UPDATED 10:00 EDT / JANUARY 05 2020

IOT

Autonomy at sea: Ocean drones net hard-to-get data for weather, fishing and more

The Earth’s oceans cover 70% of its surface. They supply half the oxygen we breath. They influence rainfall all over the world. They can send catastrophic tsunamis on shore in coastal communities. They’re home to the fish that make up 20% of the protein we consume. And those are just some things we know about them.

Man has long sought to know what wonders lay hidden in the deep sea. Whatever odd or marvelous things may be down there — treasures, sunken vessels, bizarre, undiscovered animal life — light isn’t among them. The pitch darkness and pressure of the ocean’s depths put limits on exploration. Now, technologists are using sonar data and artificial intelligence to see into the world’s oceans.

Since it impacts us land dwellers in so many ways, we can benefit from increasing our knowledge about it, according to Sebastien de Halleux (pictured), chief operating officer of Saildrone Inc. It’s a very hostile, dangerous environment, which is why we know so little about it, he added. The typical ships and buoys we’ve deployed to study it have left much a mystery. The data they collect from these bodies covering 70% of the planet is tiny compared to the amount of on-land data we have.

Ideally, we would be able to take photographs of the deep sea to make out its topography, and lifeforms. Unfortunately, the light from a camera’s flash can’t travel through miles of ocean water. But it is possible to pick up sounds from the deep with sonar sensors. Saildrone’s fleet of about 1,000 autonomous surface vehicles uses these sensitive, sound-detecting devices.

These bright orange sailboats are typically 23 feet long with a sail more than 16 feet tall. They harness wind power for propulsion and solar power for onboard electronics. The sonar data they collect, processed with machine learning, can produce a “statistical biomass distribution” or a 3D rendering of the seabed, according to De Halleux.

“We use sound instead of light, but with the same principle … which is that we send those pulses of sound down and [listen to the echo] from the seabed or from fish or critters in the water column,” De Halleux said. “We paint the ocean with sound.” 

De Halleux spoke with Dave Vellante and John Walls, co-hosts of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, during Amazon Web Services Inc.’s re:Invent conference in Las Vegas. They discussed how oceangoing drones are advancing the collection and application of sea data. (* Disclosure below.) 

This week, theCUBE spotlights Saildrone in its Startup of the Week feature.

Investing in impact

Founded in 2014, Saildrone has been a revenue-generating company since day one. It got started with funds from venture capitalists interested in supporting projects with long-term impact. Increasingly common “impact funds” enable financially viable startups that address social or environmental matters — like sustainable agriculture, public health, microfinance and the like.

More than 1,300 investors manage $502 billion in impact-investing assets, globally, according to Global Impact Investing Network. Some argue that investing for good and for profit in one shot is less effective than investing for pure return and then donating money. However, research from McKinsey & Co. on impact investments in India demonstrates that they have been able to reach return goals.

Saildrone’s customers include governmental organizations like NASA and NOAA, as well as research universities. It’s involved in Seabed 2030, a UN-backed initiative to map the ocean floor by 2030. Its drones have no fuel, no engine, and no carbon emission, making them 100% environmentally friendly, De Halleux explained. And they offer a relatively easy, inexpensive means to capture ocean data.

“No government has ever come and told us, ‘We have enough ships or enough data,'” De Halleux said. “If you’ve got a coastline, you’ve got a data problem.”

Autonomy at sea, analytics on land

The autonomous vehicles don’t require any crew members to operate them. They leave the dock on their own, sail around the world for up to a year, then return to the same dock. A drone harvests all the energy it needs from the environment: wind for propulsion and solar power for electricity, which powers the onboard computers, sensors and the satellite link that tells it when to return to shore.

Programming the boats to navigate the open sea by themselves can prove challenging. To respond to objects and events at sea, their algorithms must train on the lots of data. So the drones take millions of pictures of the ocean environment. Then Saildrone’s technologists train use these or other data sets to train algorithms to recognize a boat on the horizon, a bird, a seal or other objects.

“In some hard cases, when you have a whale under the Saildrone or a seal lying on it, we have a lot of fun pushing it on our blog and asking the experts to really classify it,” De Halleaux said. “You know, what are we looking at?”

For the majority of Saildrone’s boats, analysis of captured data takes place onshore. In fact, the drones themselves have little autonomy. The actual data analytics and inferencing — powered by the AWS cloud — takes place onshore within Saildrone. Once a drone sends data on its environment back to shore, engineers crunch it and optimize the route. They then send instructions via satellite back to the boats. The data from a drone’s 25 sensors may be processed into products such as weather forecasts. The Saildrone Forecast application renders a picture of the Earth and predictions about weather. 

The company just announced a 72-foot ship called the Surveyor that has onboard compute. All sonar data is processed on the ship with machine-learning and artificial-intelligence technology and sends the finished product back to shore. “Because no matter how fast satellite connectivity’s evolving, it’s always a small pipe so you cannot send all the raw data for processing on shore,” De Halleux explained. 

As advanced as this technology sounds, it is actually more affordable than previously available systems for gathering ocean data, according to De Halleux. In the past, some less developed nations could not afford the tools to accurately forecast weather or establish fishing quarters.

Now they can. And this is part of delivering the impact — it’s leveraging this amazing infrastructure and putting it in the hands, with a simple product, of someone, whether they live on the islands of Tuvalu or in Chicago,” De Halleux said. 

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of AWS re:Invent. (* Disclosure: AWS sponsored this segment of theCUBE. Neither AWS nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE

A message from John Furrier, co-founder of SiliconANGLE:

Your vote of support is important to us and it helps us keep the content FREE.

One click below supports our mission to provide free, deep, and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU