The Inventor Of Roomba Has A New Robot That Sucks Up Invasive Fish

The Guardian robot collects lionfish so we can eat them—before they destroy ecosystems by chowing down on all the local fish.

When Colin Angle visited Bermuda in 2015 and first saw lionfish–an invasive species that has decimated local fish populations with its ravenous appetite–it wasn’t the first time he had heard about the problem. But it was the first time that Angle, who is the cofounder of iRobot, the tech company that makes the Roomba, started to think that he might be able to help solve it.


“The dive operator challenged me when I came out of the water and said, ‘You build robots–why don’t you create a robot that can go and help control the lionfish problem?'” Angle tells Fast Company. “Rather than dismiss that as ‘Yeah, whatever,’ it was, ‘Okay, would that work?'”

“There’s a demand for lionfish, and if you can catch them, people will pay you.” [Photo: courtesy Robots for the Environment]
Within a year, he and his wife, Erika Ebbel Angle, founder of a biotech startup and the director of the education nonprofit Science from Scientists, had launched a nonprofit called Robots in Service of the Environment–and a team started to develop a robot called Guardian, designed to drive up to a lionfish, stun it, and suck it inside the robot, which can collect multiple fish before returning to the surface. The fish can then be sold to restaurants.

It’s one among many attempts to control an out-of-control population. The fish, which is originally from the Indo-Pacific, has no native predators in the Atlantic, where it started to appear–possibly released by people who had bought the fish as pets–in the 1980s. The fish eats other fish voraciously; within half an hour, it can consume 20 other fish. In five weeks, one lionfish can reduce the population of fish on a reef by 80%. A single lionfish can live up to three decades, and spawn as many as 2 million eggs in a year. In some locations, there may be as many as 1,000 lionfish per acre.

In Florida, Whole Foods stores now carry the fish–which tastes a little like a cross between grouper and mahi mahi–to try to encourage more fishing, and fishing tournaments have tried to encourage it more. But lionfish don’t respond to bait, and can’t easily be caught with nets. Spearfishing works, but only in shallower waters, and the fish are often found in caves deeper than humans can dive.

“There’s a demand for lionfish, and if you can catch them, people will pay you,” says Angle. “Our challenge is that down in deeper water or in areas that aren’t commonly accessible to people, we can’t [catch them]. And that’s what happens with areas where they are proliferating at the greatest rates. So a robot that could go down two, three, 400 feet would be an incredibly powerful tool.”

“It’s actually less about the technology and more about the business model.” [Photo: courtesy Robots for the Environment]
One challenge was developing a robot that could catch the fish affordably. “You’re not going to go change a population by capturing one lionfish with a half-a-million-dollar robot,” he says. He realized that he could apply manufacturing expertise from iRobot–making low-cost robots like Roomba–to the new robot. The nonprofit is aiming for a cost of $1,000 or less. “The fully capitalized operating cost of running that type of robot is so low that you could make money at this.”


After considering a range of designs, the team moved forward with a prototype that drives directly up to the fish (the lionfish, as a top predator, has no fear and won’t swim away; other fish, with a normal amount of fear, aren’t captured by the system). Two arms stun the fish, and then the robot sucks the fish inside a tube. The current prototype can catch up to 10 fish before returning to the surface to deliver its catch.

“A robot that could go down two, three, 400 feet would be an incredibly powerful tool.” [Photo: courtesy Robots for the Environment]
The robot is controlled remotely by an operator at the surface who watches the action via a camera. Unlike another robot designed to kill starfish (also an invasive species), the Guardian can’t hunt down fish by itself. But that makes sense economically, Angle says.

“It’s actually less about the technology and more about the business model,” he says. For the project to work and control the species, it had to work at scale, and that means collecting the fish for sale; the starfish robot is designed to simply kill starfish. If someone is on site to collect the fish anyway, they can control the robot, making the design simpler and less expensive.

“Part of the funding model and scalability of the project could be tied to the gamification of the operation of the robot.” [Photo: courtesy Robots for the Environment]
After a successful Kickstarter campaign, the nonprofit is currently working on the next generation of the design, with simpler operation. This version will work with a laptop, a single tether, and some batteries, and will be easier to ship and possible to operate with less training. As a prototype, it’s still expensive, but the team plans to use it to demonstrate that it’s possible to reliably capture a certain number of fish each hour. Next, the engineers will work on reducing cost, focusing on making the most expensive part–eight thrusters that drive the robot through the water–less expensive.

Ultimately, when the robots hit their target cost, the nonprofit hopes to sell them for sport fishing along with commercial fishing. People may even operate the robots remotely. “Part of the funding model and scalability of the project could be tied to the gamification of the operation of the robot–you could pay to operate a lionfish from your iPad, sitting in your office wanting to do good,” Angle says.

The lionfish robot is the first of several robots the nonprofit hopes to develop to tackle environmental problems; others might include robots that fight poaching, or collect ocean plastic. “We believe that low-cost robotics is a technology that enables a variety of innovative solutions that have not heretofore been feasible or viable,” he says. “The lionfish is step one.”


About the author

Adele Peters is a staff writer at Fast Company who focuses on solutions to some of the world's largest problems, from climate change to homelessness. Previously, she worked with GOOD, BioLite, and the Sustainable Products and Solutions program at UC Berkeley.