The Tech > Emerging tech > Hardware > Satellites - (11)

Satellites can be used for a range of public good tech solutions, from providing imagery to humanitarian responses to monitoring environmental degradation.

Showing 11 Results

Carbon Mapper

We leverage remote sensing technology to detect, pinpoint, and quantify methane and carbon dioxide (CO2) emissions at the scale of individual facilities.

Google MethaneSAT partnership

Just like how we use AI to detect sidewalks, street signs and road names in satellite imagery to display helpful information in Google Maps, we’ll also use AI to identify oil and gas infrastructure, like oil storage containers, in our imagery. Then, we’ll combine it with EDF’s information about oil and gas infrastructure to locate where the emissions are coming from.



United States of America (the)

"an end-to-end open-source framework for creating georeferenced pedestrian networks from aerial imagery"



United States of America (the)

A better and faster way to track methane

SpaceX Starshield

Starshield leverages SpaceX's Starlink technology and launch capability to support national security efforts. While Starlink is designed for consumer and commercial use, Starshield is designed for government use, with an initial focus on three areas:

GiveDirectly Hurricane Relief

In partnership with Google and Propel, GiveDirectly is automating cash aid to Propel users living in areas deemed damaged by satellite imagery

The State of Satellites

This is an exploratory overview of current and upcoming sources of data, processing pipelines, and data products. It is aimed at helping non-experts explore and harness the unfolding revolution of Earth observation, with an emphasis on understanding current capabilities and project development considerations.

Facebook AI mapping for humanitarian aid and development

Building on our previous publication of similar high-resolution population maps for 22 countries, we're now releasing new maps of the majority of the African continent, and the project will eventually map nearly the whole world’s population. Using weakly- and semi-supervised learning.

Back to Top