NDE 4.0 Podcast Transcript
Episode 24 — AI & Autonomous Robots: The Future of NDT
Our Guest: Dustin Whitehead
Nasrin Azari: [00:00:00] Hello and welcome to floodlight Software’s NDE 4.0 podcast. In this series, we interview various experts in Industry 4.0 concepts, ideas, issues, and technologies as they relate to non-destructive testing and inspections. This show is designed to explore the biggest challenges and opportunities for the future of NDT guided by some of the smartest people in the industry.
Nasrin Azari: So be prepared for a thought provoking discussion and to learn something new in the next 30 minutes. Hope you enjoy the episode.
Nasrin Azari: Today I am honored to welcome Dustin Whitehead to the podcast. Dustin has been a leader in the [00:01:00] non-destructive testing and inspection industry for well over two decades. He has held executive positions across multiple verticals. Overseeing revenue growth, commercial strategy, and international expansion.
Nasrin Azari: Currently, he serves as president of advanced inspection and the C-E-O-O-C-M-O of Sub CNDT and engineering, while also spearheading the development of NDT Nexus, a next generation digital ecosystem for the testing, inspection, certification, and compliance sector. Dust Dustin’s career has focused on advancing inspection technologies.
Nasrin Azari: Particularly robotics ultrasonic testing, phased array, and automated digital radiography while bridging the gap between field application and commercial strategy. His leadership encompasses scaling companies, mentoring the next generation of technicians, and driving the adoption of innovative solutions to enhance operational safety and efficiency across the energy infrastructure and industrial sectors.
Nasrin Azari: In recognition of his contributions, Dustin was [00:02:00] elected to the Board of Directors of the American Society for Non-Destructive Testing in 2025. In this role, he continues to shape industry standards, influence workforce development, and integrate advanced technologies into inspection practices, thereby instilling confidence in his leadership and the industry’s future.
Nasrin Azari: A couple of weeks ago, Dustin posted a commentary about how AI is already being written into regulatory standards and that at some point in the maybe not so distance, future AI assisted inspections might become part of required procedures. He and I followed up afterwards with a great discussion about his recent visit to Carnegie Mellon’s Robotic Institute and the various ways that AI is and will be transforming the NDT industry.
Nasrin Azari: And I immediately knew he would be a fantastic guest on our NDE 4.0 podcast. So here we are. Welcome to the podcast, Dustin. I am so glad to have you.
Dustin Whitehead: Yes, no, thank you. I’m glad to be here.
Nasrin Azari: Awesome. So in this [00:03:00] format for this podcast, I’ll pose five questions to you, specifically around AI and hopefully generate some really interesting conversation and I’m looking forward to hearing.
Nasrin Azari: Some of these topics that we’ve chosen. So let’s jump right in. The first topic is how is AI transforming the analysis of terabytes of ultrasonic inspection data?
Dustin Whitehead: That is actually a wonderful question and so glad you asked. Um, if you look at the evolution specifically with ultrasound testing and most of the advanced techniques that are used throughout the industry.
Dustin Whitehead: The biggest problem encountered is data collection size.
Speaker 3: Mm-hmm.
Dustin Whitehead: Right? So it’s always been a big trade off between how much data you need and how much data is actually useful based on the technology available. So I mean, even if you go back and read some of the early reports in the eighties of some of [00:04:00] the technology development, they’re like, this is going to be the future of inspection.
Dustin Whitehead: We just don’t have the ability to process that data. We don’t have the transfer speeds that actually make the data collection a viable time efficient way to go about the inspection. Because of course, with any part under consideration, we don’t get called out just to go make sure pipe might be okay or you know, make sure our vessel is built to the specifications.
Dustin Whitehead: No, they want us to go out and actually cement, like, Hey, this is the code that it was built to per this code. Here’s the information, the data that we have available that then led to this interpretation of the status of the vessel. Be it needing repair, good to go, or I wouldn’t even buy it if I were you.
Dustin Whitehead: Um, so based on that, you know, anytime you have a transition. Throughout the realm of, you know, we went from analog to digital and everybody’s like, oh, I hate digital. Digital will never [00:05:00] be the future. My analog scope is the best thing I’ve ever seen, which I literally just had a conversation with somebody yesterday about.
Dustin Whitehead: Um, but that trans, that transformation has led to new faster inspection times, more reliable inspection data. Um. A huge transition into, you know, away from the paper and more into digital reporting.
Speaker 3: Yeah.
Dustin Whitehead: So when you look at the history of all of these processes coming together. Inspection companies are like, we have all this data, but we don’t have anything to do with it.
Dustin Whitehead: Like, what good is it to
Nasrin Azari: us? Right. Right. I’ve heard that from a lot of companies that, yes, and it’s fascinating. I mean, I think you’re right with the ability to process, you know, using AI technologies to be able to process that data well now that data becomes useful. And so, um, having the ability to store it somewhere where it makes sense and it’s easily accessible is gonna be important.
Nasrin Azari: But then, you know, being able to, or. [00:06:00] Kind of, you know, knowingly capturing all of that beta data, saving it so that you can actually perform more, um, extensive analysis of the data, I’m sure is gonna result in, I’m assuming, more accurate results, maybe more informed results, maybe just better, you know, resulting recommendations.
Nasrin Azari: Oh,
Dustin Whitehead: 100%. And if you look at, take for example, I would say just over the course of a. Five year span. I have countless hard drives just full of data, be from, you know, welds, vessels, um, corrosion mapping, data tank inspection data, and it just sits there not doing anything for anybody. Well, one of the things that the next generation of software that should be coming out very soon, um, will actually allow you to not only take that data.
Dustin Whitehead: To formulate formula, a better decision [00:07:00] matrix based off of what the data shows, right? Because everything right now is oftentimes left up to operator interpretation.
Speaker 3: Mm-hmm.
Dustin Whitehead: Well, as you see a lot of the. I don’t know. I call ’em the rock stars of NDT. Mm-hmm. Um, we’re not getting younger and a lot of people are starting to transition out, take on more management, uh, positions within companies or, you know, retiring and just enjoying Sure.
Dustin Whitehead: You know, the time that they have. Um, a lot of that knowledge is just lost. Mm-hmm. So being able to take historical data and throw it into a program and say, okay. And not even tracking it by. I mean there, there’s a lot of application there as to condition monitoring of vessels and you know, really able being able to elevate the amount of data collected, processed, and was the inspection technique right in the first place.
Dustin Whitehead: How reliable are the fitness for service calculations? You know, risk [00:08:00] based inspection, like, are we inspecting reasonable interview intervals? There’s just a whole lot that goes to it. But you know, when we look at upskilling the future generations, taking that historical data and processing it and being able to relay that information without having a person to physically do it, is gonna be a game changer.
Dustin Whitehead: So if you look at someone like the training techniques, a lot of these are going to be, I don’t wanna say automated, but we’ll call it subsidized by machine learning, having the ability to go into a machine and you know. Actually walk through a realistic type scenario using real data from previous inspections and being able to do it without actually having to put yourself in the environment when the data was collected or having to go through, you know, any of the processes do that.
Dustin Whitehead: So it really makes training a lot easier. The data’s process faster and you’ll start to see a lot of technicians will literally level up. Yeah, just based off the data that all these companies already have.
Nasrin Azari: I think [00:09:00] that that, um, that’s really fascinating. And, you know, one of our other questions that I wanted to get into today, I think is very, uh, aligned with what you just started talking about.
Nasrin Azari: That that question is how does, so I, and I know that there’s a lot of concern across the industry about. People’s jobs being replaced by these AI systems and you, along with some of the other sort of more forward. Thinkers around ai, I think are of, kind of, of that, of that belief that AI can be a, an assistant or an augment to the workforce versus a replacement.
Nasrin Azari: So, uh, one of the questions that we have here is how does AI enhance inspector capabilities while still preserving the workforce?
Dustin Whitehead: So that actually ties right into kind of what we were talking about with phase array. So when Phase array came out in the early two thousands, [00:10:00] mid two thousands and really started gaining a lot of traction, everybody’s biggest push was, oh, phase array is gonna res replace radiography.
Dustin Whitehead: It’s the next big thing. Radiography. ISS gone. All the radiographic inspectors got, you know, upset. They were like, am I still gonna have a job in 20 years? Well, here we are 20 years later and there’s still just as much radiography going on.
Speaker 3: Yeah. As
Dustin Whitehead: it was before. Um, so it’s not a replacement, it is more of a, a system, just as you said.
Dustin Whitehead: Um, and not only in, you know. Kinda helping the inspectors with their level of knowledge, be it, you know, through, you know, agent queries or whatever they’re looking to do to learn more about the inspections they’re doing, but being able to really expand on the realm of knowledge that they’re able to present
Speaker 3: mm-hmm.
Dustin Whitehead: At one time. Um, it really helps cross [00:11:00] over some of the different technique variables. So. If you look at a radiographic inspector going out and doing a 6 53 inspection with a 6 53 inspector, like there’s a huge language gap in there mm-hmm. That may not be able to convey information in an effective manner to where the API can make a valid decision based off the radiographic report.
Dustin Whitehead: And the radiographic report might not be written in a useful way for the 6 53, but you can run it through an agent and query. Hey, this is what that report says. This is what his report says. How do we culminate this and provide the client with a deliverable that’s actually insightful,
Nasrin Azari: right?
Dustin Whitehead: Without them having to go and look at each report.
Dustin Whitehead: So it really helps crossover, do a lot of crossover in between. That makes the technician a much more valuable asset, not based off of just what he knows, but the ability that he has to kind of cross outside the traditional knowledge gaps that most encounter.
Nasrin Azari: Yeah, that makes a lot of sense and that’s a [00:12:00] really interesting perspective.
Nasrin Azari: Um, you know, I’ve, we. Quite a few conversations I’ve had in this podcast have been around, you know, using AI as an assistant versus thinking that it’s gonna replace. And I, I truly believe that that’s the way the industry’s going and I actually think it’s gonna be a benefit given the shortage of sort of like you were talking about, where there’s a lot of knowledge that’s kind of phasing out of the industry as people retire, move on to their next thing.
Nasrin Azari: Um. And I think there’s a way to preserve that, uh, a little bit. And like you said, take advantage of, let, allow inspectors to focus on what they really are interested in and become very proficient at that thing. And then, and then be able to use assistance to sort of bridge that gap between the other, uh, methodologies or the other, uh, functions within the testing kind of project, if you will.[00:13:00]
Dustin Whitehead: Yeah, no, it’s just a huge overall boost. I mean, even if you look at the amount of books I have on different techniques and applications and evaluation methodologies, and I have read them at one point in time over another, but being able to recall that information from the top of my head, it’s quite daunting.
Dustin Whitehead: Yes. But having age that I can queer and be like, oh, I know I read this in this book. What did it say? Instead of me spending a day trying to go through a 300 page book and find exactly what I’m looking for, I can simply query the book Yes. And get that answer in seconds versus, you know, wasting a day’s worth of time, which of course, if I had to report to somebody or I’m on site looking for this answer, they want it now.
Nasrin Azari: Right.
Dustin Whitehead: So having that ability to provide somebody information in a very time efficient manner. It’s huge. Yes. And I’m sure that there are hundreds of technicians using that probably as we speak right now.
Nasrin Azari: Right, right. Yeah, absolutely. Let’s, [00:14:00] um, let’s change our focus and talk about your recent visit to Carnegie Mellon’s Robotics Institute and what cutting edge robotics research is making its way into industrial inspection.
Dustin Whitehead: That is actually it. It was, it was a very, very eye-opening. Uh, true to say the least. There was some of the coolest toys I’ve ever seen in my life.
Speaker 3: I bet.
Dustin Whitehead: Um, it was like the things that we tried to build outta Legos back in the day, um, that they’re actually putting in place. And it was really interesting.
Dustin Whitehead: There was one of the groups that was presenting, um, a status update on their thesis, and part of their thesis was creating a. Robotic vision system that could lift, uh, valves and flanges intended for subsea use and provide a as-built 3D model that would then transition into an as-built CAD [00:15:00] drawing for the part.
Dustin Whitehead: ’cause there’s a huge difference between what’s called out and manufacturing and what actually gets delivered. Okay.
Speaker 3: Um,
Dustin Whitehead: and so they had developed this arm. That basically act as a, a turnstile. So would take the part, pick the part up and move it around the camera lens. That would then go inside the part externally from a part and look for defects.
Dustin Whitehead: And one of the things that was brought up during that, um, thing was, well, have you thought about, you know, checking for cracks and welds or incorporating, you know, some radiography or ultrasonic of testing to this too, to give you an entire volumetric inspection in the park? And they’re like, no, we were just asked to build something like this for.
Dustin Whitehead: A customer, we don’t know exactly what he’s gonna use it for.
Speaker 3: Mm-hmm.
Dustin Whitehead: But having the ability to know what’s out there and bridging that gap between what, what’s available tomorrow and what we need today is gonna be huge for the future of the industry. ’cause there’s so many things, so many developments going on that may not [00:16:00] apply directly to what we do in NDT currently.
Dustin Whitehead: But if you don’t know what’s out there that could be adopted or brought into the space. There’s not gonna be a much of a evolution per se. Right.
Nasrin Azari: Right. Or it will just take longer. Right. I mean
Dustin Whitehead: mm-hmm.
Nasrin Azari: It’ll take longer if you have to, if you feel like you’ve gotta build, you know, come up with your own ideas versus sort of borrowing from other industries or from universities that are performing this amazing research and they, they, they don’t know how, how, you know, the best ways to use the.
Nasrin Azari: Technologies that they’re coming up with. I mean, you’re sort of bridging that gap between here’s some really awesome technology, here’s some, here’s a real industry problem, and here’s how we can actually take that technology and put it to good use.
Dustin Whitehead: Oh, 100%. And I mean, there was a, there was a group of students that was working on a un completely unrelated development, which will be really cool when it comes out.
Dustin Whitehead: Um, but one of the mainstays of the product [00:17:00] was collision avoidance. And it wasn’t collision avoidance in the sense of like trying to prevent a car crash. Okay. But just how can you augment, say, a blind person’s ability to avoid stepping over a curb or running into a light pole or hitting sign that’s too high.
Dustin Whitehead: And one of the things that, that really brought to my mind is how when we’re going out and looking at, you know, the next generation of products that are in the market, how do we take the techniques that we have today, integrate those next gen technologies, such as collision avoidance, so that when you’re working in a refinery, your scanner doesn’t run into a nozzle and fall off a tank 300 feet.
Dustin Whitehead: Mm-hmm. Like there, there’s a lot of, you know, kind of hazards that go in or associated with that. I mean, even then, the ability to detect a drop object right after it’s dropped and deploy a safety net. Like there’s so much utilization that you can take these outside technologies and apply ’em to our industry to make the industry safer, make the industry more reliable, [00:18:00] and make it a much more effective industry overall.
Nasrin Azari: Right?
Dustin Whitehead: Because of course, nobody likes when inspections miss something or run into a pipe rack.
Nasrin Azari: Right. Or break something.
Dustin Whitehead: Yes, yes. Or the robots
Nasrin Azari: break something. Yeah. Yeah, I can see that. I mean, it’s funny. It’s, it’s, it’s, this is one of those things where people talk about ai, like it’s this one thing and it’s really a compilation of so many different things, and they’re very specialized.
Nasrin Azari: Um, and so sometimes it’s, you know, we almost, we almost minimize the. Both the variability of the different types of AI that they, that there will be and the impacts that the, that those ais will have on, you know, not just NDT industry, but tons of industries, um, are go going to be affected. And it’s just affected and I think sometimes.
Nasrin Azari: It’s, I think it’s gonna blow us away, uh, as new technologies [00:19:00] are created and born and developed and proven out. I think, I think we’re gonna, we’re gonna be just blown away.
Dustin Whitehead: Oh, yeah. Well, and I think there, there’s two things to that. I think a lot of it has to do with. I don’t wanna say the movie industry over glamorizing what AI is.
Dustin Whitehead: Yeah. So you think iRobot, you think, you know, independence Day of, oh my God, these robots are gonna come down and take over the world and, you know, humans are gonna be caught in the matrix. Um, it’s not like that at all. I think, you know, when we look at upskilling and supplement. The thing that affects humans the most when making in, in being able to make a purely logic based decision.
Dustin Whitehead: It’s very hard to not be emotionally involved in the decision making process. It’s very hard not to let environmental factors, uh, factors affect your ability to make decisions. If you’re tired, you’re sleepy, you [00:20:00] stayed up too late, you know, there’s a lot of things that go into everyday decision making processes.
Dustin Whitehead: I think the big adder there is being able to bounce your decision off of somebody and saying, Hey, this is what I think in a purely based on this data set decision, what would the outcome be? Right?
Nasrin Azari: Yeah.
Dustin Whitehead: Humans have a very hard time making a very analytical decision, whereas a computer has no emotional attachment to the decision it’s making at all, um, which is a good and bad thing, but I think overall, just having another opinion.
Dustin Whitehead: A lot of the times, or having the ability to learn more information outside of your current set of knowledge will really benefit everybody. Oh, definitely. In definitely
Nasrin Azari: as a whole. And it’s, it’s funny, I, I think you’re absolutely right. You know, you’ve got your ai that, that has zero emotion and that makes it really good at certain things.
Nasrin Azari: But then you’ve got the hu let’s take advantage of the fact that, that as you, as you mentioned, humans. [00:21:00] There’s, it’s very difficult to take the emotions out of our decision making. Well, let’s use that as our own advantage, right? That there’s certain, there’s certain, you know, areas where that’s gonna benefit us and where the AI is gonna be at a disadvantage.
Nasrin Azari: And let’s sort of keep that in mind, I guess, as we figure out the roles that each pe, each person plays and each technology plays. Um, hundred
Dustin Whitehead: percent. And I mean, even if you look, there are humans that are very. I don’t want to, we’ll just call it TAUs of their decision making at time. Mm-hmm. But it’s also a tool for them of, am I being too stubborn in my decision making choices?
Dustin Whitehead: Is there a better alternative or a better way to relay the information? So it helps really both sides of the coin. So while it can’t have a genuine sentiment on the decision it makes, it can act if it
Nasrin Azari: Yeah, that’s true.
Dustin Whitehead: That’s true. So I mean, it’s kind of been a very interesting. It’s very fascinating.
Nasrin Azari: Yeah. Thinking about all these topics is so fascinating, [00:22:00] and I know we’ve gotten a little bit off track here. So let’s move to our next question, which is, how is AI revolutionizing inspection, reporting, and real time adaptation?
Dustin Whitehead: That is amazing. So if you look kind of going back into what we were just talking about, about the human’s emotions, the fatigue factors, uh, there’s so much that goes into writing a report.
Dustin Whitehead: And a lot of technicians may or may not realize that those reports can actually be used in court as legally bound documents in the event of a failure or catastrophe. And it happens all the time. And one of the things that, you know, without taking that into account when you write a report. Some people go in and just put the bare minimum.
Dustin Whitehead: Some people put too much. I’ve literally seen a report, this is no joke, it was an internal inspection report on a high pressure vessel that said, looks good, will [00:23:00] work for a long time. That was what the report said. Nothing else but taking, you know, the AI machine learning portion of that feeding reports through, because.
Dustin Whitehead: A quality director is inundated with things to review, to ensure reporting, compliance technique, compliance certification, like there’s a whole lot there. And so one of the things that we’re able to do with machine learning is take it through and run it through a very specific set of parameters of does this report meet this criteria?
Dustin Whitehead: Does this report, you know, it’s checking all the boxes without necessarily having to have a person go through and make sure all those boxes are checked. So you’re able to process a hundred reports in 30 minutes and get a report out that says, Hey, we need to review this report, this report, and this report.
Dustin Whitehead: So now you just get an entire day’s worth of work and distilled it down to 30 minutes. Yeah. So it gives you really actionable insights moving forward. [00:24:00] And as you get through the whole workflow process of that, that decreases billing times, it decreases chargeback for NDT companies, I mean. It just benefits everybody overall.
Dustin Whitehead: And you can give the technician real time feedback of, Hey, you just heard this report in it needs to be reviewed.
Nasrin Azari: Right?
Dustin Whitehead: So really bridging that gap between report submission, report review and acceptance, or rejectance into a almost real time scenario is gonna completely change the industries. ’cause now you can train your technicians faster, you can correct them faster, and you can prevent incidents from occurring.
Dustin Whitehead: I won’t say prevent them from occurring, but you can help to mitigate the, the chance for something to happen due to a reporting issue.
Nasrin Azari: Yeah. Yeah. I mean, I love the, I love the idea of, or just the kind of the thought process around all the efficiencies that we’re gonna be gaining in different areas and how that’s gonna [00:25:00] allow us to improve the processes and, you know.
Nasrin Azari: I, I think that there’s a lot of interest in, you know, making sure that NDT is seen as a value, a valued part of the construction or manufacturing or maintenance process for assets instead of this sort of pain in the neck thing that has to be done in order to get the seal of approval. Right. And I think that’s kind of part of what we’re seeing here is, um.
Nasrin Azari: Just being able to provide a lot more value quicker and at less cost to the organization that’s providing the inspection work at the end of the day.
Dustin Whitehead: Oh, 100%. And I think it’s going to be, this might be a little far out there. I think it’s gonna help to almost level set the industry.
Speaker 3: Mm-hmm.
Dustin Whitehead: Because a lot of the top tier companies are top tier because they gate keep knowledge.
Dustin Whitehead: They [00:26:00] gatekeep personnel. They gatekeep ideas. Well, when you have the ability to query large data sets in a very time effective manner, you’re not relying on those people that are tied up in an ivory tower to pass a decision along. And you’ve got the small companies that maybe want to improve but don’t know how.
Nasrin Azari: Yeah,
Dustin Whitehead: so it really gives, it, it kinda levels the field there. And I mean, even helping to standardize technician abilities across the field. I mean, there’s a, there’s a wide breadth between certified inspectors that I think this will really help kind of bridge as we move forward. Uh,
Nasrin Azari: yeah, that’s, that’s awesome.
Nasrin Azari: Um. So as we’re talking about the workforce and as we’re talking about, um, the roles of, I inspect the inspector changing and kinda the ways that AI is affecting that, our last question is around the autonomous future. What [00:27:00] breakthroughs are bringing us closer to fully autonomous inspection robots?
Dustin Whitehead: Ooh, that’s a great question and it, it kind of ties back into the whole.
Dustin Whitehead: The conversation as a whole. Um, yeah, so if you look at like the C collision avoidance, the reporting time for inspection, setup, deployment, you know, really making an efficient use of NDE, um, is gonna be huge. ’cause I mean, we are the pain in the butt. Like we’re more the afterthought when it comes to most projects.
Dustin Whitehead: So like, oh yeah, and now we have to hire MDE to come in and make sure our pipeline’s compliant or make sure we can start our unit up. We don’t want that. A lot of it has to do with the recording lag, the, you know, gaps in training, the ability to be efficient in our, our, our application. Um, and especially as we move forward with the kind of workforce, I mean, that’s gonna be a big thing that a lot of the majors are gonna start bringing up.
Dustin Whitehead: So, with that in mind, looking forward to, you know, what we’re trying to do specifically is [00:28:00] build robots that really help people do the job job more efficiently. You know, minimize setup time, minimize reporting time, minimize accident occurrences. Really getting to a point where you can go out and put a, we’ll call it a tool on a pipe or a vessel, and it goes and plans out the inspection.
Dustin Whitehead: It really goes in and it’ll be able to provide a CAD drawing as build drawing. It’ll actually pinpoint. Corrosion is instead of having to have, you know, do a general screening and then follow up with a in-depth corrosion map and then going back and manually verifying it, you can get the robot to automate those tasks and actually make it get smarter the more that it’s used.
Dustin Whitehead: And you know, that’s one of the things that we’re currently working on is, you know, making a tool that makes the inspector’s life easier. Right, because the worst thing you can do is, you know, you go into the office Monday morning, they’re like, oh, hey, you got 18 tanks to inspect this week, which everybody knows, happens all the time, and it isn’t possible.
Dustin Whitehead: But getting to a [00:29:00] point where you can deploy a tool that makes your life easier, gets the job done, and then you just simply go back, review the reports, review the data, look at anything that’s flagged. I mean, it’s literally going to make. Inspection. A much more attractive, not only career, but a much more beneficial tool to the end users.
Nasrin Azari: Yeah, and the other thing that, that I’ve heard other people say about using robots, particularly smart robots, is the ability to make it safer for the human inspectors as well. Obviously if you, you spend a lot of money on a, on a, on a. On an inspection robot, you obviously want it to continue operating for quite some time, but if an accident were to occur, if there’s a dangerous situation or like bridge inspections, I think about all the harnessing that has to happen.
Nasrin Azari: If you have the ability to send up a robot, um, [00:30:00] that might actually be faster because you don’t have to put the sort of safety infrastructure in place or the same amount of safety infrastructure in place and it. And, and it might involve less people, which would also be cost effective. What do you think about that aspect of robotics?
Dustin Whitehead: Oh, it’s gonna be huge. So that, that’s actually one of, one of my favorite topics is, you know, I have the ability throughout my inspection, through career to go to a lot of places. It wouldn’t be the places that most people would want to go to. It wouldn’t be working in conditions that most people would wanna work.
Dustin Whitehead: Um, so at height, hanging from the bottom of an oil rig or working on top of a wind tower. Um. Those are traditionally things that people don’t like to do.
Speaker 3: Right?
Dustin Whitehead: And so finding people, especially to do those tasks or work inside of a Coke drum or, you know, even, even just refinery in general, I mean, there’s a lot of things that are not so good for humans to be in, right?
Dustin Whitehead: They don’t affect the, so. Being able to take those [00:31:00] robots and fill those, I don’t want to go in there type scenarios. Mm-hmm. Um, is gonna be a huge push forward. But then, you know, kind of in the next gen of stuff is the safety interlocks in the robotics. So whereas robotics were very robotic in their movement, you know, even 10 years ago, we’re now able to integrate safety algorithms to where if the robot senses itself in an unsafe situation, it stops.
Dustin Whitehead: Which wasn’t something, you know, that’s pretty cool. They would literally run until they broke themselves to death.
Nasrin Azari: Yeah.
Dustin Whitehead: Um, but, you know, to kind of facilitate that is a lot of the, well, I mean even the automated systems that we use for a lot of our inspections today, they’re a one-way communication operator tells us where, where to go.
Dustin Whitehead: And it goes, it does the inspection task and it stop.
Speaker 3: Mm-hmm.
Dustin Whitehead: Um. So kind of bridging that gap with the feedback loop of where you deploy the robot and then the robot robot starts to [00:32:00] sense, you know, there’s a change in voltage that shouldn’t be there. There’s system noise that shouldn’t be there. It can actually create a data flag that then feeds back to the operator and says, Hey, this motor’s acting in a abnormal status.
Dustin Whitehead: What do you want to do?
Nasrin Azari: Yeah.
Dustin Whitehead: You know, do you wanna stop the robot? Do you wanna continue the robot? But really getting that feedback. And of course, as you keep running that system. It gets better and better at detecting unsafe conditions, be it an operating state and environmental hazard. I mean, there’s all sorts of things that go into it, but the tools actually have the ability to start learning, and then of course can gain that autonomous state, which isn’t necessarily true a hundred percent, but it gets to a state where it can operate in a much more.
Dustin Whitehead: User friendly, safe, and mechanically inclined state.
Nasrin Azari: Right.
Dustin Whitehead: So
Nasrin Azari: that’s super interesting.
Dustin Whitehead: Well, and of course now you can [00:33:00] integrate, you know, sensors so you don’t have somebody on the side of the tower. That gets caught in a, you know, H two s mm-hmm. Because something went up there and you’re on ropes. And of course anybody that does rope work knows it’s not easy to climb and, and out of piping no matter how fast the safety guy tells you to get down.
Nasrin Azari: Right, right, right. But having
Dustin Whitehead: the robots with the ability to. You know, have a gas detector so you can send it into a place before you send a human. Right. If you have to send a human at all. And bridging that gap to where we don’t have to go into those environments anymore, right. And put our stuff at risk is gonna be a huge factor moving forward.
Dustin Whitehead: ’cause nobody wants to go out and get hurt at work.
Nasrin Azari: No, no. Nobody wants to send their people in places where they could get hurt and, and also the other thing about. You know, kind of just along those lines, robots can make very precise movements, whereas, as you were saying, people can’t, especially if they’re panicked.
Nasrin Azari: Mm-hmm. You know, if they’re panicked about something, then they may not make the best decisions. Whereas the [00:34:00] robots are more likely to make the better decision and get out of harm’s way easier. Um, you know, without kind of freezing, I suppose. So, mm-hmm. Yeah. Um, super interesting. I mean, I, my, my mind is kind of going in a million different directions and I expect that, that the minds of our audience are also going in a million different directions.
Nasrin Azari: Um, so thanks Dustin for a super informative, I learned so much in our conversation today, so this is really fun for me. Is there anything else that you’d like to add before we sign off for the day?
Dustin Whitehead: I think that’s pretty much it. Of course, I could sit here and probably talk for another two hours,
Nasrin Azari: I think.
Nasrin Azari: I think you and I both, I think you and I both together. Um, so thank you so much, Dustin, for joining. It was, like I said, really, um, very educational, informative, and a pleasure to have you on the podcast today.
Dustin Whitehead: Yes, thank you. It was my pleasure as well. I had a great conversation.
Nasrin Azari: Good, and, and thank you also to our audience for joining today.
Nasrin Azari: We will be [00:35:00] posting this podcast episode both on our website and in our new LinkedIn discussion group. And so this is a place where both Dustin and I will jump in and answer any questions you have or response any of your commentary, because I’m sure that there will be some based on the conversations that we had today.
Nasrin Azari: So you can off offer your own thoughts and commentary and then, and then start some. Some, you know, follow on conversations and I think that will be really fun. So I encourage you to participate in that discussion because sometimes those discussions often add a lot more value to the topic than we were able to cover today.
Nasrin Azari: I’d also like to remind our listeners that we welcome feedback as well as nominations for future guests. So to do that, you can send a message to us through the Contact us form on our website. www.floodlightsoft.com. Thanks again for joining us and we’ll see you next time.[00:36:00]
Nasrin Azari: And that’s a wrap for today’s discussion. Head on over to our NDE 4.0 podcast page for more interviews like this one, and reach out if you have any questions, feedback, or ideas that you’d like to share. Thanks very much and have a great day.
For more expert views on NDE 4.0, subscribe to the Floodlight Software blog at floodlightsoft.com.