Now that the Mayflower Autonomous Ship (MAS) has successfully completed its maiden voyage across the Atlantic, it’s time to check in with Brett Phaneuf, Managing Director of the MAS project and Edward Milne, Technical Sales Manager for the EMEA region at Hexagon’s Autonomy and Positioning division. They provide fascinating details about the voyage itself, share parallels with the original Mayflower voyage in 1620, and they also discuss what’s next, both for the MAS project as well as the marine research industry at large. Brett and Eddie also discuss how data from autonomous marine vessels is becoming more accessible, accurate and reliable and even wax philosophical about a not-so-distant future when fleets of autonomous vessels will be linked together in an AI-driven sea-faring network, accomplishing common research goals.
BK: Hello and welcome to this episode of HxGN Radio. I’m your host, Beth Keener. And in today’s episode, we are revisiting the Mayflower Autonomous Ship and talking about its maiden voyage across the Atlantic, with Brett Phaneuf, managing director of the Mayflower Autonomous Ship Project and president of Submergence Group. And we’re also joined by Edward Milne, technical sales manager for the AMEA region at Hexagon’s autonomy and positioning division. Wow. You guys both have quite the resume just in title. So, thank you both for joining us today.
BP: You’re welcome. It’s great to be here.
EM: Hello, Beth. Yeah, great to be here.
BK: Awesome. Well, welcome, guys. Let’s dive right in. And yes, I was being punny with that.
We’ll start with Brett. Brett, okay, so we had previously chatted and we chatted about the Mayflower Autonomous Ship, or MAS, as you call it. And while the vessel was still getting ready, that’s when we had our conversation. And it’s travelling from Plymouth, England, to Plymouth, Massachusetts. So how does it feel now that it’s arrived?
BP: Oh, I get asked that question a lot and I’m never sure what to say. And largely it’s a great relief. We just spent a lot of time focused on a lot of effort to get it to this point. So, we are I mean, it’s everybody expects me to be jumping up and down with excitement. It’s just really more catharsis, right? So, it’s just, aah, I can take a breath after five, six years of working on this project.
And then the other feeling is now that the ocean didn’t claim it for its own, I have to continue, I have to continue to deal with it. So, it’s relief that one aspect of the work is over and then trepidation that there’s more to come. But all in all, it’s overwhelming, the good feeling.
BK: So, when you say the ocean didn’t claim it, where is its resting place right now?
BP: Well, right now it’s at Woods Hole Oceanographic Institution in Woods Hole, Massachusetts. And we got into Plymouth by a rather circuitous route. But then after it spent some time there, we moved it down to Woods Hole to install some new instruments as we’re going to go out and do a little bit of oceanographic research later this fall and try some new things with the vehicle.
BK: Exciting, exciting.
So, what did the journey look like for MAS to travel from the United Kingdom to the United States, and are there any parallels with the original Mayflower journey 400 years ago?
BP: Oh, well, yeah, there’s a few parallels I suppose. We had to turn back at least once, as they had to turn back twice. So, a couple of false starts is fair game. And then everything went really well at the beginning. We got, I don’t know, 10 days, 12 days out, a little more, and the ocean cooperated. It looked like a lake. We’d look out into the cameras through the Mayflower because we monitored it 24 hours a day, and it looked like just a millpond out there.
And then we developed a couple of little issues that forced us into the Azores, so we had to stop in Horta and resolve a problem. And then it was about another thousand miles west of the Azores, 1,100 miles west of the Azores, on our way to Virginia, actually, which weirdly, is what the original Mayflower was supposed to do as well in 1620. So, we had decided to go to Virginia because we missed the 400th anniversary in Plymouth, and we wanted to get to D.C. ultimately. About 1,000, 1,100 miles west of the Azores, we ended up developing intermittent faults with a couple of the instruments on board, and we didn’t know what they were. And the closest landfall at that point then was Halifax. So, we decided to turn to Halifax and we got all the way up there. We went through some pretty vicious weather, but we were able to get into Halifax and then make some repairs. And then down to Plymouth, Massachusetts, without incident.
Interestingly enough, there’s sort of some debate about whether or not the original Mayflower, when it had been blown north from its intended landfall in Virginia, it actually had gone along the coast of Nova Scotia and stopped for provisions. I won’t comment on that, but because I don’t know. But if so, then that’s another parallel.
But yeah, it was a little bit less of a direct route than we hoped, but a touchdown is six points no matter how you get the ball across the line. So, we’ll take it.
BK: Okay. So, there were quite a few parallels. Brett, thanks for sharing that. An exciting journey too.
So, Eddie, now moving over to you, what role did Hexagon | VERIPOS equipment play in the journey, and how did the different pieces of technology work together to create an autonomous solution?
EM: Okay. Thanks, Beth. Yeah. So VERIPOS, we supplied the GNSS system for the Mayflower Project. So, the system provided the accurate, robust and reliable positioning for the vessel through its journey across the Atlantic. So basically, allowed the autonomous system on board the vessel to navigate either its preprogramed path or also, if it autonomously decided it had a different path to take, to allow it to actually navigate those paths as well to make sure it basically got from the U.K. across to the United States.
So, with regards to the different technology, um, the multi-constellation and multi-frequency GNSS receiver that’s integrated into the LD8 units that were installed on board, those coupled with the Apex 5 corrections and enabled the positioning to use more than 30 satellites any one time. And this allowed the vessel to be positioned continuously within four centimetres throughout the whole journey across the Atlantic.
So, the advanced interference and spoofing detection on the LD8 unit was also able to identify any interfering or spoofing signal and warn the vessel systems of its presence, allowing the autonomous vessel systems to take appropriate action and to mitigate any problems in this regard.
With regards to the Apex 5 PPP correction service, not only does it remove the errors in the GPS, Glonass, Galileo, Beidou and QZSS satellite systems, it also provides further cheques to the GNSS positioning to detect any erroneous data from specific satellites or constellations and removing these from the positioning solution.
The VERIPOS corrections were received through multiple communication satellites simultaneously, and also use diverse communication systems to ensure the correction of signals, what all of us received on the vessel, no matter the location of the vessel or the environment that it was in.
And finally, the quantum software is used to remotely configure the LD8 units for any changes that are required at sea. But its main task was to provide detailed and precise status information to ensure monitoring personnel regarding the performance of the LD8 units and status of its solution. So, with us focused on any fault condition like the end user remotely monitoring the system to identify any issues easily and take appropriate action to mitigate and resolve these in good time.
I think the main point to note in this is that there’s no personnel on board an autonomous vessel to resolve the issues, observe if the vessel’s offshore. So, this is why a robust system such as the LD8 and the Apex 5 service are essential to continue autonomous operations.
BK: So, Eddie, what you’re telling me is Hexagon had a really small job.
EM: Yeah. Well, we are part of a whole system. And it was a major, you know, the absolute positioning is the start of everything. If you don’t have position at sea, you’ve got no landmarks, nothing to actually measure yourself against. So, it’s not like you could follow a map when you’re on a road. It’s the GNSS positioning is absolutely essential back for offshore positioning. And so, yeah, that’s where the start of everything is for all the rest of the sensors that are on board the Mayflower.
BK: Pretty incredible. And so, I know that the primary goal of the entire voyage was to learn. Now, we’re not only learning about technology enabling autonomy, but we’re also learning about ocean and climate science and more.
So, I guess this would be a great question for Brett. What did you learn? And then Eddie, if you wanted to piggyback on that, I know from a technology standpoint, it’s really very interesting what you learnt as well.
BP: Sure. So, I mean, gosh, we learnt a lot of things. We’re still analysing the data. We’re nowhere near ready to publish it. We collected so much information about marine mammals at sea, the elemental composition of the ocean, fluorometry, temperature, depth, salinity, all sorts of different things that we wanted to integrate and publish, but we just couldn’t get enough bandwidth through the communication satellites we had to send it all off. So, we’re in the process now of sort of decompressing all the data from the archive on the ship. We’re going through it all and parsing it out to various scientists that want to look at it.
But I’d say, you know, really what I learnt, my personal takeaway from this is really more in the technological area other than, you know, not to steal Eddie’s thunder there. I mean, I’m interested in, you know, I guess the big takeaway for me was that the ship itself is always going to be a ship. It’s always going to have problems. And design for our reliability and resilience at sea is really important. And Eddie alluded to that a few minutes ago. There’s no one there to fix it.
But also, that these things are possible, that we’re not in the realm of theatre or fantasy. These things are possible now. Long transit, trans-oceanic voyages, long deployments at sea into the deep ocean with unmanned systems are doable, and they drive down the cost immensely to do the kinds of research we need to better understand the climate dynamics of the planet and to really understand the ocean.
So, there’s lots of the ocean we’ve never looked at. And every time we look in the ocean, we learn new things. And there’s no exception on this journey. We’ve learnt lots of new things. But for me it’s really about the sort of underscoring the ability to do research in this new way and have it produce meaningful results. I mean, that’s the big takeaway. What we need is more ships, not Mayflowers per se, but more ships like Mayflower that are unmanned, autonomous and capable of collecting vast amounts of data. And we’re moving now to the point where that can be processed, that data can be processed on the ship, on the edge, and create information that’s immediately indirect, directly actionable by scientists on manned research vessels, of which we have few by comparison to the number of smaller unmanned systems we could deploy at a significantly lower cost. I think that’s what this mission is really about, is moving us out of the sort of realm of fear and fantasy. And the idea that it can’t be done, the idea that it’s dangerous, the idea that it’s prohibitively expensive, none of those things are true. And so, we’re hoping it’s a watershed moment for moving forward, both with regulators and scientists alike, to get these things out in the ocean and learn more about our planet.
EM: So yes, Brett says, you know, this autonomy is not something that’s, it’s something for the future. It’s happening now, you know. Autonomous and remotely operated ships, as you know, they are growing. They’re growing in numbers, but also in the size of the vessels as well, that are being kind of belting at the moment.
So, for me, Brett mentioned the data links and the data coming back from these vessels and being able to actually do all the computations on board the vessel and limiting what’s coming back is important across these satellite communication links because the more data you bring across, the more the satellite links basically cost. So, you know, reducing that bandwidth is very important, but also making the system as robust as possible. GNSS been out for many years now and people basically take it for granted that it is going to work. But you have to, you know, it works when you build it into the system to start with that, you know, you’ve got the robustness and the redundancy in the systems because you’ll always get some kind of failure offshore and you just have to make sure that you’ve got enough systems kind of there and the systems that are there are tested and you know that they’re robust to make sure that they’re just, you know, they’re not going to fail. But if they do, then you’ve got things there to back them up and be able to make sure that the positioning continue. At the end of the day, if you’ve got a vessel in the middle of the ocean that loses its GNSS and you lose its position, trying to find that vessel’s going to be a very interesting project.
BK: To say the least, right?
EM: Yeah, definitely. So, it’s, yeah, the main thing is making sure that you don’t lose our positioning to start with.
BK: So, you guys are mentioning and touching on the data. And as you had mentioned earlier, Brett, there are so much of it that was collected that couldn’t be simultaneously transferred back to where you guys were collecting from, and so you’re just still digesting it now. And I also know that with all of this research in the ocean, that we are constantly trying to understand how the ocean can work to our benefit and life on land, but then also how to preserve our ocean. And so, the data you collected was for marine mammal health, climate research. And then also, could ocean waves be a sustainable energy source.
So, Brett, I know that there’s still so much you are digging into, but can you share any early insights of the research that you have from MAS?
BP: Well, I’m sorry to say I can’t right now. I’m a little bit detached from it because I’m on holiday, believe it or not. But we’re still deconvolving the data. It was quite voluminous. We did get back in real time sort of basic information about water temperature and salinity, and occasionally we could pull some acoustic data off, and when the system keyed that there were whales nearby. But we just haven’t had the time now to deconvolve it all. Of course, the ship came in and we pulled all the data out. There’s still more to extract, actually, I found out yesterday, right going into everybody’s holiday season. So, there’ll be a lot more to tell in September.
And again, you know, not to veer away from ocean health, which is a large part of why we wanted to do this, what we have learnt is a tremendous amount about building stable software systems, containerised, compartmented, highly modular reconfigurable software that can deploy on a minimal compute system on the edge. That sounds like gobbledygook, I suppose. But we were able to do some incredibly sophisticated sort of updates to and critical failure management to the system as a result of this sort of leading edge design of its compute architecture through a very low bandwidth satellite connection.
And the other thing we learnt that’s really important and I don’t want to downplay this, aside from the tremendous value in having precise positioning, which means without which the data basically is you have to have position in time. And if you don’t have that, then it’s of no value whatsoever. And that’s, you know, depending on VERIPOS for that has made it, you know, very reliable part of the system, being able to know where we are and what we got. But beyond that, communications is critically important.
Right now, there’s a lot of angst about the deployment of these types of things, sort of unwarranted fear of potential accidents. But these are still quite small compared to the ocean and the ships that traverse the large ocean. But the communications being able to monitor the vessel real time, even though it’s through a low bandwidth system, we expected to have about 50% uptime on the satellite, so we had to build a very smart vessel. But in fact, it was north of 95, 97, maybe 98, 99% uptime, we’re looking at the metrics now. And it was just an incredibly stable connection, more stable than this link that I have with you guys for this discussion, it seems. And, you know, that’s a huge success technologically that people underplay all the time. I think we take this level of communication and interconnectedness in our society for granted, and it’s certainly not something mariners take for granted.
But I think we can now, actually, and it’s getting better—and it’s not getting better incrementally. It’s getting better by orders of magnitude on a week to week, month to month basis, with a concomitant drop in cost, or at least a flat cost relative to vastly increased bandwidth. So, the utility of these vessels and the capability to understand what they’re doing and to extract the data in real time is literally now upon us. So, the next time we do a long transoceanic voyage, we’ll be able to actually stream the real-time data off the boat so that in real time we can integrate it to climate models and give it to scientists, so there’s no latency between when we get to shore and they get the data. They’ll get it in the moment, broadcast right off the ship, at less cost than if we had to do it that way today with an order of magnitude less bandwidth.
So not to dodge the question on early insights, but those are my big takeaways, right? So, the technological leaps are going to allow us to get the data and to validate the data and compare the data sets with other either extant data from the same region or from other devices in the same region, to do cross calibration between various vehicle sensors and systems so that it’s all sort of comparing apples to apples and to ask the ship really profoundly complex questions that we’d like it to get data to resolve because of that increasing bandwidth in terms of communication. So, hold that space but give us a little more time, and I think it’s going to be very impressive what can be done in real-time. So that was a long-winded way to not answer your question.
BK: It was perfect though, because honestly, it directly ties into Eddie here, who I believe, you know, with running the VERIPOS technologies through Hexagon has probably more to add in regard to how precise positioning empowers these marine research groups, like those working with MAS, but then also just positioning in general and how that continues to play into to the overall success of projects like these. Eddie, do you want to weigh in?
EM: Yeah. So basically, as Brett mentioned a few seconds ago, that being able to geolocate all the data that the marine researchers are actually logging is essential. If you aren’t actually locating where your data is, then if you’re doing wildlife monitoring or, you know, defining migration routes or looking at populations of wildlife or looking at the sea temperatures in certain locations, etc., then you’ve got to geolocate all that information. And the more accurate that you can do that, the better.
Also, if you’re actually trying to map any sea floor or looking at sea levels, then you could use the precise positioning, the vertical element of precise positioning, to actually measure the height of the vessel and the mean sea surface at any one time or where the vessel is. So again, you can actually use all that data to take your data means and see how it varies over the time.
So yeah, the precise point positioning, as you’re getting farther and farther down into the lower levels of the accuracy, then the data is becoming more and more accurate and reliable.
BK: Okay. So, MAS has received a lot of public attention. Were there any surprising groups who reached out to discuss the potential of autonomous technology in their own marine applications?
BP: Hmm. That’s an interesting question. Surprising groups. Gosh, we, I mean, there were some by which we were flattered. We talk a lot with the U.S. Coast Guard and the U.S. Coast Guard Research Centre about how to implement autonomous technology in furtherance of the U.S. Coast Guard’s mission, not just in protecting the coast, but also in doing scientific research at sea, specifically focused around sort of mitigating the risk of life at sea and search and rescue. We’ve got some contacts from my favourite one, NASA, because I’m, you know, I sort of do marine because I can’t do space. So, I’m a frustrated space enthusiast that didn’t figure out how to get into that field. So, I do the underwater part, which is sort of the sometimes thought of as the poor cousin, but I think it’s harder. But talking with NASA’s been really exciting and the JPL [Jet Propulsion Laboratory] guys about how we could use the ship and others like it interlinked, almost as if the idea that, almost like there’s satellites that float in the ocean in communication with a web of smart satellites with edge processing in space, the idea being that while these satellites are doing research, ocean and state of climate, their AI [artificial intelligence]systems, which may be individuals or distributed across a space network, a space based network, can communicate with the ship and other ships’ AI systems that are similar, and then come up with common research goals and do anomaly detection and re-task each other dynamically without human input to investigate things that are unusual. So that’s kind of interesting, right? It’s a little bit of a Skynet feel to it, but not intended to, for sure.
But yeah, those are really exciting. We got a lot of requests for people to study mammals, to help the marine mammal studies. That’s interesting. It wasn’t unexpected. We didn’t get anything, really—can’t think of anything really crazy or wacky that we got. I kind of wish there were some wacky ones. That would be fun.
We’ve got asked about maybe doing something in some really remote areas using this ship, Mayflower or something like it, using its AI to map Tristan da Cunha, which is the most remote populated island in the world. It’s only reachable by ship from South Africa. And I guess the ship goes every three weeks back and forth. And I kind of want to go there and do that one. But the whole point of an unmanned vessel is, I don’t go there and do that; the vessel does. So, if you like to travel, it’s a bit of a letdown. But other than that, it’s—but nothing totally crazy.
We did get asked if we could go up to the Arctic and do some work up there, climate research up there. But it’s so weather dependent. The ship wasn’t built for ice and we’re going to try to avoid the Titanic’s fate at all costs. So probably won’t go into the high Arctic, but that would be exciting. The last time I was up there, it was the first time I’d seen a walrus in the wild, right next to my room. So that was interesting.
BP: No. But I wish I had more to say that was insane. But no, I think that might be bonkers enough, so I’ll leave it there.
BK: Eddie, have any final thoughts for us? I’d love to know your thoughts.
EM: Well, yeah. I’m with Brett there. Thankfully, we’ve got no insane ones, but, you know, autonomy and autonomous ships, it’s just growing. And so, yeah, there’s the Mayflower certainly helped promote what VERIPOS can do in the autonomous frame. And so, yeah, we are, you know, we are still talking to people about creating a more commercial kind of side of things on autonomous vessels and fleets. So yeah, it’s definitely, it’s growing and it’s going to be just more and more, you know. There will be people at sea for a long time, but this is probably the fastest growing industry for the future.
BK: Well, you’re leading me right into my final question. And I just want to know what you guys’ parting words are for what we can expect in the future or what we should look forward to in the future for autonomous travel throughout the ocean, and anything that you think will be helpful as far as continuing to move forward in protecting our ocean life, and how those two can connect and will continue to connect.
BP: You know, that’s a really interesting question, because I have been perplexed by something, I guess, for a while since we started doing Mayflower. So, working with IBM, we set up a web portal that most people have seen, where we broadcast the camera feeds and other data from the ship in real time that we can support with the satcom. And, you know, I’ve said this before. At 2:00, 3:00 in the morning, I’d wake up, “I wonder what’s happening to the ship.” You know, I’d roll over, I’d log in, and I’d see if everything was okay. And, you know, at one point there were 30,000 people watching. I could see how many people were looking at the camera feed.
BK: That’s incredible.
BP: Yeah. Well, I kept saying to myself, who are these weirdos, right, that are watching this from all parts of the world? And, you know, and I’m one of them, right? And I talked to so many people who in government offices, in businesses that said, “Oh, you know, we get emails, when is it coming back online? We have it on all day. It was great. Everybody was just watching it.”
And what I realised, I kind of think is that first and foremost, people have a very anachronistic view of the marine industry, right? I think they all sort of feel like it’s Mutiny on the Bounty or Das Boot. Right? There’s no real understanding by the general public of the level of technology and complexity and risk in the marine enterprise. It is incredibly safe, though. It has become incredibly safe. But there’s this sort of almost romantic allure of danger at sea that’s not well informed by the state of the current technology and industry. And, you know, we depend on it; 90% of all freight moves at sea, right? Everything you order on Amazon has been on a boat, right? But people, by and large, are detached from it, right? It’s a small sector of society that works at the sea, but all of society depends on the sea in many different ways, for trade, for food, for climate modulation, all of these things.
And I think the way you, what I think is that people have a basic, profound fascination with the unexplored and with the ocean in general. There’s something intrinsic to humans and it isn’t fed well. And, you know, a similar fascination with the unexplored with space, but it’s much more fed in the popular psyche, right, with science fiction, among other things. And so, I think if you want people to care more, you need to show them what it is they’re caring about. You need to find ways for them to experience it, the vastness of it, the complexity of it, the power of it, that don’t necessarily put them in peril or cost them a fortune to experience. And so even something as simple as a camera feed of the middle of the Atlantic Ocean at night with a full moon in front of the vessel, or when the boat was crashing through 12-foot waves and several waves actually overtopped the entire vessel. So at least at one or two points in our mission, the entire Mayflower Autonomous ship was under water, right?
BP: Yeah. And that’s okay, right? We understood that that could happen. And we built submarines for a day job, so we know how to deal with it. Didn’t go very deep, but it was awesome to watch, right, when the wave rolled over the highest camera on the highest mast.
But there’s something very engaging. And I think that, I don’t mean to be, you know, I’m not a romantic by nature, but I think that that’s one way you get people not necessarily excited, but fascinated, and they can see it and they can experience something that they probably have no other way in their life to experience ever. So, I think we don’t do a very good job of that generally. And so, I took that away from it. And I think that’s an important factor in instilling respect for and a desire to care for the ocean in the average person. I don’t know. Maybe that sounds trite, but that’s sort of one of the things I took from this whole experience.
EM: Yeah. So that’s all good stuff from Brett there. And trying to follow that up, it’s, you know, I think from my sight, from my thought, it’s condoned on the commercial side. It’s as Brett said, everything that you can probably purchase, certainly, you know, in the U.K, you know, island state, and even kind of globally, has probably all been at sea at some point. So, the amount of things that you buy are transported by sea. And that’s probably, that’s one of the big things that autonomy will come into is making these vessels, which are massive vessels, autonomous and, you know, taking the need for people to be at sea, which, you know, obviously to improve safety for sea. You know, it’s a very safe place to sea, but it can have its dangers. And taking the person out of there will remove the risk to any people.
So, I think that autonomy at sea is something that’s definitely, it’s coming, it’s here now, but it’s only going to grow. I think autonomy throughout all parts of our lives is going to grow. I guess that, you know, international laws and all this kind of stuff is going to have to kind of catch up with this autonomy at sea, because there’s so many different organisations and jurisdictions that have to be kind of overcome for vessels to be roaming around the sea without any humans on board. But it will come.
BK: You guys have, this conversation could go on forever. I feel like there’s so much to learn and so many fascinating parts of what you guys have done. And it’s wild to be having a follow up conversation about the full voyage of the Mayflower Autonomous ship. And I just wanted to thank you guys for your time today and for sharing those insights and your thoughts on how this impacts technology, our ocean life, and truthfully, the world, and then the potential of a future autonomous travel in the ocean. So, thank you both for joining me.
And for our listeners out there, you can learn more about the Mayflower Autonomous ship at mas400.com, and about VERIPOS, assured positioning solutions, at veripos.com. Thank you, guys, so much for listening.
And gentlemen, it has been a pleasure.
BP: Pleasure’s all ours. Thank you very much.
EM: Thank you, Beth.