Smarter Satellites: AI-Powered Earth Observation Revolutionizes Data Collection
For decades,satellites have functioned primarily as passive observers,diligently capturing images of Earth regardless of their content. But a recent breakthrough, spearheaded by a collaboration between NASA’s Jet Propulsion Laboratory (JPL), UK-based Open Cosmos, and Dublin’s Ubotica, is changing that paradigm. This innovation introduces intelligent, on-board processing that allows satellites to decide what to image, dramatically improving the efficiency and value of Earth observation data.
A Landmark Test in Space
Earlier this month, the team successfully tested this new approach aboard a compact satellite built by open Cosmos. The key component? A machine learning processor developed by Ubotica. Here’s how the test unfolded:
The satellite scanned 500km ahead, capturing a preview image.
Ubotica’s AI instantly analyzed the scene for cloud cover.
if clear skies were detected, a high-resolution image was taken.
If clouds were present, the image capture was skipped, conserving valuable resources.
This seemingly simple process represents a significant leap forward. As Ben Smith of JPL,which funds the Dynamic Targeting work,explains,”if you can be smart about what you’re taking pictures of,then you only image the ground and skip the clouds.” This translates to a much higher proportion of usable data for scientists.
From Passive Collection to Intelligent Analysis
Traditionally, satellites beam back vast amounts of data – the good, the bad, and the cloudy. Scientists then spend considerable time and effort sifting through this backlog. Brian Quinn, chief strategy officer at Ubotica, highlights this inefficiency: “Until now, satellites have merely acted as passive data collectors.”
this post-processing delay can be critical. For example, identifying events like wildfires or harmful algal blooms could take days.As Quinn noted in a recent NASA article, this delay hinders rapid response and informed decision-making.
Expanding the Capabilities: Beyond cloud Detection
The potential applications of this technology extend far beyond simply avoiding clouds. NASA, Ubotica, and Open Cosmos envision a future where satellites can proactively identify and monitor:
Wildfires
Volcanic eruptions
Severe storms
This real-time detection capability will be invaluable for disaster response, environmental monitoring, and scientific research.
Building on Past Successes
This latest test isn’t an isolated event. It builds upon a strong foundation of collaboration:
2021: Ubotica demonstrated real-time AI cloud detection aboard the International Space Station (ISS) in partnership with JPL.
* 2024: Open Cosmos launched HAMMER, an AI-powered satellite featuring a hyperspectral camera and Ubotica’s machine learning processor.These previous successes paved the way for the recent in-orbit exhibition, solidifying the viability of intelligent satellite imaging.
This new era of ”edge computing” in space promises to deliver more timely, relevant, and actionable insights from our planet’s orbit. You can expect to see this technology rapidly integrated into future satellite missions, transforming how we understand and respond to our changing world.