WASHINGTON — Northrop Grumman says a new investment into small startup Deepwave Digital will allow it to push data processing much closer to the point of collection, decreasing the amount of data that needs to be transported and getting products to war fighters faster.

Northrop Grumman will install Deepwave Digital’s artificial intelligence solution on airborne and on orbit payloads, said Chris Daughters, the company’s vice president of research, technology and engineering for aeronautics systems. With the AI incorporated into the payload, the satellite or aircraft will no longer need to send data back to the ground to be processed. Instead, the payload will process the data itself.

The innovation, said Daughters, is in incorporating the AI software with the hardware used in various Northrop Grumman payloads.

“Deepwave Digital really has an innovative architecture that blends artificial intelligence with some advanced hardware in the RF [radio frequency] domain, whereas a lot of artificial intelligence in the past has always been focused on, I’ll just say, mining and scrubbing really big data or maybe doing customized things with video or audio,” said Daughters.

In other words, the AI is able to process data at the edge, at the point of collection, instead of sending it down to Earth to be processed independently.

“We have a lot of RF payload systems that collect information or collect intelligence for the war fighters and the government as a whole, and by embedding this innovative hardware and AI software, we are able to essentially filter and optimize and scrub and prioritize the data much nearer to the point of collection for that information,” said Daughters.

AI has been key to the military’s efforts to manage the torrent of data collected by government and commercial satellites, which is simply too much for human analysts to sort through on their own. While progress has been made in using AI to process satellite imagery, military officials are now tackling a related problem: the sheer amount of raw data being pushed out over the networks for processing at remote locations. To reduce the bandwidth needed for satellite imagery, U.S. Army officials say the military needs to shift to edge processing, meaning applying AI to the data and then sending the finished product out over the network instead of all of the raw data.

“Edge processing is something that we’re very interested in for a number of reasons. And what I mean by that is having smart sensors that can not only detect the enemy, [but] identify, characterize and locate, and do all those tasks at the sensor processing,” said John Strycula, director of the Army’s task force focused on intelligence, surveillance and reconnaissance, at an October AUSA event.

“If I only have to send back a simple message from the sensor that says the target is here ― here’s the location and here’s what I saw and here’s my percent confidence ― versus sending back the whole image across the network, it reduces those bandwidth requirements,” he added.

Daughters said Northrop Grumman is working to incorporate Deepwave Digital’s AI solution into products in development or in limited production in the “very near term.”

“The capability exists now. We just need to integrate it in with the systems,” said Daughters. “Our research and technology organization is already looking at how we could be injecting this hardware and AI capability in some of the systems that exist right now or very near term.”

While the company is applying the AI to both space and airborne systems, the technology can be more easily integrated with airborne systems, where payloads can be more easily accessed or swapped out.

About Nathan Strout

Nathan Strout covers space, unmanned and intelligence systems for C4ISRNET.

Share:
More In Intel/GEOINT