How NASA Is Testing AI to Make Earth-Observing Satellites Smarter

How NASA Is Testing AI to Make Earth-Observing Satellites Smarter

5 min read

Preparations for Next Moonwalk Simulations Underway (and Underwater)

Cloud cover image taken above earth
Cloud cover can keep optical instruments on satellites from clearly capturing Earth’s surface. Still in testing, JPL’s Dynamic Targeting uses AI to avoid imaging clouds, yielding a higher proportion of usable data, and to focus on phenomena like this 2015 volcanic eruption in Indonesia Landsat 8 captured.
NASA/USGS

A technology called Dynamic Targeting could enable spacecraft to decide, autonomously and within seconds, where to best make science observations from orbit.

In a recent test, NASA showed how artificial intelligence-based technology could help orbiting spacecraft provide more targeted and valuable science data. The technology enabled an Earth-observing satellite for the first time to look ahead along its orbital path, rapidly process and analyze imagery with onboard AI, and determine where to point an instrument. The whole process took less than 90 seconds, without any human involvement.

Called Dynamic Targeting, the concept has been in development for more than a decade at NASA’s Jet Propulsion Laboratory in Southern California. The first of a series of flight tests occurred aboard a commercial satellite in mid-July. The goal: to show the potential of Dynamic Targeting to enable orbiters to improve ground imaging by avoiding clouds and also to autonomously hunt for specific, short-lived phenomena like wildfires, volcanic eruptions, and rare storms.

This graphic shows how JPL’s Dynamic Targeting uses a lookahead sensor
This graphic shows how JPL’s Dynamic Targeting uses a lookahead sensor to see what’s on a satellite’s upcoming path. Onboard algorithms process the sensor’s data, identifying clouds to avoid and targets of interest for closer observation as the satellite passes overhead.
NASA/JPL-Caltech

“The idea is to make the spacecraft act more like a human: Instead of just seeing data, it’s thinking about what the data shows and how to respond,” says Steve Chien, a technical fellow in AI at JPL and principal investigator for the Dynamic Targeting project. “When a human sees a picture of trees burning, they understand it may indicate a forest fire, not just a collection of red and orange pixels. We’re trying to make the spacecraft have the ability to say, ‘That’s a fire,’ and then focus its sensors on the fire.”

Avoiding Clouds for Better Science

This first flight test for Dynamic Targeting wasn’t hunting specific phenomena like fires — that will come later. Instead, the point was avoiding an omnipresent phenomenon: clouds.

Most science instruments on orbiting spacecraft look down at whatever is beneath them. However, for Earth-observing satellites with optical sensors, clouds can get in the way as much as two-thirds of the time, blocking views of the surface. To overcome this, Dynamic Targeting looks 300 miles (500 kilometers) ahead and has the ability to distinguish between clouds and clear sky. If the scene is clear, the spacecraft images the surface when passing overhead. If it’s cloudy, the spacecraft cancels the imaging activity to save data storage for another target.

“If you can be smart about what you’re taking pictures of, then you only image the ground and skip the clouds. That way, you’re not storing, processing, and downloading all this imagery researchers really can’t use,” said Ben Smith of JPL, an associate with NASA’s Earth Science Technology Office, which funds the Dynamic Targeting work. “This technology will help scientists get a much higher proportion of usable data.”

How Dynamic Targeting Works

The testing is taking place on CogniSAT-6, a briefcase-size CubeSat that launched in March 2024. The satellite — designed, built, and operated by Open Cosmos — hosts a payload designed and developed by Ubotica featuring a commercially available AI processor. While working with Ubotica in 2022, Chien’s team conducted tests aboard the International Space Station running algorithms similar to those in Dynamic Targeting on the same type of processor. The results showed the combination could work for space-based remote sensing.

Since CogniSAT-6 lacks an imager dedicated to looking ahead, the spacecraft tilts forward 40 to 50 degrees to point its optical sensor, a camera that sees both visible and near-infrared light. Once look-ahead imagery has been acquired, Dynamic Targeting’s advanced algorithm, trained to identify clouds, analyzes it. Based on that analysis, the Dynamic Targeting planning software determines where to point the sensor for cloud-free views. Meanwhile, the satellite tilts back toward nadir (looking directly below the spacecraft) and snaps the planned imagery, capturing only the ground.

This all takes place in 60 to 90 seconds, depending on the original look-ahead angle, as the spacecraft speeds in low Earth orbit at nearly 17,000 mph (7.5 kilometers per second).

What’s Next

With the cloud-avoidance capability now proven, the next test will be hunting for storms and severe weather — essentially targeting clouds instead of avoiding them. Another test will be to search for thermal anomalies like wildfires and volcanic eruptions. The JPL team developed unique algorithms for each application.

“This initial deployment of Dynamic Targeting is a hugely important step,” Chien said. “The end goal is operational use on a science mission, making for a very agile instrument taking novel measurements.”

There are multiple visions for how that could happen — possibly even on spacecraft exploring the solar system. In fact, Chien and his JPL colleagues drew some inspiration for their Dynamic Targeting work from another project they had also worked on: using data from ESA’s (the European Space Agency’s) Rosetta orbiter to demonstrate the feasibility of autonomously detecting and imaging plumes emitted by comet 67P/Churyumov-Gerasimenko.

On Earth, adapting Dynamic Targeting for use with radar could allow scientists to study dangerous extreme winter weather events called deep convective ice storms, which are too rare and short-lived to closely observe with existing technologies. Specialized algorithms would identify these dense storm formations with a satellite’s look-ahead instrument. Then a powerful, focused radar would pivot to keep the ice clouds in view, “staring” at them as the spacecraft speeds by overhead and gathers a bounty of data over six to eight minutes.

Some ideas involve using Dynamic Targeting on multiple spacecraft: The results of onboard image analysis from a leading satellite could be rapidly communicated to a trailing satellite, which could be tasked with targeting specific phenomena. The data could even be fed to a constellation of dozens of orbiting spacecraft. Chien is leading a test of that concept, called Federated Autonomous MEasurement, beginning later this year.

News Media Contact

Melissa Pamer
Jet Propulsion Laboratory, Pasadena, Calif.
626-314-4928
melissa.pamer@jpl.nasa.gov

2025-094

Powered by WPeMatico

Get The Details…
Anthony Greicius

NASA’s Hubble, Chandra Spot Rare Type of Black Hole Eating a Star

NASA’s Hubble, Chandra Spot Rare Type of Black Hole Eating a Star

6 Min Read

NASA’s Hubble, Chandra Spot Rare Type of Black Hole Eating a Star

2 Elliptical galaxies appear as fuzzy white oval. At right is a smaller fuzzy purple blob. The background of space is black, speckled with tiny white dots.
NASA’s Hubble Space Telescope and NASA’s Chandra X-ray Observatory team up to identify a possible intermediate-mass black hole.
Credits:
NASA, ESA, CXC, Yi-Chi Chang (National Tsing Hua University); Image Processing: Joseph DePasquale (STScI)

NASA’s Hubble Space Telescope and NASA’s Chandra X-ray Observatory have teamed up to identify a new possible example of a rare class of black holes. Called NGC 6099 HLX-1, this bright X-ray source seems to reside in a compact star cluster in a giant elliptical galaxy.

Just a few years after its 1990 launch, Hubble discovered that galaxies throughout the universe can contain supermassive black holes at their centers weighing millions or billions of times the mass of our Sun. In addition, galaxies also contain as many as millions of small black holes weighing less than 100 times the mass of the Sun. These form when massive stars reach the end of their lives.

Far more elusive are intermediate-mass black holes (IMBHs), weighing between a few hundred to a few 100,000 times the mass of our Sun. This not-too-big, not-too-small category of black holes is often invisible to us because IMBHs don’t gobble as much gas and stars as the supermassive ones, which would emit powerful radiation. They have to be caught in the act of foraging in order to be found. When they occasionally devour a hapless bypassing star — in what astronomers call a tidal disruption event— they pour out a gusher of radiation.

The newest probable IMBH, caught snacking in telescope data, is located on the galaxy NGC 6099’s outskirts at approximately 40,000 light-years from the galaxy’s center, as described in a new study in the Astrophysical Journal. The galaxy is located about 450 million light-years away in the constellation Hercules.

Two elliptical galaxies at upper right and lower left appear as fuzzy white ovals. At bottom center is a fuzzy purple blob outlined by a white box. White lines angle outward from the corners of the box to a blow-up of the purple blob. At the center is a white dot labeled HLX-1. The main image also has two bright stars with prominent diffraction spikes. The background of space is black, and is speckled with tiny white dots.
A Hubble Space Telescope image of a pair of galaxies: NGC 6099 (lower left) and NGC 6098 (upper right). The purple blob depicts X-ray emission from a compact star cluster. The X-rays are produced by an intermediate-mass black hole tearing apart a star.
Science: NASA, ESA, CXC, Yi-Chi Chang (National Tsing Hua University); Image Processing: Joseph DePasquale (STScI)

Astronomers first saw an unusual source of X-rays in an image taken by Chandra in 2009. They then followed its evolution with ESA’s XMM-Newton space observatory.

“X-ray sources with such extreme luminosity are rare outside galaxy nuclei and can serve as a key probe for identifying elusive IMBHs. They represent a crucial missing link in black hole evolution between stellar mass and supermassive black holes,” said lead author Yi-Chi Chang of the National Tsing Hua University, Hsinchu, Taiwan.

X-ray emission coming from NGC 6099 HLX-1 has a temperature of 3 million degrees, consistent with a tidal disruption event. Hubble found evidence for a small cluster of stars around the black hole. This cluster would give the black hole a lot to feast on, because the stars are so closely crammed together that they are just a few light-months apart (about 500 billion miles).

The suspected IMBH reached maximum brightness in 2012 and then continued declining to 2023. The optical and X-ray observations over the period do not overlap, so this complicates the interpretation. The black hole may have ripped apart a captured star, creating a plasma disk that displays variability, or it may have formed a disk that flickers as gas plummets toward the black hole.

“If the IMBH is eating a star, how long does it take to swallow the star’s gas? In 2009, HLX-1 was fairly bright. Then in 2012, it was about 100 times brighter. And then it went down again,” said study co-author Roberto Soria of the Italian National Institute for Astrophysics (INAF). “So now we need to wait and see if it’s flaring multiple times, or there was a beginning, there was peak, and now it’s just going to go down all the way until it disappears.”

The IMBH is on the outskirts of the host galaxy, NGC 6099, about 40,000 light-years from the galaxy’s center. There is presumably a supermassive black hole at the galaxy’s core, which is currently quiescent and not devouring a star.

Black Hole Building Blocks

The team emphasizes that doing a survey of IMBHs can reveal how the larger supermassive black holes form in the first place. There are two alternative theories. One is that IMBHs are the seeds for building up even larger black holes by coalescing together, since big galaxies grow by taking in smaller galaxies. The black hole in the middle of a galaxy grows as well during these mergers. Hubble observations uncovered a proportional relationship: the more massive the galaxy, the bigger the black hole. The emerging picture with this new discovery is that galaxies could have “satellite IMBHs” that orbit in a galaxy’s halo but don’t always fall to the center.

Another theory is that the gas clouds in the middle of dark-matter halos in the early universe don’t make stars first, but just collapse directly into a supermassive black hole. NASA’s James Webb Space Telescope’s discovery of very distant black holes being disproportionately more massive relative to their host galaxy tends to support this idea.

However, there could be an observational bias toward the detection of extremely massive black holes in the distant universe, because those of smaller size are too faint to be seen. In reality, there could be more variety out there in how our dynamic universe constructs black holes. Supermassive black holes collapsing inside dark-matter halos might simply grow in a different way from those living in dwarf galaxies where black-hole accretion might be the favored growth mechanism.

“So if we are lucky, we’re going to find more free-floating black holes suddenly becoming X-ray bright because of a tidal disruption event. If we can do a statistical study, this will tell us how many of these IMBHs there are, how often they disrupt a star, how bigger galaxies have grown by assembling smaller galaxies.” said Soria.

The challenge is that Chandra and XMM-Newton only look at a small fraction of the sky, so they don’t often find new tidal disruption events, in which black holes are consuming stars. The Vera C. Rubin Observatory in Chile, an all-sky survey telescope from the U.S. National Science Foundation and the Department of Energy, could detect these events in optical light as far as hundreds of millions of light-years away. Follow-up observations with Hubble and Webb can reveal the star cluster around the black hole.

The Hubble Space Telescope has been operating for more than three decades and continues to make ground-breaking discoveries that shape our fundamental understanding of the universe. Hubble is a project of international cooperation between NASA and ESA (European Space Agency). NASA’s Goddard Space Flight Center in Greenbelt, Maryland, manages the telescope and mission operations. Lockheed Martin Space, based in Denver, also supports mission operations at Goddard. The Space Telescope Science Institute in Baltimore, which is operated by the Association of Universities for Research in Astronomy, conducts Hubble science operations for NASA.

Share

Details

Last Updated
Jul 24, 2025
Editor
Andrea Gianopoulos
Contact
Media

Claire Andreoli
NASA’s Goddard Space Flight Center
Greenbelt, Maryland
claire.andreoli@nasa.gov

Ray Villard
Space Telescope Science Institute
Baltimore, Maryland

Powered by WPeMatico

Get The Details…

GVIS History

GVIS History

4 Min Read

GVIS History

Three NASA employees sit in a large room filled with glowing television and computer screens.

As part of NASA Glenn’s Scientific Computing and Visualization Team, the GVIS Lab has a storied visual and technological history. 

Credits:
NASA

GVIS: the ICARE Era

In 1982, a $20 million supercomputer was brought to NASA Glenn. Scientists at NASA Glenn were becoming increasingly reliant on computer simulations to test their experiments. Advancements in computer technology allowed a different type of testing environment — one that revolved around virtual models and data over physical observation. The benefits of this method included a decrease in costs, a decrease in associated risk, faster turnaround, and more data.

But this method of experimentation created a problem: With data-point counts somewhere in the millions, it was a challenge for scientists to even begin to look at their own collected data. In short, there was simply too much data to be analyzed. To solve this problem, NASA Glenn built the Interactive Computer Aided Research Engineering system (ICARE) in the center’s Research Analysis Center.  

Taking up several rooms, consisting of 22 total workstations, and costing a grand total of $20 million, the ICARE system was a way for scientists to examine their data through the aid of supercomputer visualizations. Using both graphical and modular methods, ICARE’s visualizations revealed and shared information in ways that traditional methods could not match. 

The construction and implementation of the ICARE system was revolutionary to both the center and NASA as a whole. Before 1982, NASA already had an established interest in powerful computers; however, the ICARE system took NASA into the era of supercomputing. ICARE also brought increased attention to the value and power of scientific visualization. 

The Creation of GVIS

In 1989, it was time for an upgrade. NASA Glenn wanted the latest scientific visualization technology and techniques for its scientists, so the center expanded the Research Analysis Center to make room for the new Graphics and Visualization Lab (GVIS). The GVIS Lab acquired cutting-edge graphics technology, including studio-quality TV animation and recording equipment, stereographic displays, and image processing systems. Later, the High-Performance Computing Act of 1991 provided funding and opportunities to add high-speed computing, virtual reality, and collaborative visualization to its fleet of tools.

A rectangular white device sits in the very back of a large, beige, mostly empty, windowless room.  Some rolling chairs and brown tables sit near the device in the back of the room.
The secure supercomputing space that would eventually become the Graphics and Visualization Lab (GVIS), shown in 1989.
NASA

During this period, the GVIS Lab was responsible for assisting NASA Glenn scientists who needed help visualizing their data. The lab was also tasked with inventing new visualization techniques and promoting NASA Glenn’s activities though tours, videos, and other outreach programs. Some of the techniques the lab developed included particle tracking, iso-surface contours, and volume visualization. Tour guests included school children, corporate VIPs, local and national politicians, TV news media, and researchers from other national labs. Using state-of-the-art recording and editing hardware, the GVIS Lab regularly shared work both inside and outside of NASA.   

As other labs and researchers began to gain access to their own scientific visualization tools, the GVIS Lab shifted its focus to experimenting with virtual reality- and augmented reality-based visualizations.

GVIS Now

Today, the GVIS Lab has the same mission that it had in 1989: to apply the latest visualization and human interaction technologies to advance NASA’s missions. The team takes pride in pushing the limits of scientific visualization and computer science, helping fellow researchers make sense of their data, and inspiring the next generation through demonstrations and presentations. Computational technology has come a long way since the days of ICARE, but GVIS has continued to explore current and cutting-edge technologies. 

In addition to scientific visualization and experimental computational technologies, the GVIS Lab now also specializes in virtual design, interactive 3D simulations, natural user interface development, applications of computer science, and mission scenario visualizations. The team uses the latest edition of 3D programs and VR devices to experiment with how these systems can be used to visualize data, pushing their input and output capabilities. 

With all this technology, GVIS also supports the visualization of a wide variety of 3D data and models such as CAD, point clouds, and volume data. Additionally, the lab is capable of high-impact data visualization, web-based visualization, time-accurate data representation, and designing and testing CAD models in virtual reality.

A large group of NASA employees stand smiling in front of a skeleton of a dinosaur.  One employee in the middle holds a sign reading: "NASA".
The Graphics and Visualization Lab (GVIS) team attends a STEM outreach event at the Cleveland Museum of Natural History.
NASA

Public Engagement

Outside of the lab, GVIS has a longstanding history of taking its technology demonstrations across the city, throughout the country, and around the world. The team has extensive experience organizing, presenting, and facilitating STEM-based educational outreach for a variety of different events and venues. Inside the lab, GVIS supports the education and career exploration of its high school and college interns through mentorship, community engagement opportunities, and access to cutting-edge technology.

Contact Us 

Need to reach us? You can send an email directly to the GVIS Team (GRC-DL-GVIS@mail.nasa.gov) or to the team leader, Herb Schilling (hschilling@nasa.gov).

Share

Details

Last Updated

Jul 23, 2025

Powered by WPeMatico

Get The Details…
Ann Koppitch

Feeling the Heat: Perseverance Looks for Evidence of Contact Metamorphism 

Feeling the Heat: Perseverance Looks for Evidence of Contact Metamorphism 

2 min read

Feeling the Heat: Perseverance Looks for Evidence of Contact Metamorphism 

A color photo from the surface of Mars shows very rocky terrain in the foreground sloping downward from left to right. The ground ranges in shades from orange-tan to brown, while the rocks are mostly gray. A trio of hills rises up in the distance.
NASA’s Mars Perseverance rover acquired this image of the boulders along the contact at Westport, using its Mastcam-Z Left Camera, one of a pair of cameras located high on the rover’s mast. The rover acquired the image on July 10, 2025 — Sol 1560, or Martian day 1,560 of the Mars 2020 mission — at the local mean solar time of 11:23:38.
NASA/JPL-Caltech/ASU

Written by Melissa Rice, Professor of Planetary Science at Western Washington University

Following a short break for the July 4th holiday, Perseverance drove westward to a site called “Westport,” where the clay-bearing “Krokodillen” unit meets an olivine-bearing rock formation. It is possible that the olivine-rich rocks are an intrusive igneous unit, meaning they could have formed when molten magma from deep within Mars got pushed upwards and cooled under the surface. If that’s the case, Westport could preserve a dramatic moment in Mars’ history when hot, molten material intruded into existing rock formations.  

Those intrusive processes are common on Earth, and the heat of the intruding magma can fundamentally alter the surrounding geology through a process called “contact metamorphism.” The heat from the intrusion will “bake” nearby rocks, creating new minerals and potentially new environments for microbial life. Conversely, the intrusive rocks get rapidly “chilled” where they meet preexisting solid rock formations. 

At Westport, Perseverance is looking for evidence that the Krokodillen rocks at the contact were baked, and that the olivine-bearing rocks at the contact were chilled. Images from the Mastcam-Z instrument reveal that the contact is littered with intriguing dark, rubbly rocks alongside lighter-toned, smooth boulders. Both rock types are proving challenging to study. 

The dark fragments are too small and rough for Perseverance’s standard abrasion techniques, but the rover cleared off the surface of a rock called “Holyrood Bay” with its gas Dust Removal Tool (gDRT). Perseverance also tried to abrade a nearby boulder named “Drake’s Point,” but the rock shifted to the side, causing the abrasion to stop short. The science questions here are compelling enough, however, that Perseverance will keep trying to look within the rocks at this important boundary. 

Share

Details

Last Updated
Jul 22, 2025

Related Terms

Powered by WPeMatico

Get The Details…

Curiosity Blog, Sols 4607-4608: Deep Dip

Curiosity Blog, Sols 4607-4608: Deep Dip

3 min read

Curiosity Blog, Sols 4607-4608: Deep Dip

A grayscale photo from the Martian surface shows a landscape of undulating but mostly smooth terrain, in light gray. A plateau on the right side of the foreground drops down toward the left about mid-frame, in a gentle curve. A ridge across the upper middle of the frame creates a V-shaped horizon line. Beyond that in the distance, filling the bottom of the V, a jagged line of rocks is brilliant white, looking like snow-capped mountains.
NASA’s Mars rover Curiosity acquired this image, looking toward the upper slopes of Mount Sharp, using its Left Navigation Camera (Left Navcam) on July 20, 2025. Curiosity captured the image on Sol 4605, or Martian day 4,605 of the Mars Science Laboratory mission, at 18:58:26 UTC.
NASA/JPL-Caltech

Written by Deborah Padgett, MSL OPGS Task Lead at NASA’s Jet Propulsion Laboratory 

Earth planning date: Monday, July 21, 2025

Curiosity continues our exploration of the fractured boxwork terrain on the slopes of Mount Sharp. After a successful 5-meter drive (about 16 feet), our rover is resting in a hollow on its way to a boxwork ridge viewpoint. Over the weekend, Curiosity began an atmospheric observation with the SAM instrument, which will continue into today’s plan. Because the SAM instrument is complex and powerful, it uses a great deal of energy when it operates, causing what we call a “deep dip” in the battery charge level. This means that we have to wait a bit after the SAM observations complete for the battery to recharge enough for Curiosity to observe its surroundings with other science instruments, or move its arm or wheels. For this reason, the plan today does not include a drive, and contact science at this location will be done on the second sol of the plan. 

On Sol 4607, Curiosity will begin the day with SAM atmospheric composition activity, which will run for several hours. After it finishes, we will use the rover’s navigation camera to perform a cloud altitude observation, looking for cloud shadows on the upper reaches of Mount Sharp, and clouds drifting by overhead at the zenith. Overnight, Curiosity’s battery will recharge, allowing us to perform a targeted science block on the morning of Sol 4608. This starts with Navcam observations of dust opacity across the floor of Gale Crater, then a measurement of dust in the air toward the Sun with Mastcam. Curiosity then turns Mastcam toward the ridge ahead to obtain a 15×1 mosaic on target “Cueva De Los Vencejos Y Murcielagos (Cave of Swifts and Bats).” Afterwards, Mastcam will look back along Curiosity’s tracks, hoping to see freshly broken rocks and determine the texture of disturbed ground. Next, ChemCam’s laser spectrograph will zap a nodular rock pillar named for the famous high-altitude “Lake Titicaca” bordering Bolivia and Peru. A second ChemCam observation with the RMI telescopic camera will study stratigraphy on the Mishe Mokwa butte with a 5×2 image mosaic. Mastcam will finish off this science block by looking at the pits left behind by the ChemCam laser on target “Lake Titicaca.”  

In the afternoon, Curiosity’s arm will reach out to brush the dust from the bedrock target “La Tranquita,” then observe it with the MAHLI microscopic imager and APXS. MAHLI and APXS will also investigate plate-like rock formations at target “Aqua Dulce.” A third target with more complex rock structures dubbed “Paposo,” after a natural monument along the Pacific Coast of northern Chile, will be imaged only by MAHLI. The next morning will include another targeted science block. Curiosity will then drive away toward the next viewpoint in the boxwork terrain of Mars.

Share

Details

Last Updated
Jul 22, 2025

Related Terms

Powered by WPeMatico

Get The Details…