Skip to main content

Edge Models Provide Researchers with New Insight into Marine Ecosystems

Wednesday, June 12, 2024

Analysis of large datasets aboard a sailing drone off the coast of Gotland is giving marine ecologists real-time insights into the happenings of the Baltic Sea.

– Thanks to edge models, we don’t have to wait for the drone to come ashore before we can access the data, says Jonas Hentati Sundberg, researcher in marine ecosystems at the Swedish University of Agricultural Sciences.

The sonar is a familiar tool for Jonas Hentati Sundberg. It provides important information about fish stocks, waves, and the seabed. But the tool also has its weaknesses: it collects an extensive amount of data, making it time-consuming to analyze.

With the help of decentralized processing and analysis right at the data collection point, this limitation has been overcome. For Jonas Hentati Sundberg and his colleagues, this means new opportunities to research marine ecosystems. In the long run, it could also lay the foundation for a new way of managing marine ecosystems:

– Fishing quotas are currently based on forecasts and set on an annual basis. With real-time data, decisions could be made in real-time. Legal and management frameworks need to be adapted to handle this type of information and swift decision-making, says Jonas Hentati Sundberg.

Sofia, Jonas and Joakim posing for the camera on a boat, blue sky

From left: Sofia Nguyen, Jonas Hentati Sundberg and Joakim Eriksson

Complete Post-Analysis Complemented by Real-Time Updates on the Most Important Data

For several years, Jonas Hentati Sundberg has used a Norwegian-built sailing drone to conduct his research. The drone autonomously navigates the waters between Öland and Gotland, collecting sonar data, among other things, which is later analyzed.

But when the drone was deployed this spring, it was equipped with a new combination of hardware and software: a Raspberry Pi and self-developed models for processing and analysis.

– With edge models, we filter the collected data in real-time. The most interesting data is sent directly to shore every ten minutes, via satellite link. The rest is analyzed later when the drone comes ashore and is emptied of data, says Joakim Eriksson, AI Developer at AI Sweden.

Sofia Nguyen and Joakim Eriksson working on computers inside a ship

Sofia Nguyen and Joakim Eriksson, AI Sweden

This isn’t the first time that Joakim Eriksson and Jonas Hentati Sundberg have worked together. Two and a half years ago, they developed models that automatically count nesting guillemots on Stora Karlsö. Last winter, a chance conversation revealed the potential for a new collaboration – on drones. In the drone project, solutions similar to those for counting guillemots will be used to eliminate time-consuming manual work.

– But we aren’t just limiting manual work, we’re also reducing the risk for human error, which makes our research results more robust, says Jonas Hentati Sundberg.

The new technology will allow researchers to study the ecosystem in a new way: in real-time for the most critical events, and historically for a more complete understanding.

– What we want to know in real-time could be data indicating an ecosystem collapse or unexpected new patterns that we haven't seen before. This would allow us to react immediately, such as by activating additional sensors to gather more information, says Jonas Hentati Sundberg.

The comprehensive, complete analysis funded by the Arctic Research Foundation and Stockholm Resilience Centre will be supported by artificial intelligence.

The AI models currently used by Joakim Eriksson and his colleague Sofia Nguyen can, for instance, differentiate between the seabed, waves, and fish. "This allows us to already calculate the fish mass under the drone.‘’

"The next step is to develop algorithms that can identify individual fish and perhaps even determine their species," says Sofia Nguyen, AI developer at AI Sweden.

Sofia Nguyen and Joakim Eriksson, AI Sweden

Sofia Nguyen and Joakim Eriksson, AI Sweden

Sailing drone at sea

Sailing drone at sea

FACTS: Decentralized AI as a Tool for Real-Time Decisions

The parallel development in sensor technology, federated learning, and decentralized AI opens up new analysis possibilities in many sectors and industries. Thanks to increasingly energy-efficient, precise, and otherwise improved sensors, more data can be collected. AI's role is to reduce the amount of information that needs to be transmitted wirelessly. Through models that can run on less powerful hardware, it becomes possible to filter data directly on the sensor, only sending valuable information onward.

AI Sweden has led many projects in decentralized AI and federated learning, including in the automotive industry, finance sector, healthcare, and space.

"When you look at all these projects together, it becomes clear how they create value for our partners, as knowledge builds on projects that follow one another and also involve different industries and sectors. This drone project is one example, another is how Zenseact used knowledge from the FedBird project that Jonas and Joakim were also involved in," says Johanna Bergman, Director Strategic Development at AI Sweden.

The Goal: Robust, Generalizable, and Possibly Commercialized

Joakim Eriksson and co-worker Sofia Nguyen handle the AI components in the drone project. They note that the current drone has its limitations:
"For instance, we need to conserve electricity. In the long run, we would like to develop our own drone to optimize the hardware conditions necessary to run more advanced AI models onboard," he says.

Another way forward is to install AI models on SLU's research vessel Svea, where the sonar systems are more advanced, and the connectivity and electricity supply are better.
"Perhaps we could even install the solution on commercial ships and use them as a sensor network," says Jonas Hentati Sundberg.

The drone project has attracted international interest. A trip to Coats Island in Hudson Bay, Canada, is booked in the weeks following Midsummer, where the technology will be tested. A similar collaboration is also in the works with a research group in Australia. The goal is to take the next step in development together with them.

"We want to make our solution more robust and generalize it so it works on more hardware platforms and for more ecosystems. Eventually, we might even commercialize it," says Jonas Hentati Sundberg.

From the Young Talent Program to the Baltic Sea

In the spring of 2021, Joakim Eriksson graduated from the ABB high school in Västerås, where he studied a technical program with a focus on artificial intelligence. That fall, he was among the new students in the first cohort of AI Sweden's Young Talents program.

Sofia Nguyen participated in the Young Talent program in the fall of 2023 after graduating from Platengymnasiet and now works as a developer at AI Sweden.

"In the Young Talent program, we give young people with potential the opportunity to learn about AI and machine learning and work on real projects. Time and again, we have seen that these talents can truly contribute and create significant value. Joakim and Sofia are excellent examples of this," says Sofia Hedén, Head of Talent Programs.

"The program has been incredibly educational, and I appreciate AI Sweden's trust in my abilities and the opportunity to contribute, despite my young age," says Sofia Nguyen.

A picture of Joakim Eriksson

Joakim Eriksson, AI Developer at AI Sweden

Sofia Nguyen

Sofia Nguyen, AI Developer at AI Sweden

Sofia Hedén

Sofia Hedén, Head of Talent Programs at AI Sweden

You might also be interested in:

Stockholm location
The increase in AI applications is leading to a fast rise in demand for AI experts. The business...
Edge Learning Lab
Edge Learning Lab is a unique environment that enables professionals and students from our partners...
More than you think. Inside AI Sweden’s Data Factory, Zenseact has used a dataset from the Swedish...
Baltic Seabird Dataset
The dataset consists of 2000 hours of video footage of guillemots on a ledge on Stora Karlsö, Sweden...
Urban traffic scene with a pedestrian crossing, viewed from the perspective of a driver
This dataset is provided by Zenseact with the intent of bringing real-world challenges to academia...