Aquaai's Robotic Fish Are Harnessing Data to Save Our Seas
Fish farmers constantly face the challenges of unused feeds and fecal waste (among other things). Meanwhile, the use of antibiotics and pesticides that aquaculture sometime uses can have adverse effects on the marine environment. But a startup called Aquaai is hoping to solve this pollution problem using a new kind of robotic fish. The company'’s invention uses predictive analytics, AI, VR, computer vision and biomimicry to surface precise data that can allow fish farmers to make smarter decisions while saving time, money, and the environment.
SeaAhead recently caught up with Aquaai co-founder and CEO Liane Thompson to discuss how the company’s robotic fish are helping fish farmers and others in bluetech to make more informed, and more environmentally sound decisions.
Tell me a little bit about your background and how you got involved with Aquaai
I come from a media background. I was previously [an executive producer] with the New York Times, covering conflicts all over the world. I was running a very large team of about 300, putting together documentaries, reality, delivering programming — over about 100 hours of TV a year for all major broadcasters. We were, at that time, the largest production company in New York City. I put people like Anthony Bourdain on television, three prime time Emmy nominations for different series and programs I’ve created or ran as an executive producer. I went out to Israel during the peak of the Intifada, just because as a former conflict reporter, right after 9/11, I was a little bit tired of sitting in Manhattan and I went to the Middle East to open up the quasi-TV bureau. We did some programming for Discovery and on Suddam Hussein… I stayed in the Middle East, ended up quitting the Times and covering different areas, I was in Haiti for an earthquake. I had always been with young companies (young media startups). I always sort of had my own [media] startups [in Budapest, and in Israel].
At that time I was doing a story on the Q (from James Bond) when I met Simeon Pieterkosky, who is a robotocist/anamotronix and originally from Cape Town, South Africa who was also in Israel for a period of time. He was obsessing over this robot fish because his daughter had come home from school and said “Dad you have the save the seas, stop building land-based robots.” He had already built about 10 land-based robots. He was one of those guys that could build anything and had been doing that when I met him. When I met him he was looking at different systems. He realized the systems in place were unattainable because of cost and needed to incorporate more biomimicry… so he thought: “What systems could work in the ocean to be available to many people to make a bigger impact and change?” He had essentially promised his daughter to dedicate his time to making robots that could go into the water. That’s when I entered the picture, we became a couple and came to my home in the U.S. Rather than going into media, I said let’s do a startup with your invention. We chose the robotic fish and formed a company around it and dedicated ourselves to “saving the seas”. I had no idea in my life that I would ever have a robotics company.
Tell me about the underwater vehicles you’re using-- are they AUVs or ROVs?
We want to ultimately have a robotic fish swimming all over the ocean, swimming and gathering data and providing that data — whether its to conservationists who can’t afford to send a dive team out, to space agencies who want to monitor temperature shifts affordably, globally… even now where we’re swimming in the arctic circle in Norway where we’re working with sustainable fish farmers.
Our current partner and customer is [a company] that provides salmon to Whole Foods. The units in the water right now are autonomous, in Norway, because the farmer wanted data 24/7, without intervals — and because it’s also Northern Norway where their farms are very sophisticated (they already had electricity in their cages). So we just tap into the existing electricity with a tether, and we’re sending nonstop information. Those units are tethered (ROVs). The next units we’re delivering in a couple of months are battery powered (AUVs, batteries last about 8-10 hours). We have both options.
They’re all programmed, and we use AI for navigation. The units learn based on what they’re swimming with, in this case the salmon. They’re very efficient. The whole idea of having the robotic fish [was about] trying to be unobtrusive. The existing habitat (other fish) adopt our robot and swim alongside it. We have been able to achieve efficiency by having it swim like a real fish. They’re also very maneuverable which is great as we start to look at use cases outside of aquaculture. The units are built to be scalable, if someone wants the battery to last longer we can scale it and use a larger battery.
The sensors are all plug and play. The user needs to be able to understand it, it needs to be simple, it needs to be affordable. [Simeon] thinks and designs these with a manufacturer’s brain. The fewer parts you have, the fewer issues you have with maintenance. We needed something that can handle rough waters, be affordable to enough people can use it.
We don’t sell the units, we’re B2B. So we lease the units, and then we provide the data through a subscription. We have 3 cameras on board. And we collect salinity, oxygen, pH, depth, temperature, gps, sonar. So depending on what use case, the units are meant to be scalable/plug and play to to collect that data. The units are 90% recyclable. Most of it we 3D print.
So the customers lease the units and subscribe to be able to get the data from you? Are you doing anything with the data other than providing it to your customers? What impact are your customers able to have with the data you provide?
Over time, the goal would be for us to provide predictive and preventative analytics. We want to be able to have different sensors and other add-ons to provide more information to the farmer. Right now we can see the fish swim and monitor behavior and water quality, so the farmers are able to decrease their feed (e.g. if the fish aren’t eating there is no purpose in continuing to feed them). That also helps then keep the ocean cleaner and not wasting food.
In Norway it’s also regulatory to provide different information at three and six meters every day, so we are able to go in and run a mission (e.g. in the morning the fish checks the bottom to see if there is a lot of waste, or it does a net inspection run to see if there’s a hole in a net to prevent any escapes and it can send an alert—it’s an early warning system as well). We can gather that information at three and six meters so the farmer can get those readings and send that information directly to the regulatory agency to which they need to provide that data. We help do that as well. We’re helping the farmer be more aware and active. [Fish] farming is a very busy job — we’re there to provide the information they need to make smart decisions so their stock is healthy. There are already cameras and sensors, but they are static. Our platform/data acquisition is superior to that, it’s much more dynamic.
What has been your biggest challenge jumping into the bluetech space?
I can answer that from two perspectives — one from Simeon, the mind behind the machine. Having built so many robots on land, working underwater is a whole different animal. It’s so challenging to work underwater.
From my perspective, the challenge is to make an affordable unit. We want to create an affordable system that can be used widely. We’re trying to hit that sweet spot between the surface and ~25 meters down where we believe a lot of work needs to be done (e.g. from runoff, coastal flooding, disaster scenarios, for coral reef restoration, for teams finding it more and more difficult to finance expeditions/research projects).
From a business perspective, neither of us come from the U.S. startup world — I don’t have a network of entrepreneurs and investors. It’s such a new space, and a space that most people don’t understand.
In addition to those in the aquaculture field, who would be considered a customer that would lease the unit?
Global reinsurers/insurance agencies. They view our systems as a risk mitigator and post-disaster technology. For example, they may send someone out to a flooding scenario in Houston, or somewhere in Florida, and sometimes sending out those divers to get proper readings is a very dangerous job. This is a job a robot could do, maneuvering around all the debris and gathering data to provide back to reinsurers.
Also ports. You have different needs in ports because shipping is huge globally. Ports are struggling, environmentally, with traffic, monitoring sea bank shifts, et cetera. They all have ROVs, but they usually sit on the floor because you still need somebody to drive it. With an AUV you can set GPS coordinates and program the mission and it goes and does it.
What’s next for Aquaai?
We are honored that, about a year ago in London, the European Space Agency featured us as a new data source for monitoring temperature changes for combating climate change. Once we move our operations to Europe, we want to engage more with the European Space Agency to provide that data to them.
We also have a mini lab in Norway, up near where our units are currently swimming; we will continue to grow that out and our customer base in Norway. We also want to grow the company, and enter other markets in Europe given that Europe has strongly embraced the idea of climate change and the need to focus on our oceans and ocean sustainability. They’ve been very welcoming (they being folks out of France, Ireland, Norway) to our technology. We see ourselves growing with those that are dedicated to having a positive impact on our oceans.
We are also very school-focused, and we’re talking with different institutions to boost their curriculums. We want to donate one of our robots so the students have that to learn off of, and educate the next generation of ocean warriors. We’re also always working on our technology. We’re always upgrading, seeing what can be added, what can be improved. We are always upgrading our units. As a young, early stage startup, we are always looking for partners who are like minded and on board with our mission.
Hannah Armstrong is SeaAhead’s associate editor. This interview has been edited for length and clarity.