Ocean Cleanup: Clean up the oceans with smart boats

Cleaning up the oceans is a huge undertaking, especially for a single nonprofit, but have Microsoft on your side is a nice bonus.

The Ocean Cleanup she's born in 2013 with the aim of cleaning up the large mass of plastic waste in the Pacific. Since then, the project has also embraced the goal of cleaning up the oceans by preventing new waste from entering the ocean. This means cleaning up the rivers that carry many of the pollutants. In 2018, The Ocean Cleanup has partnered with Microsoft to develop machine learning models to track and recognize plastic. Since then, The Ocean Cleanup's plastic "collectors" have been scanning the rivers and learning more and more to recognize the plastic to be cleaned. 

A fleet of interceptors to clean up the rivers

The fight against plastic pollution is active on two fronts: the one already present in the ocean and the one that moves in the ocean through the rivers. Plastic in the oceans tends to form masses such as the Great Pacific Garbage Patch, the immense island of waste in the center of the Pacific as large as the Iberian Peninsula.

These masses then break down into microplastics, which can be harmful to marine life and ultimately to people, penetrating everywhere. Hunting for this waste with ships and nets would be expensive, time-consuming and would take enormous quantities of fossil fuels. Instead, the project is working with what it calls a passive system for cleaning. A fleet of solar-powered Interceptors that autonomously collect plastic from rivers before it can reach the ocean, moving with the river currents.

The Ocean Cleanup: How an Interceptor Works

After the first interceptors already deployed in Jakarta and Malaysia, The Ocean Cleanup hopes to use interceptors on 1000 of the world's most polluting rivers within five years. Vietnam and the Dominican Republic are the next targets. Each interceptor has a capacity of 60 cubic meters (65 cubic yards) of waste and weighs nearly 50 tons. Despite its enormous size, the interceptor is scalable, can operate in most of the world's polluted rivers, and requires only minimal human contact with dangerous pollutants. It's a kind of catamaran, where the water continues to move with the current while concentrated plastic flows through it.

The current moves the debris onto a permeable conveyor belt, leaving only pollutants behind. Depending on the weather, current and other factors, a single interceptor can collect more than 5 tons (11.000 lbs) of debris in one day. Once full, the interceptor brings the waste to shore for sorting at a local facility.

This is where Microsoft's help comes in

In 2018, the Microsoft employee Drew Wilkinson contacted The Ocean Cleanup via email to explain how the tech giant could help the nonprofit at its annual hackathon. “Microsoft has immense computing resources that could really help you track your efforts at a fraction of the cost using AI”, he wrote in his outreach email.

Thus, during two corporate hackathons, Microsoft employees built a machine learning model to monitor plastic flowing through rivers with interceptors. Before Microsoft's help, the nonprofit had one person identify debris images on its own, an inefficient and tedious process. The nonprofit shared the images with Microsoft volunteers, who used the machine learning model to identify the plastic versus other debris like leaves or branches. In one summer they were able to label more than 30.000 photos.

Now they are working on a similar model for ocean photos as well. Ships in the ocean have a similar cleaning method and bring debris back to a control center for evaluation. Now, The Ocean Cleanup is working to recycle the collected plastic into new products to keep the nonprofit financially sustainable.