know how we fit into the AI infrastructure

[00:02:24] because we're pumping out a lot of data, and there are people at headquarters who would like to use that data for different companies. And we are increasingly seeing impressive robotics demonstrations, especially at conferences like this. But what is the biggest gap between a successful demo and a machine that can reliably every day, work reliably every day in a working factory or indeed industrial environment? Probably 10 or 12 years of hard experience in the field.

[00:02:51] You'll see what the customer's site looks like and what their problems are and then trying to get the robot to be functional and robust out there. And you mentioned visiting a client's site. You don't have to mention any names here, but are there any use cases that would really bring it to life, the kind of measurable difference that robotics is making? Well, we've got one that we've got on the screen now showing a robot painting of a ship. So this is the Navy's fastest cargo ship.

[00:03:21] This is the fastest cargo ship in the world and a very aggressive hull. It came in and needed to be turned back into service very quickly. So they brought on the robots, our robots. We stripped it, painted it, and found that we saved about 25% of the material because we put more material on the ship and less in the air. So there was some left over. Were there any other big differences there? You've mentioned materials. What about cost time to paint the ship?

[00:03:51] Cost was about three or four times faster than typical with human crews. So much faster that way. There's still humans involved, monitoring it. The coating thickness was much more consistent. So plus or minus 10% over that whole three-week period versus 27%, plus or minus for the human crews. So we have a lot of people from the human crews that we've hired to be robot operators. Brilliant. And that is just one example of one industry.

[00:04:19] Any other use cases you can share or what you're sharing here? Well, we've got another one. We were called in. There was a plant conversion to a data center. I know data centers are a bit of an issue now, but this one wanted to convert an old factory. It was a great space, but it's very dirty. They wanted to clean the ceiling and coat the ceiling, but they wanted for some reason to run 12,000 volts live, keep that live in the ceiling.

[00:04:44] So they needed something that would go up and target the ceiling, but miss every 12,000 volt line. People were hesitant to do that, to be up near the line spraying solution. So we called in our robots and they did it and did it faster than the human crews. And I've got to ask, what is harder today? Improving the robotics hardware itself or actually integrating the robot and AI into some of those traditional existing industrial systems, workflows? Oh, yeah.

[00:05:13] Now you're getting into a whole different area. So we can make the robots do a lot in the field, but it's ever easier to integrate into the systems now with AI. So we find that our software for developing the robots, for getting to do new tasks is about eight times faster than it was 18 months ago. Just the AI tools are great for that part. The other part and why we're at this show is even though we're doing a lot with the robots and we're gathering a bunch of data,

[00:05:42] the companies don't have a way to consume that data yet. So we're here with IFS because they're that bridge that provides the pipeline between what we're doing on the edge and what the headquarters needs. And what is the key message you want people to walk away here? So for people that can't attend, can't visit your booth, what is it that you want them to walk away with? Well, we find mostly that a lot of people still don't believe the robots can do the things that we have them out in the field doing.

[00:06:10] So I think letting them know that robots can be out there painting, coding, installing fasteners at height. They can go into dirty, dangerous, confined spaces that people don't want to go for explosion risk. So the robots are out there. They're not necessarily humanoid. They're not necessarily walking on four legs. And I think that the robots will be out there in many forms.

[00:06:38] So it may not look like a robot at first blush, some of them, but they're doing real work. And are there any other examples of an AI-powered robot that's also performing useful work at scale rather than just a pilot or proof of concept? You mentioned a few already, but are there any others? That's a good question. We know of, there are probably 100 that we know of that are at that prototype phase or at the pilot phase.

[00:07:03] And I think they will succeed, but it's going to take some years to really work out the minor kinks because you need it. You can't have it being, you can't have it failing in the field. It needs to be always operational. It needs to be doing the right task and not making mistakes. And when it comes to exploring the art of the possible there, what happens when an environment becomes unpredictable? How can these systems deal with exceptions, edge cases, physical viability? I'm sure you get a lot of these types of things. That's where the AI comes in. Yes, yes, yes.

[00:07:32] So our robots are designed to adapt to the environment they're in. So they go in, they learn the environment, and they adapt to it. In fact, we started out with a robot system that would take a BIM model. It's an architectural model of a building, would execute a paint plan and go into the building and do it. But we pretty quickly found out nothing was ever built as it was designed. So we had a customer who came in and he said, can you just make it, have it go in the room and just paint stuff? Don't worry about how it was built. It's just painting what's actually there.

[00:08:01] So we took about nine months to actually make the robot, a lot of the smarts to make it like a dumb painter. Painter's pretty smart. So we'd just go into our room. It would assess it. It would see the walls. It would just go wall to wall. And it would communicate with the operator. It would say, should I paint this? Should I avoid this? And then it would learn. And so after part of a room, then it's on its way. Yeah. It's really good.

[00:08:27] And you raised a great point there because we're not talking about replacing humans with robots. Far from it. And you've given a few examples already. But where do humans ultimately, where will they always need to step in? And what does that effective collaboration look like between human worker and that AI-powered robot? What's that look like? So I'll give you two cases. One is the ship painting that we did. So that one, the robot's doing the painting. Traditionally, a person is up there in a hazmat suit.

[00:08:56] They're covered in paint and blowback. And it's a nasty job. So with our robots, that same person is involved. But they're on the ground in an air conditioning command center. So they haven't had complaints so far for their union. So we engage with the unions as sort of our first, say, here we're coming. We want to bring to the market. And so they're all about people, right?

[00:09:23] So a lot of people doing the construction tasks, they age out around late 40s or so because their shoulders are blown out. They've got back injuries. They can still drive a robot just fine. And they've got the expertise they're bringing to it. So a lot of those are the unions like us because we're keeping a lot of those guys in the workforce. And then the new, the young folks, they're using it as a pitch. They can say, you know, look, you can drive a robot. So come work for our union.

[00:09:54] Brilliant. And it's so true what you're saying because I know, I'm very good friends with a professional painter. His shoulder went in his 40s because of the repetitive strain example. And when it comes to measuring improvement, what kind of metric should buyers use to distinguish that general, genuinely useful industrial robot from just another tech demo? Any metrics there that you would? I mean, at the beginning, you give some great examples of the ship. And there's a great metrics there. So all of our clients are in the manufacturing space or in the heavy industrial space.

[00:10:23] So they aren't going to adopt a new technology unless it pays off. Yeah, yeah. And our focus has been on the worker safety. But so the other things were a bit of a surprise, the material savings, the consistency that's just something that a human can't really do without assistance from the robot. I think those were really found out when we got out in the field and experimented and had the robots deployed. And it's an ever-improving cycle.

[00:10:51] So you have the expert who's driving the robot still giving advice of, you know, robot, do this thing. Do this better. You know, that works. And so much has changed since the last IFS Unleashed. And I suspect everything you were talking about then has advanced so much. If I could fast forward 12 months, two years, what's your grand vision for everything?

[00:11:10] Well, I think IFS really sees the value in connecting the hardware that's out in the field gathering real data to, you know, the management team that needs to see what's going on in the world, right, with their systems. So we're finding, I think this is the start of that connection of not just knowing that you have a machine out there, that you have an asset, and you know it's got some preventative maintenance coming due. That was last year or two years ago.

[00:11:37] So now it's tapping into the data that that machine is gathering all the time, you know, every day. Exciting times ahead. And for anyone listening that would like to find out more information about anything we talked about today, where should they go? Check us out at ascendrobotics.com. That's A-S-C-E-N-D robotics.com. Awesome. I'll include a link to that and indeed your personal LinkedIn and encourage people to carry this conversation on. But thanks for joining me today. Very good. Thank you.

[00:12:04] So a big thank you to David from Ascend Robotics there. What I particularly enjoyed about this conversation was hearing about robots that are already doing difficult and dangerous jobs. Rather than simply performing for an audience at a tech conference. And I've seen much of that this week. If you have a look on my LinkedIn, you'll see many examples of that.

[00:12:26] And the ship painting example really stood out, especially the improvements in speed, material efficiency and coating consistency. But at the heart of all this, yes, we're talking about robotics, but I think there's a very human story that is even more interesting. David explained how experienced workers who might otherwise have had physically demanding careers because of injuries and they're able to continue using their knowledge while operating robots from a safer environment.

[00:12:56] And I also loved his observation that buildings rarely match their original designs. So the robots have to learn to work with what is actually in front of them. Another reminder, I think, that the industrial innovation must account for the unpredictable nature of the real world. So I'll be sharing more conversations from IFS Unleash. Follow more on Tech Talks Daily wherever you listen. And as always, I'll be back again tomorrow with another guest. But thank you so much for taking the time to listen.

[00:13:26] Bye for now.