Vice Adm. Robert Sharp, Director of the US National Geospatial Intelligence Agency (NGA), on Monday highlighted the ongoing efforts at the agency to leverage Artificial Intelligence (AI) and apply Machine Learning (ML) algorithms to enable GEOINT analysis at scale. “The foundation GEOINT business has continuously evolved over the past several years as it gets faster, more precise, and more resilient,” he said.
Delivering his keynote address at GEOINT 2022 Symposium here in Aurora, Colorado, on April 25, Vice Adm. Sharp specifically underlined that the Federal Budget gives NGA the operational control of Project Maven’s GEOINT AI services and capabilities from the Office of the Under Secretary of Defense for Intelligence and Security. That includes responsibility for labeled data, AI algorithms, Test and Evaluation capabilities, and the platform.
“This move brings together two major Department of Defense AI and ML efforts under one roof,” he said, adding that the agency saw this opportunity to provide strategic leadership and a corporate approach to GEOINT AI investments.
In the coming months, NGA will be calling on the industry to collaborate move forward together, “so we can deliver GEOINT at the pace that our warfighters and decision makers need”.
“We have to be able to keep up with rapidly emerging digital trends. We have to be able to accelerate our ability to provide detections at the speed of mission, to give our customers tactical, operational and strategic advantage,” he said.
The Director brought attention to the agency’s latest Tech Focus Areas, which was distributed as part of the conference packet. The updated version starts with a graphic that explains where it is heading, both as an agency and as a community. The graphic gives an idea of how its workflows are changing from human-initiated to machine-initiated.
“We need to make these changes in order to increase our speed, and stay ahead of the growing amount of GEOINT created each and every day,” he said.
The graphic shows that the data types and sources are changing — from U.S. Government-prioritized to all data sources that come from different parts of the innovation ecosystem.
“These concepts aren’t new, but we’ve never put them on paper this clearly or publicly before. We realize these changes won’t be easy, but as I’ve told Team NGA: ‘Hard is authorized’. We all know that we need to continually change and improve, in order to deliver on our mission.”
Beginning next fiscal year, NGA will have complementary Computer Vision efforts that will deliver automated GEOINT detections to both intelligence analysts and warfighters, in its key role as a combat support agency.
As part of its efforts, NGA will also provide the subject matter experts — humans who can train the machine, evaluate it, and make sense of the output. “We will bring together those disparate, sometimes siloed communities of Machine Learning experts, data scientists, GIS experts, and imagery analysts, to improve A.I. model performance, develop standards, and lead interoperability efforts for the GEOINT community,” he said.
Three major projects
The Director highlighted the three Foundation GEOINT projects the agency had been working on.
The first one on his list was the backup to GPS, “which is prudent for us all to be planning for these days”. The initiative is called Magneto, and is about Magnetic Navigation, or MagNav for short. Explaining how the backup GPS will work, the NGA Director said the system uses the magnetic field frozen in the Earth’s crust. To provide navigational data. The system is not quite as accurate as GPS, but could serve as a great passive system for all weather, day-and-night, and, as the agency claimed, pretty much impossible to spoof or jam.
“As a backup to GPS, it offers resiliency. We’ve already got partners in the Navy and the Air Force requesting MagNav products – they already see the value, and we think others will too,” he said.
To add here, the NGA Magneto project, though heartening, is not THE alternative to the GPS that the US government has been working on. The responsibility currently lies with the Department of Transportation which has been developing a terrestrial PNT.
The call for GPS backup has gained momentum over the recent years with the GPS system found to be increasingly vulnerable – open to jamming and spoofing. The Ukraine crisis has brought the need in sharper focus with repeated instances of GPS jamming being reported from Ukraine and the areas surrounding it by Russian forces.
HawkEye 360 Detects GPS Jamming Along Ukraine Border Prior to Russian Invasion
GPS is crucial to critical infrastructure in the United States, as they depend on precise timing and navigation signals from the GPS satellites. If the GPS system failed, all kinds of infrastructure would critically suffer, including transportation and telecom networks.
To make matters worse, Russia has already developed a terrestrial PNT system (which can be used when signals from space – whether GPS or GLONASS are not available). A few other countries including China, South Korea, Iran and Saudi Arabia also have terrestrial PNT. The US overdependence on GPS for PNT has been a concern for many years. Back in 2004, President George W. Bush first called for the establishment of an alternative backup system for GPS, but it was only in December 2018, President Donald Trump signed into law the National Timing Security and Resilience Act of 2018, which tasked the Secretary of Transportation with establishing a terrestrial backup timing system for GPS within two years.
On January 15, 2021, during its last days in office, the Trump Administration issued yet another policy directivehighlighting the importance of the GPS, and underlining the need to develop an alternative PNT. The Memorandum on Space Policy Directive 7 (SPD-7), established the implementation actions and guidance for United States space-based PNT programs and activities for America’s national and homeland security, civil, commercial, and scientific purposes.
Arctic gravity data
The second project update the NGA Director spoke about was the gravity data collected by a Coast Guard team which recorded how the Arctic is changing and what implications these changes could have on the planet.
“Things are changing rapidly in the Arctic, and that part of the planet is taking on increased strategic importance – this is critical data for us to have,” he said.
In addition to being a climate risk, the fast-melting Arctic has opened up new shipping routes. This has changed the geopolitical implications around the Arctic, which was till recently characterized by cooperation. The current political tensions could also see Russia taking a tougher stance on its earlier ambitious claims on the Arctic, leading to higher military activity on both sides, thus endangering the already fragile ecology of the region.
Ukraine Crisis Could Have Direct Impact on the Arctic
The Ukraine crisis has further added fuel to the fire, as seven members of the Arctic Council quit in view of the sanctions against Russia, the current chair of the council.
Dr. Stacey Dixon, Principal Deputy Director of National Intelligence, is scheduled to give further details about the Arctic and its geopolitical implications on Tuesday as part of her keynote.
Source Maritime Automated Processing System
The third and the foundation project update that Vice Adm. Robert Sharp shared with the audience as part of his keynote was about the Source Maritime Automated Processing System. SMAPS for short, the system basically speeds up and simplifies navigation warning alerts, and is slated to become operational by the end of this month.
“Our Maritime Safety Watch desk gets almost a thousand messages a day on Safety of Navigation issues: icebergs, man overboard, missile and rocket launches, military exercises, pirates, sea monsters, you name it,” he said.
The desk takes all those individual reports, organizing and validating them, before issuing warning broadcasts. The team issues about 30 to 40 warning broadcasts every day.
SMAPS works with natural language processing and basic machine learning, to improve the system, so that the officers can focus on analysis, and turn over the data administration to the software. That reduces the time it takes from receiving incoming messages to sending out alerts to the Navy, the Coast Guard, and commercial mariners by up to 50 percent. “It also reduces all the standard human error. This effort is literally going to help save lives, and it’s a great example of the power of partnering with machines,” he added.
