Chris Yip  0:01  
Welcome to Tell Me More: Coffee with Chris Yip, the official podcast of the Faculty of Applied Science and Engineering at the University of Toronto.

Each month, I sit down with someone from our vibrant global community to talk about what places them at the heart of designing bold solutions for a better world. You'll meet students, professors, and alumni who are making a difference across a range of fields, including some where you may not expect to find them. My guest today is alumnus Don Scott, a graduate of our engineering science program. Over the past few decades, Don has made a name for himself in the field of marine technology, designing systems for all kinds of ocean going vessels. Most recently, he's been exploring the application of artificial intelligence to unpiloted underwater vehicles, creating marine analogues and the more familiar airborne drones we're starting to see in a variety of applications. 

Don, welcome to the podcast.

Don Scott  1:00  
It's good to see you, Chris. Thanks for having me.

Chris Yip  1:03  
It's gonna be great to catch up and hear about the amazing stuff you've been doing over the past few years. I always start these with this question about when and how did you realize that engineering was for you?

Don Scott  1:16  
Oh, we're going right back to the beginning, I guess, right?

Chris Yip  1:18  
All the way back, all the way back to grade one.

Don Scott  1:22  
To be honest, I really wasn't sure about anything until I was in my 20s. To be honest, I know I was drawn to adventure and being outside and the wilderness and being on the ocean. And I was always looking for a path that took me in that direction. I think I was pretty lucky. My father, he was actually a professor at U of T, like doing a deep ocean geology. So when I was in my high school, formative years, he was diving to the bottom of the ocean in deep sea submersibles. So that kind of really showed me the possibilities of what was out there, what you could do, like I really enjoyed the humanities, but I had this affinity for math and science, it's how my brain works. So that's sort of the path that I wanted to go down.

Chris Yip  2:09  
So why did you pick U of T and EngSci in particular? 

Don Scott  2:11  
Well, obviously grew up in Toronto. My dad was professor at U of T so it was - it sort of seemed like the natural choice. You know, back then, like EngSci the program was kind of legendary, right? (Chris: Right.) You know, it was a challenge. If you were like a high performing student in math and science, that's where you wanted to go. What I remember the most about applying for EngSci back then, it was the freedom that it provided, the two years of general study and then two years of specialization. You know, that really appealed to me, this sort of broad education. You know, my parents would say I was just putting the decision off for another 2 years but the reality is I wasn't ready to narrow my choices going into university, like I didn't want to close any doors. I wanted my university experience to open doors and that's really what EngSci provided.

Chris Yip  2:11  
And I'm trying to remember back then, right? Because it was a few years ago that you were going to EngSci and I was going into Chem. I don't remember there being an option in kind of, you know, we're going to talk about marine autonomy and underwater vehicles. I don't remember that being an option. You remember which option? I don't remember what option you picked.

Don Scott  3:20  
Yeah, I wasn't even sure which one. I had a pretty good idea that's where I wanted to go. I knew I wanted to work in the ocean environment and ocean science and marine engineering. And like you said, there just wasn't an option I, I kind of made my own to be honest. Because we sort of had that little bit of that freedom. I ended up taking the geophysics option, because that was the right combination of the hard maths and physics with the earth sciences, because that's what a lot of the marine telepresence was about back then. It was about deep sea ocean exploration but it also you know, you were studying EM fields, you were studying microprocessors, you were studying signal processing and all the electrical engineering side of things as well.

Chris Yip  4:07  
I know that between second and third year, you did something pretty cool. Do you want to share that with our listeners?

Don Scott  4:12  
I decided to take a year off. It wasn't part of the regular professional experience year but I knew I wanted to take the year off and actually go out and work. And I had the opportunity to go work as an intern down at the Woods Hole, kind of the deep submergence lab with that group. It was a pretty exciting time. They were right in the middle of their Titanic exploration phase. The team I ended up work working with was a group called Marine Telepresence and they were focused on this idea of remote ocean exploration. That's where I met I guess my my first mentors guy called Dana Yoerger. If you think of that time, Bob Ballard was like the face of the Titanic adventure and Dana was the quiet guy working behind the scenes making all that happen, the engineer. Well, him with a team obviously. You don't get anything done in the ocean individually. And so that was the group that I was working with. It really cemented my desire to work in this area. It was fun, exciting, just an overall incredible experience. It really galvanized my desire to work in this area.

Chris Yip  5:23  
To sort of move in and into the space where you've got such a challenge - I guess, right? - to work in the marine environment. Also back then, right? So technology obviously has moved a lot in the last, I'm gonna say three decades. 

Don Scott  5:37  
That's about right. (laughing) 

Chris Yip  5:39  
Were there other undergrads at Woods Hole at the time?

Don Scott  5:42  
The summer before I was actually working on a ship. It was a big oil bulk or cargo carrier that was transiting from the Canadian High Arctic over to Europe, to the North Atlantic, and that was quite the experience. So that's the story, it's like, I found out about being able to go down to Woods Hole by a telex on that ship. (Chris: What's Telex?) What's Telex? Yeah, everyone's gonna go "what the heck's a telex?". Anyway, yeah, so I wasn't I wasn't part of like the academic group. I was part of sort of the corporate group and so we were developing some technology and products for ocean exploration in the area of ocean acoustics. So that's sort of really how I got started in doing work in ocean acoustics.

Chris Yip  6:25  
Did you actually do another year after third year? Or did you just come back and then was straight third year, fourth year, and then right into work full time?

Don Scott  6:33  
So right at the end of that year with Woods Hole and Dana, I was working on a National Geographic project, and I met my second mentor, this guy called Marty Wilcox. And so he was in the corporate side of the world. He's actually a whole a really interesting character in himself, like he's the father of the real time fetal ultrasound. He had sold his business and had a no compete clause so we applied all of his medical ultrasound technology into ocean acoustics. So believe it or not, they're very closely related. Anyway so I was at a bit of a crossroads when I was coming back for third year. Was whether it was to continue working with Dana, you know, in the sort of more academic stream, or work with Marty in the sort of more corporate side. I chose to go with Marty, it was really r&d. We were doing a lot of creative research and development in ocean acoustics. And really, when I came back for third and fourth year, I was actually working for Marty during the school year, you know, as I could, and then I went back to work with him in between third and fourth year, and then continued to work with him upon graduation.

Chris Yip  7:42  
Was that still at Woods Hole?

Don Scott  7:45  
No, no, he was actually in Virginia so on the Chesapeake Bay. The work with Marty was all in ocean acoustics and ocean imaging, making some some really neat things, couple of patents, things like that. It was all very novel and interesting. My experience is I sort of kind of understand how I like to work on new things and we had sort of gotten to the point where we're building product and not really doing so much new innovation anymore. And I ended up being part of a panel for like, a product innovation thing for a company called Sippican and they said they were having an issue with some of their technology, and they were trying to think of new products. And I said, Well, why don't you just do this with your existing product? And they said, Oh, we talked to people, we we can't do that, and I say, Well, I'll do it. And they said, really? I said, Sure. Basically, I started up there sort of r&d innovation group, which like basically call it new product development. But you know, they were based down in Cape Cod and I didn't particularly want to move to Cape Cod, believe it or not. And so I sold them on this idea of remotely working, you know, and saying, if you want to hire the best people for this thing, they may not necessarily want to move to rural Massachusetts. You know, this is back in the early 90s. Fax machines, dial up modems and Federal Express was how we communicated. Back to the story. They said, Yeah, we'll try this innovation group for a year. Right? See how it works. I ended up working with them for about 10 years. Who knew that remote working worked really well and you didn't actually have to come into the office?

Chris Yip  9:30  
All right, let's go into the kind of cool stuff. So now Submergence Group and it's marine AI? Can you give us a little bit of a background on what that is and what it means?

Don Scott  9:39  
Yeah, so they're very related but two very different paths. How's that sound? So Submergence Group, is the parent company, we've made our job or careers, basically doing novel, creating novel vehicles for various navies of the world. Everything from full size autonomous submarines that are being used as target practice, to swim or delivery vehicles for the call to dry combat submersibles, or US special forces. And then you have marine AI, which is this new entity. I  had the general feeling that our engineering team was getting, you know, a little bored, at least well I was getting bored. How's that sound? (Chris: Oh, there you go.) But I'm probably the first one to get bored. I was starting to feel it. This was like six or seven years ago and I said, You know, there's this whole thing called artificial intelligence, I think we really should get into this. Let's learn about it, let's get smart about it, and let's see where we can apply it in the marine domain. So that's when we started really started pursuing what that meant, and where we could apply this technology, or this capability in the marine domain. And that was really the beginning of marine AI.

Chris Yip  10:53  
Everybody thinks about AI in the context, or at least in terms of vehicles, right? You tend to think of cars, because this is the classic ones of cars, driving on the road, trying to avoid hitting pedestrians or other cars and all those sorts of challenges. But in the marine space, as I think about it, you realize there's very different kinds of AI challenges compared to a car and naively I think it would be simpler, but I think it's probably not. 

Don Scott  11:18  
In a lot of ways it is simpler, because the decisions don't have to happen as quickly. We're moving quite a bit slower typically, however, I would never say working in the marine domain is easy. A car driving the road doesn't have to worry about the road undulating. You know so there are those types of challenges. Those are technical challenges, like in terms of operating within that space, and we have a pretty good handle on that. Where things are different, so we split the operating space into multiple regions so they all have their own challenges, like approaches to ports is incredibly complex. It'd be like, you know, driving in downtown in New York City or something. Actually, that's pretty well regimented so imagine, like... (Chris: Downtown Beijing) downtown Beijing, there you go. And then there's sort of like the coastal, the coastal area where things aren't quite as busy but you still have land and islands and currents that you need to worry about, and small boat traffic. And then there's the open ocean, which is relatively easy. Like you can see long distances, you know, what's going on, there's not a lot of traffic. So we treat those all very differently. But I guess, speaking more generally, like everyone understands what we call the rules of the road for driving a car, right? You know, stay in your lane, stop at a stop sign. (Chris: Generally.) Most people should, yeah, so there are an equivalent set of rules for operating a vessel at sea. And we actually call them the rules of the road, officially they're the collision regulations, but it describes the interaction between ships at sea. They're not incredibly complex. They're somewhat ambiguous sometimes, because a lot of it depends on situation but they are definable. So as long as you understand the situational context that you're in, you can actually make the right decision or should be able to make the right decision. That's assuming everybody's behaving properly, right? And that's where things start to get complex as when people or other vessels aren't behaving as they're expected. So you need to be able to understand that situation and react to it appropriately.

And I think in the context of boats, marine, it's also you can't stop.

Yeah, there's there's a lot of interesting dynamics at play, obviously, like you have restricted ability to move, you're what we call advance and transfer, which is, let's say you're turning to the left, you're in a car, you just turn left, right? In a boat, you keep going for a very long distance, at least in a large ship, you go a very long distance before you're actually going in the correct direction so you need to accommodate changes like that. What this all gets down to eventually though is going back to this idea of artificial intelligence, or augmented intelligence, just kind of what we'd like to use the term and sort of autonomous vessels. It's understanding the perception, the understanding of the situation, so you can make decisions, because that's sort of where I was getting to earlier is that we have the capability and understanding on how to make a correct decision but what I think is our biggest limiter right now is perceiving or understanding the environment that we're in, like, where the other ships are, where the other hazards are, things like that.

Chris Yip  14:46  
I guess that's part of the AI is that it has to sort of understand the forces that the ship is feeling at a given instant and being able to predict which way to go or how to adjust, right?

Don Scott  14:56  
Yeah, I've often described the ocean as this hostile dynamic environment. Someone - can't remember who wrote it - wrote it as a cold, indifferent force that is constantly trying to frustrate you. And so you do, you do have to live with that, right? It's going back to the analogy of cars or whatever. You would drive a very different way in the summer than you would in Canadian winter and hopefully your autonomous vehicle would accommodate those changes. So much in the same way that an autonomous ship needs to behave in a certain sea state as well.

Chris Yip  15:29  
And autonomy being different for different ships if you're dealing with a freighter versus a tug, or even a ferry that's going back, I think of the world's simplest probably autonomous system, which would be that Billy Bishop ferry that goes back and forth.

Don Scott  15:44  
One thing we didn't really touch on, I can tell you why the Billy Bishop is not autonomous and I'm pretty sure it has to do with regulation. We have the technology, what we don't have is the regulatory mandate to do something like that and I think rightfully so. I always talk about how AI and/or augmented intelligence, we're in the trust building phase, especially for marine autonomy. It's so new, and people don't really understand it yet so we're being I think, fairly rightfully cautious in terms of how we approach it. Like I said, we're building trust in these systems and might have little ways to go still.

Chris Yip  15:59  
I know and agree, I think you sort of nailed it. I mean, you know, imagine back when they first thought of cruise control in a car, right? You know, should we rely on this? And what happens if it does, you can't turn it off or something like that? Technology is getting better and better. I think part of the regulatory is probably the fact that you can have autonomous but the problem is when you have humans doing things separately, which are entirely unpredictable, augmented intelligence has to figure out the chaotic behavior of a random human deciding to do something silly, right? 

Don Scott  16:51  
That's the toughest part of our problem, actually. We can have a whole swarm of autonomous vessels and they'd all act perfectly with each other. When you put one human on a paddleboard in there, all chaos breaks loose.

Chris Yip  17:03  
That's right. So let's talk a little bit about the cool project that you've been working on that sort of just wrapped up in a sense, right? Which was really the Mayflower Autonomous Ship.

Don Scott  17:14  
Yea, it was a huge...I was going to say a huge project, but it was a lot of fun, is really what this project was. So our engineering facility is in Plymouth, England, Plymouth, United Kingdom, and the primary owner of the company is from Massachusetts from Boston area. So there's a bit of a connection between Plymouth, UK and Plymouth, Massachusetts, which is just south of Boston, by the way. Anyway, he was meeting with the Plymouth City Council like seven or eight years ago about how to commemorate the 400th anniversary of the original voyage of the Mayflower from Plymouth, England to the states. And someone came up with this idea of building a replica ship, and he quite, bit of an acerbic Boston guy, he said, it was a stupid idea. So someone said, Okay, well, what do you think we should do? And he said, We should build a ship of the future. We're going to build an autonomous ship that's going to recreate this voyage and it'll be ready and, you know, by 2020. And then he got out of that meeting, and he gave me a call. We've been working together for 30 years, this guy called Brett Phaneuf. Anyway so he gave me a call and said, Hey Don, can we build an autonomous ship that will sail across the Atlantic in four years? And I said, Sure, why not? (Chris: There you go.) And, and that was the genesis of the of the project. It became much more than that and you know, this was all happening at the same time, we wanted to get into artificial intelligence in the maritime domain. So it all sort of came together, as it was an excellent sort of proof of concept for us to develop that technology but both Brett and I have backgrounds in marine science as well as interest in advocating for preservation of the world's oceans. So it was an opportunity also to showcase disability to remotely collect data for ocean science. And the program took on this sort of dual nature where we were advancing the technology for developing remotely autonomous vehicles, marine vehicles, but also developing a platform for collecting ocean data. We didn't make it in 2020. We got about four or 500 miles offshore and developed a mechanical issue, (Chris: Okay.) And instead of pressing on, we decided the smarter thing would be to bring the boat back. And then last summer, we made a second attempt, and she sailed across the Atlantic. We made a quick stop in the Azores because of a mechanical problem, (Chris: Okay.) and then we actually made land in Halifax and then continued down from Halifax down to Plymouth, Massachusetts later on, so she made it across.

Chris Yip  20:01  
That's awesome. And I was gonna actually ask, can you give us a sense of sort of how, how many different prototypes did you go through in order to get to the version that made it across?

Don Scott  20:11  
You know, I had mentioned Submergence Group, the parent company. So we have, you know, a couple of decades of building ocean going vessels, (Chris: Right.) so we, we had a pretty good sense that we were going to be able to build a boat that would survive the difficulties of that type of a transit. So we weren't too worried about that. What was all new was the perception, being able to see what was around us and understanding what was around us. And then the ability to make decisions in terms of navigation and safe operation. I really don't like what I call PowerPoint engineering where you...I got this great idea but you don't actually do anything with it. And that's really the ethos of our company is we really like building things and putting them on the water and working with them in the water because we figured that's where you learn, that's where you experience. That's where you test everything and understand what the problems are. So that was the approach we took, which was we're not going to just think about this and do it all the simulations and everything, we're actually going to build the boat, we're going to put it in the water. It's funny, our young engineers, were We're not actually going to do this, are we? Are we actually going to...like we're nearing the end of the project, right? Like, it was like, Okay, we're getting ready to launch date and there was this real reticence from some of the younger engineers like, are you? Are you? Are you actually going to send it? That's why we built it, guys. This is why we're doing this. And he said, Yeah, but it might not come back. I said, I know but you know what? Can't stay here forever. So that's that's definitely how we work. You need to actually get on the water, you need to get wet, get seasick, you know, learn what you're doing and work in that environment.

Chris Yip  22:01  
And I think you mentioned along the way, you're also doing ocean sampling, right?

Don Scott  22:05  
Yeah so I think that was kind of a unique part about Mayflower is that we had a fairly significant payload area. We were pretty much open to anyone who wanted to put anything on the boat. It was kind of funny, because we'd say Okay, so you know, the scientists would say, Oh, I can put an experiment on, great, how much space do I get? And we say, Well, we'll give you, I don't know, like a square meter. Is that good? And they thought they were gonna get, you know, like, space for a little mini computer, anyway we gave them lots of space and lots of power and things like that. So we were, you know, constantly sampling the water for water chemistry. There was actually a really neat experiment on this called hyper taste was basically a tongue. It was developed to look for fake wine and whiskey and they adapted it to look at the chemical properties of water so they had to take this sampling device that was designed to sit on a desk, and marinize it. So that was a big thing. So some of the other experiments were like marine acoustics. We had transceivers in the water listening for marine acoustics and using machine learning systems to identify different types of whales, things like that. I'm forgetting half of them, and I apologize to the ones I forgot. Like I said, one of the main focuses was to develop this as a platform for ocean science.

Chris Yip  23:28  
What's the next step for Mayflower? Actually physically, where is it right now?

Don Scott  23:32  
So as I mentioned, my first mentor of my careers, Dana Yoerger at Woods Hole. He's still there, incredibly active, and the boat is actually with Dana. (Chris: Cool.) So Woods Hole is not taking over, but they're going to be operating the vessel for ocean science and research. It's kind of full circle for me. (Chris: Yeah, that's awesome.). I feel I'lll keep going in circles but I've come to a nice sort of full circle moment with Dana. He's really what got me into this career. Like really, like I said, really galvanized my interest in doing this work in marine engineering and now I'm working with him again. It's been great. 

Chris Yip  24:12  
So what's the next step?

Don Scott  24:13  
Oh, so we're outfitting the boat. She's gonna go out and voyage in the spring as part of a research program off the coast of the US. I'm trying to think exactly what they're doing. I believe they're looking at the biology of the ocean. I think the idea is that they're looking at following the apex predator, following the sharks, which leading them to smaller fish, leading to smaller fish, leading to smaller fish, that idea. They're looking for a vessel that can move and react to situations and it can help them sort of get to where they want to be so they can actually do the ocean science. It's actually a really interesting time to be an ocean engineer. What we're capable of doing now just wasn't possible three or four years ago. The increases in compute power of edge devices, the advances in machine learning and deep learning, our ability to operate remotely over satellite communications. All of these things are sort of coming together this convergence point that's allowing some pretty amazing things to happen. Me and my team we're pretty excited being really at the forefront of a lot of this technology as this small, young dynamic team, we're able to act very quickly and implement things very quickly. So there's lots of interesting things going on. So yes, the technology that we've developed for Mayflower and marine AI is definitely being pushed into the broader marine domain. We're working on a project now with DARPA, or we're doing the executive autonomy for it's called the No Manning Required Ship, NOMARS. So the idea is truly a ship of the future. If you were to design a ship, that did not require a human onboard, what would it look like? And then they're actually building it. And so part of that is, okay, if you're actually going to build a ship that has no one on board, you need to be able to operate it and that's where we come in, right? We provide the executive autonomy for operating this vessel. And I think that's where the future is for a lot of this technology. We're not going to be replacing people but I think what they need to start working on is how could you have a ship without people on board?

And as you were talking, I started thinking about all the other areas where we're getting autonomy for exploration and stuff and I think we've talked about this before, which is all this sort of, you know, space exploration on other planets, right? You already have the autonomous drones, in a sense, flying around on Mars. You know, they're starting talking about other planets where they definitely have a marine environment where they're looking for underwater submersibles to go in and start exploring and...

That's actually Brett's dream. Is to...is it...Asedulous? I'm going to get the name wrong.

Chris Yip  26:50  
The moon? Oh, yeah. The moon.

Don Scott  26:52  
The moon, the water moon, yeah, I think he's talked about this a couple of times and I think one of his dreams is to put a autonomous marine vessel on a spaceship, land on a remote planet and explore the ocean.

That, to me is just astounding to think about.

What's astounding is to think that we could actually do it.

Chris Yip  27:12  
In your lifetime. 

Don Scott  27:13  
Yeah, we could. Like there's no reason why we can't. We have the capability, we just need the will. And we need someone who's willing to take the risk to do it. This is why I like working with young engineers, right? Because they'll hear this idea and they'll say yeah, why not? What I worry about is that we lose that ability to dream like that and have that vision. And I think you'll stay successful as an engineer if you maintain that, that ability to do that.

Chris Yip  27:43  
I think that's the that's the key to this whole process, right? Success is all about having fun, and continue to have fun as you go through all the things that have happened. Hey, Don, this has been an absolutely...just a blast to to catch up with you and hear about your, just your journey, right? For the listeners. I mean, Don and I have known each other for many, many years now and it's been great to catch up and see where you've gone thinking back to where we were together in high school. But now what you're doing now is just astounding.

Don Scott  28:15  
Well, let me turn it back on you, Chris. It's been amazing to see what you've done with the U of T program. I love the idea of the way you connect with the students. So congratulations on everything that you've been doing as well.

Chris Yip  28:27  
Again, thanks. Thanks so much, Don, for taking the time today to be to be part of our podcast.

Don Scott  28:32  
Thanks, Chris. It's been a lot of fun talking with you.

Chris Yip  28:35  
Thanks again for listening to Coffee with Chris Yip. If you want to catch up on past episodes, or make sure that you don't miss the next one, please subscribe. We're on Apple Podcasts, Spotify, and more. Just look for Coffee with Chris Yip. You can also check out @uoftengineering on Twitter, Facebook, Instagram and LinkedIn for more stories about how our community is building a better world. And finally, if you'd be inspired to join us, we'd love to welcome you. Whether you're thinking of taking a degree or working with us on a research project, you can find us online at engineering.utoronto.ca. Or you can visit our beautiful campus in Toronto, Ontario, Canada. I hope I can join you for coffee soon.

Transcribed by https://otter.ai