Flood detection from space, supported by AI


The broader the vision that one has about the extent of a flood, the better panorama one can have of that scenario to design any emergency measure.

Researchers from the Image Processing Laboratory (IPL) of the University of Valencia, in collaboration with the University of Oxford and the Phi-Lab of the European Space Agency (ESA), developed an AI-based flood detection system that works from space and is already in orbit.

WorldFloods, an in-orbit system that monitors floods

According report the same University of Valencia through the portal of its Science Park,

On June 30, this system was projected into space from Cape Canaveral, aboard Space X’s Falcon 9 rocket. The operation was carried out by the company D-Orbit, which has previously worked with companies to put microsatellites into orbit. .

WorldFloods, the presented system, is currently housed in Nebula, a satellite of just 1 kg in mass. The AI ​​algorithms present in this technology can be executed on this support thanks to the presence of radiation-tolerant chips.

This model was developed in the FDL-Europe research incubator, with funding funds provided mainly by the European Space Agency.

The main purpose of WorldFloods implementation is to reduce the time it takes to trigger an emergency response to natural disasters of this category, providing a system of images that are updated almost in real time. In addition, thanks to the support of artificial intelligence, the use of resources is optimized, reducing the cost of implementing this technology, with respect to what it would involve executing a project like this using other means.

“On-board processing offers a solution to decrease the amount of data to be transmitted by reducing large sensor images to smaller data products.”said Gonzalo Mateo, a member of the IPL.

WorldFloods relies on the system CubeSat, which unlike other satellite models, provides a response in a matter of hours and not days.

Although the investigative report after this project, published in the journal Scientific Reports of Nature, it recognizes that the transmission of data from these platforms to terrestrial receivers is conditioned according to the power and bandwidth restrictions of the CubeSats, however it presents a useful advance that can contribute to the management of emergencies derived from major floods.