The AI Revolution Enters the Water: How Radmantis Is Transforming Aquaculture
Robert Huber, Co-founder of Radmantis, and Senator Elizabeth Warren discussing Radmantis' technology.
Radmantis was founded by two academics looking to use their book smarts to create solutions to real problems. The result? A robotic fish management system that uses AI and machine learning to interact with fish in a way that is not harmful to them.
Looking to expand from Midwestern fresh water to salt water, Radmantis participated in the third cohort of the BlueSwell Incubator. During the program, they took advantage of SeaAhead’s partnership with the New England Aquarium to facilitate this alteration to their technology. Since graduating from the program, Radmantis has forged a strategic partnership with Local Coho, sustainable seafood farmers, which has allowed them to improve their product offerings.
We sat down with Radmantis’s co-founder Moira van Staaden to discuss the company’s journey and its future plans.
SeaAhead: Tell us a little bit about yourself, your background, and how you ended up at Radmantis.
van Staaden: My background is traditional academic. I've been in academia for 20 years and doing everything that entails – teaching, research, and I've been pretty successful. I’ve been really well funded in terms of research, mostly from the National Science Foundation in behavioral neuroscience and STEM education. Just three weeks ago, I got back from South Africa, where I was a Carnegie fellow in science education there in Cape Town.
So, I participated in all the traditional academic activities, but just before COVID hit, I decided to try something different. I've always been interested in social communication systems and evolutionary systems and decided to go and get a business degree. I got a master's in organizational development which prepared me for entrepreneurial challenges.
SeaAhead: Very cool. How did Radmantis come to be?
van Staaden: The two co-founders are myself and a longtime colleague, Robert Huber. We’ve both been in academia for 20 some years and working on various species and behavioral systems, but always at the interface between life sciences and computing. We decided to try and find ways to apply all this expertise to real problems in the real world, not just topics of academic interest. Food insecurity, for example. If you look at population growth in east Africa, for example, you've got this massive population growth coupled with climate change, so you've got a major problem on your hands. So, we wanted to contribute in a much more direct way to mitigating those kinds of problems.
SeaAhead: Can you give us an overview of the work Radmantis does?
van Staaden: We're developing an uncrewed fish management system. We’ve combined cutting-edge AI machine learning technologies with robotics so that we can directly interface or interact with fish in a way that's not harmful to them. When you're dealing with animals, especially fish, interacting with humans is really stressful. So, anything that we can do to reduce the stress from that direct intervention is a good thing.
“When you’re dealing with animals, especially fish, interacting with humans is really stressful. So, anything that we can do to reduce the stress from that direct intervention is a good thing. ”
SeaAhead: You are involved with a number of different industries and sectors. Do you have one major product offering applicable to all of them, orare they tailored to each sector?
van Staaden: The underlying technology is the same, but it gets tailored to different kinds of applications. The main sector right now is aquaculture - generally everything's going toward robotic fish farming. So, what are the common management tasks involved in that? Feeding, harvesting and movement, but all done robotically.
The second industry would be conservation. State level conservation, or even private industry, where they are doing things like population assessment, or trying to control invasive species, which is a different technical challenge. We also began talking with folks about underwater infrared imaging. Turns out there’s a lot of interest from dual use cases in maybe applying some of that imaging technology. That's in the early stages.
SeaAhead: Your website says your product is “a technological immune system for habitats under attack.” I would love to hear you elaborate on that.
van Staaden: A human’s immune system is looking for things that are entering the system that don't belong there. It’s on guard and wants to remove these dangerous actors from the system and do that in an adaptive way. Your immune system is constantly looking at the new challenge, dealing with it, and moving on.
In the same way, we know habitats are under attack by invasive species. We have Asian carp in the US and pythons in the Everglades. We've got lionfish on the Eastern Seaboard, and Nile Perch in the African Great Lakes.
These invasives have an immediate impact on the major species, but there are also these knock-on effects these species have on the resilience and integrity of the ecosystem as a whole. So, the Radmantis technology serves as an immune system. It's always looking out for invasives. It can recognize these individuals and selectively remove them. And it's adaptive because you can retrain it to recognize different invaders.
SeaAhead: Very fascinating. What are some of the biggest challenges Radmantis has faced since its founding?
van Staaden: Anytime you are coming up with a new technology, you have to integrate it with pre-existing systems. So, we have a new technology and we're trying to integrate it into these aquaculture facilities that are based on 50-year-old technology or 50-year-old design principles. Obviously, it would be a lot easier if we could redesign these from scratch!
The other thing is, we have a system that I think can solve big problems, but you have to start somewhere. So, we're trying to demonstrate this on a small local scale, that's a critical first step to solving a much bigger problem. But sometimes people have difficulty imagining how the technology is going to impact things at scale. But it's just making that sort of mindset adjustment.
And then, really the biggest issue is proving our solution is better than preexisting solutions to problems, even though current solutions are very much sub optimal.
Typically, when people want to remove invasive species, they send vast numbers of people out into the field to manually remove animals or they use chemicals. If you're trying to get rid of sea lamprey, for instance, you throw chemicals into hundreds of rivers, to try and get rid of them. In many people's minds that is solution enough and just having a better solution is not necessary. So, how do you persuade them? How do you get them to recognize that there may be a solution better for both animals and the environment?
Co-founder and Managing Director of Radmantis, Moira van Staaden, releasing Sturgeon fingerlings into the Maumee River.
SeaAhead: What made you choose to apply to the BlueSwell Incubator program?
van Staaden: We are based in the Midwest right on Lake Erie. Given that, all our applications have really been focused on freshwater issues. But this only covers a small part of the aquaculture market.
BlueSwell is focused on coastal systems and they're also working very closely with the New England Aquarium, which has a sustainable aquaculture sub-initiative. We thought the program would be a nice way to transition our technology from freshwater to marine applications.
SeaAhead: What other ways has the program helped you shape and expand your business?
van Staaden: BlueSwell also gave us a much better understanding of the market as a whole and the value our particular solution can provide to commercial efforts. It also really helped us with answering the questions, how do we go about commercializing this? How do you value this from the perspective of industry?
The program also provided us with a lot of useful industry connections that would have been difficult to make given that we are based in the Midwest. It's a nice, small cohort, and so the opportunity to work closely with those people and to learn from them is hugely valuable because they are also doing tremendously innovative things.
SeaAhead: Very glad to hear that! Where do you see Radmantis heading in the near future?
van Staaden: What we would hope is that the vast majority of aquaculture facilities would begin to be retrofitted with this kind of technology. That's a very optimistic goal!
Longer term, we hope that instead of having to incorporate our technologies into existing systems, people will start to redesign future aquaculture facilities with these robotic solutions in mind. I think that would be ideal.