WASHINGTON — The Defense Advanced Research Projects Agency will accept proposals next month for three initiatives aimed at improving synthetic aperture radar technology in satellites.
Governments and industry are increasing investments in satellite imagery and SAR, in particular. Because SAR sensors rely on radar, they can produce images of the Earth at night and in all-weather conditions, unlike traditional electro-optical systems. The capability is especially useful for tracking movements or changes on the ground and has been in high demand during Russia’s invasion of Ukraine.
DARPA issued broad agency announcements for the efforts in February and March focused on improving SAR technology in three areas: automated object recognition; distributed radar image formation; and digital signal processing. Responses for all three are due in May.
The automated object recognition project, which DARPA has dubbed “Fiddler,” is focused on using machine learning and computer vision methods to create training data that can be used to improve existing ML algorithms.
“Performer methods will learn from real SAR images to generate or render synthetic SAR images at new imaging geometries or configurations,” the Fiddler notice states. “Performers will then demonstrate the generation of diverse training data from a few real examples to rapidly train robust SAR object detection methods.”
DARPA is interested in maritime applications for Fiddler, noting that while SAR object detection in coastal regions has improved significantly, there is a need for better methods to classify objects that are moving.
“For stationary objects, such as many terrestrial objects-of-interest, large training sets can be acquire over time to cover as many of the possible imaging variations,” the notice states. “Maritime environments are much more challenging because most objects-of-interest and the background scene are always in motion.”
The effort will include three phases and span more than three years, and DARPA plans to award multiple contracts, though the notice doesn’t mention how much funding is available for the effort.
The second project, Digital Radar Image Formation Technology, is part of DARPA’s broader vision for so-called mosaic warfare, which is focused on creating highly complex networks composed of smaller systems. Through DRIFT, DARPA wants to use clusters of SAR satellites and develop novel algorithms that “enable revolutionary advances in science, devices or systems.”
DARPA expects to make awards to multiple vendors for the program, which will have three phases over three years.
For the third effort, called Massive Cross-Correlation, DARPA seeks proposals to improve digital signal processing for SAR systems using hybrid architectures. At the end of the four-year effort, which will include multiple vendors with varied technical solutions, DARPA hopes to demonstrate the ability to process larger amounts of data and improve power efficiency.
Courtney Albon is C4ISRNET’s space and emerging technology reporter. She has covered the U.S. military since 2012, with a focus on the Air Force and Space Force. She has reported on some of the Defense Department’s most significant acquisition, budget and policy challenges.
