Fish Out of Water: Using Drones to Help African Nations Spot Illegal Fishing

AI and Jetson platform let government orgs monitor 10,000 square kilometers of ocean a day.
by Isha Salian

The undulating western coast of Africa is one of the most diverse fishing zones in the world, with more than 1,000 identified fish species. It’s also one of the most at risk.

One of the World Wildlife Fund’s “Critical Regions of the World,” the West African marine region has been overfished for decades, with rampant illegal fishing threatening marine ecosystems and the millions who depend on them.

Estimated catches in the region are 40 percent higher than reported figures, costing the governments of West African countries more than $2.3 billion a year in fishing licenses, taxes and fees.

Reversing the trend with better enforcement of fishing laws is no easy task. Monitoring vast swaths of ocean with a fleet of manned aircraft and boats is impractical for countries with limited economic resources.

But a five-person Moroccan startup is taking up the challenge. Atlan Space is helping African nations crack down on environmental crimes on land and sea with AI-powered drones. A single drone equipped with its technology can allow government organizations to scan thousands of square miles of ocean each day.

“Africa is huge,” founder Badr Idrissi said. “We are developing AI because it’s a need for us. It’s the only way we see that can help to monitor very large geographical areas.”

Two Friends Fishing for Sea Vessel Data

In 2016, Idrissi and his friend Younes Moumen began reflecting on the environmental legacy their generation will leave behind, inspired by the global climate talks going on in Morocco. Idrissi was an account executive at Microsoft at the time, and Moumen a CTO at a local startup.

The pair decided to found Atlan Space, a company harnessing autonomous drones to combat environmental depredation in Africa and worldwide, starting with the oceans. A member of our Inception virtual accelerator program, the company uses NVIDIA Tesla GPUs to train its neural networks and the NVIDIA Jetson embedded platform for inference.

To monitor illegal fishing, governments typically rely on coast guard boats or light aircraft. Both require skilled operators — and can only cover a limited area. Another option, satellites, is expensive and provides a less detailed view.

A single drone can allow them to monitor 10,000 square kilometers a day, while navigating at a closer range, under any clouds. Atlan Space deploys its AI on a Jetson TX2 board, which connects to the drone’s autopilot and communication system.

The autonomous drone flies on a path determined by the neural network as it scans the seas. Once the deep learning model spots a boat, it analyzes the image to identify whether it’s a fishing vessel.

Next, the neural network analyzes the boat’s name, flag and type of radio signals to determine whether it is legally permitted to operate in the region. All this inference work happens on Jetson, processing even if the drone loses its satellite connection during flight.

When the neural network identifies a probable unauthorized boat, it alerts authorities via a satellite message.

Back on land, the raw data collected by the drones is fed into software running on NVIDIA Tesla GPUs through the Microsoft Azure cloud platform.The deep neural networks learn from this additional data to improve and optimize for future missions.

Atlan Space partners with local nonprofit organizations and government institutions to deploy its solution in the field. The platform can work on many different kinds of drones with open APIs — so customers can use ones they already own.

A Drone’s-Eye View

Its technology will take to the skies this month in the Seychelles, an archipelago of 115 islands in the Indian Ocean, northwest of Madagascar. Fishing is a major segment of the economy — but illegal fishing costs the government millions.

Unlawful fishers may be poachers trawling for sea cucumbers or shark fins, or unauthorized boats from other nations. Or they may be local boats that break rules by using illegal nets or other unsustainable fishing practices.

In addition to the drain on the economy, illegal fishing threatens local food security and endangers fishers, who venture further out to sea to catch enough fish. “People are really putting their lives in danger just to get a decent catch,” says Idrissi.

Atlan Space was a finalist at the recent GTC Europe Inception Awards, and in June won the National Geographic Society’s $150,000 Marine Protection Prize for its project in the Seychelles.

Beyond curbing illegal fishing, Atlan Space is exploring areas like precision farming and civil engineering. AI is an essential tool for any cost-effective project to monitor such large geographic regions, Idrissi says. “I don’t see it as a trend, I see it as a necessity.”

* Main image shows a fishing vessel suspected of illegal fishing activity in Sierra Leone, detected by U.S. Coast Guard and Sierra Leonean officials in 2009. Photo by U.S. Coast Guard.