A new and better way to detect floods just got off the ground. The innovative artificial intelligence model, called “Worldfloods”, can relay flood information from space to disaster response teams below.

Deployed on hardware as part of the “Wild Ride” mission launched by SpaceX’s Falcon 9 on June 30th, 2021, Worldfloods represents an important step in bringing AI operations to space. In fact, this is the first time that a machine learning model for detecting floods will be used in space.

The research comes from an international team of eight people, including Luxembourgish researchers Dietmar Backes (University of Luxembourg) and Guy Schumann (RSS-Hydro), as part of the Frontier Development Lab Europe and was published as a Nature Scientific report in 2021.

The Luxembourg Space Agency is proud to join the University of Oxford, European Space Agency’s Φ-lab, Trillium Technologies, Google Cloud, and Intel in sponsoring the project.

The Worldfloods model is significant for a number of reasons: Floods are one of the most destructive natural disasters, and situational awareness on the ground can save lives. While developed countries have access to high-quality images that detect floods, many of the worst flood-affected regions are in low-income countries which lack such imaging resources. Worldfloods can help change that by bringing the cost of the technology down and making it more accessible.

In addition, the onboard processing feature of the Worldfloods model can help it to relay flood information in real-time, greatly improving emergency response time.

A recent article published in Nature Scientific Report details how the model works. Read it here.

Article by Luxembourg Space Agency