WHOI engineers Amy Kukulya and Roger Stokey discuss REMUS SharkCam


Interview with SharkCam Engineers Amy Kukulya and Roger Stokey from Woods Hole Oceanographic Inst. on Vimeo.


Question 1 – What is REMUS SharkCam and why is it unique?

Roger: REMUS is an underwater robot, we call it an autonomous underwater vehicle. That means there's nobody operating it, nobody telling it what to do, it operates on its own.

Amy: REMUS is good at doing a lot of things, and more or less it has been historically pre-programmed and given a set of instructions before going in the water to accomplish some tasks, and it's really good at mapping out the ocean, and looking at objects that aren't moving, but this project sort of took things to the next level, to make REMUS SharkCam a more dynamic autonomous underwater vehicle.

Roger: For this SharkCam configuration, we've specially outfitted it with some cameras and some sensors that allow it to follow a tagged animal. It’s the first time we've tracked a living thing.

Amy: The shark's not going to cooperate and say I'm a shark, I'm over here, so there's a lot of processing on-board the vehicle that's happening in order for the REMUS SharkCam to be able to keep up, and to know that the information it's hearing is in fact the shark that it's supposed to be following and filming.

Roger:  And that was a real challenge, figuring out what the animal was going to do. So we had to get inside the animal's brain a little bit. And it's based on putting a transponder on an animal, and then following it with the vehicle, the vehicle pings the transponder, it gets a response, and it knows the range and direction and it just follows it.

AMY: I knew just as much as the average person about shark behavior, but to actually develop a technology, you have to take a crash course in shark behavior, and as much as we can possibly learn, what we can cram into ourselves and then apply that intelligence into the vehicle.

The algorithms that are in REMUS that control what it does, needed to be a bit more dynamic and nimble at making decisions on its own

Roger: The initial problem was actually getting it to work at all. You have to be correct in three dimensions: latitude, longitude, and depth.

AMY: it has to solve this math problem to figure out that that information that comes back is really a shark.

Roger: We'd get to a spot, and that was three or four seconds old, but it's already too late.  So we had to forward predict where we thought the animal would be and go there, and not just where the tag was – we had to guess where the animal would be.

Amy: REMUS SharkCam is so cool because it takes people to places they can't go. Whether they’re inhospitable, or trying to blend in with the environment to get the inside scoop. Also, this technology of REMUS SharkCam is developed at the Wood's Hole Oceanographic, and it doesn't exist anywhere else in the world.

The ability to be able to do this is unprecedented, to be able to take an underwater robot and to give it a set of instructions, and for it to be able to act on those instructions and take us places never before seen, and to show the world what's happening right underneath our noses, so we can all have a greater appreciation for our own planet. This technology is the tip of the iceberg, in the sense of where we can go and what we can learn about our own planet.

Question 2: Describe the first shark attack on REMUS SharkCam.

Amy: When we started the first mission, there was just a lot of energy on the boat. People were really jazzed and excited. So there was a lot of anticipation, kind of like butterflies in your stomach before a big game, it was just fun and natural that way, but there was no anticipation of fear yet at that time.

Roger: There's always apprehension. Is everything going to work? You've spent days, weeks, hours, months, preparing for it, and finally it's show time, and is it going to work? What's going to go wrong? There's always something you don't expect, and are we ready? As Amy said, there's always butterflies.

Amy: We put it in, everything's going smoothly, the REMUS SharkCam picked up the signal on the tag right away, and we knew from our on-board ship-side computers that we were tracking, and things were working, so we were immediately excited. Then some strange behavior started coming through in our real-time data,

Amy: And Roger and I were looking at each other, and the cameras were rolling, so we didn't speak to– we learned body language pretty well, observing and seeing if this is something we should be worried about, what's happening?

We hadn't talked about what else could be happening, and we were concerned about REMUS, maybe something got stuck in the prop, some seaweed.

Roger: It was very clear something was happening down there, but we didn't know what. We had an idea that the sharks were doing something, but we really had no idea, just what they were doing.

Amy: And the vehicle finished its mission, and the sun was setting, the vehicle comes up to the surface and floats when it's done with its mission, and we went over to recover it, and we looked up, as the vehicle came up out of the water

Roger: The image on the vehicle was of teeth marks was unmistakable. Very much semi-circular, you knew in an instant exactly what had happened, there was no mistaking it, no doubt, it was just oh my god, this actually happened.

Amy: There were these massive bite marks, it was clear as day. Everybody's jaw dropped to the deck, but there was without a doubt, a shark had bitten REMUS,

Roger: The reaction to the video on behalf of everybody was just absolute shock. Everybody jumped about six feet.  Nobody expected to see anything like this.

Amy: Keep in mind, REMUS has been doing ops in a marine environment for twenty years, we’ve never been attacked by a shark. We were getting a little bit purposefully trying to get close to them this time. It still wasn't on our radar that we were going to have problems with that

Question 3: How can REMUS SharkCam help scientists learn more about shark behavior?

Amy: One of my favorite parts of this job, is that we're continually changing our applications, and we're learning a little bit about a lot of different things. In this case we worked with Greg Skomal and Mauricio Hoyas, who are super enthusiastic about everything sharks. And they were – we had to pick their brains to learn as much about what they knew in order to make the vehicle smart, and capable of keeping up and sort of play shark.

What we realized is that even the information that was out there, and really what we needed, was that the information we could get from REMUS SharkCam. We didn't even have it yet. So we had to slowly take baby steps to put this system in the water.

Roger: I had no idea when we started the project, whether the sharks stayed in the same area and sort of did circles or curly-ques or what. But what we discovered is they're actually cruising along the coast and really transiting,and that made it a lot easier, because they're maneuvering back and forth, but on the whole looking at a longer period of time, they're moving in a general direction.  And from you can estimate their speed and overall direction.

Roger: We did see different sorts of behaviors with the shark, with the respect to the vehicle. Sometimes they just bumped it, other times they munched it.

Amy: What we learned from Greg and Mauricio was that predation attacks when the shark will come from below and behind and try to disable the prepulsion, much like you see in seals, and that's what happened many times, but then there's also more passive attacks where the shark would come, sometimes from underneath or forward, or right at the vehicle, and just knock it and give it a headbutt, and that was perceived to be more of a territorial attack.

Roger: Oneof the interesting things about the attacks though, was watch it bite down on the vehicle. And we're pretty close to the crash depth of the vehicle at 90 meters, and it didn't crush the vehicle. And they would do this for ten or fifteen seconds, and then they'd swim off, you'd see them off watching, waiting for the vehicle to die. It was like they were waiting for the vehicle or whatever they thought they'd attacked, to bleed out. It was very clear this was a strategy they were following.

Amy:  So it was this really intense, nail-biting experience whether we were going to lose this vehicle or if it would a leak or it was going to get, it kept on getting munched on and ripped apart by these sharks, we didn't know how aggressive they'd continually be.

You have this new access that you never knew you could have, and the scientists are really excited, and not only are they excited, but the engineers are also equally as excited. It's that perfect marriage of an oceanographic project to get everybody on board to see something happen for the first time, and figure out where it can go from here.

In was in real time, documenting these new behaviors, seeing them, and know you're part of this groundbreaking application and technique of acquiring exciting footage so we can learn more about these beautiful apex predators.

Question 4: Has your experience using REMUS SharkCam changed your perception of sharks?

Roger: When I started the project, a lot of my feelings about sharks were colored by movies like Jaws, and Shark Week, and really I came to have a very different feeling towards them.

Amy: I just realized how fine-tuned, how their senses are so advanced, far advanced than what we can understand, and what they're doing is trying to survive.

Roger: You think of them as these ferocious predators, and certainly they are, and certainly they're very dangerous animals, but I think people have gotten the wrong impression.

Amy: What I'd like to see happen with the technology is to learn more into the inside lives of these beautiful creatures, whether they're great whites, or anything else this technology might apply to. The whole point is to learn more about them so we can protect them.

Roger: 100 million sharks are killed every year, very few people are killed by sharks. I hope of what comes out of this is a greater respect for these animals.

Amy: In the end, our primary goal is to help conserve sharks, help educate people about sharks and how important they are in the marine ecosystem.

For more information, contact the WHOI Media Office at media@whoi.edu.