Marines and soldiers at the squad level could soon have their own kind of “attention warning” system while on foot patrol ― much like modern car drivers have for lane changing on busy highways.
A combination of systems being tested soon by Marines with the 31st Marine Expeditionary Unit out of Okinawa, Japan, are helping Marines at the lowest-unit levels use artificial intelligence and sensing capabilities to know if there are threats of drones overhead or simply to remind them to check their left flank.
The experiments are the last for a yearslong program called Squad X, run by the Defense Advanced Research Projects Agency. Since at least 2017, the program delved into four key technology areas as they applied to the squad: precision engagement, non-kinetic engagement, squad sensing and squad autonomy.
Marines and soldiers will get to test run new ways to use robots to achieve squad-level missions.
Much of the past three years has been spent on the squad sensing and squad autonomy areas, said retired Army Lt. Col. Philip Root, who manages the Squad X program as deputy director at DARPA.
The goal is to put electronic warfare, sensing and networking options usually only available at the platoon or even company level in the hands of a squad of Marines or soldiers.
The long-term goal is to give a squad the capability to effectively patrol what used to be a brigade-sized area through connecting with a suite of sensors and a network of fires options.
Two key systems have evolved in their experiments, many done at the Marine Corps Air Ground Combat Training Center in Twentynine Palms, California.
Those systems are the Lockheed Martin Augmented Spectral Situational Awareness and Unaided Localization for Transformative Squads, or ASSAULTS system, and the CACI BITS Electronic Attack Module.
The BEAM Squad System finds threats using both cyber methods and radio frequencies, translating the information that’s undetectable otherwise. The ASSAULT system uses a small network of robots like hunting dogs for the squad, spotting threats and then allowing the unit to then target the threat with various means.
Marines on the 31st MEU will be working with the CACI BEAM radios as they deploy this year, Root said. He couldn’t share specific information other than to say that they will work with Marines to develop a “library of signals” relevant to their area of operations that will help Marines maintain “asymmetric situational awareness both while afloat and ashore.”
Marines with the 31st MEU are part of Commandant Gen. David Berger’s new emphasis on the Pacific, countering China and a vastly more complex signals terrain than existed only a few years ago.
The work in that region includes putting doctrinal efforts such as littoral operations in a complex environment and expeditionary advanced based operations, which could involve units as small as a squad or as large as a company to occupy small atolls for short periods, supporting naval operations by directing fires and relaying signals information.
The Squad X options predate some of those Marine Corps changes, but Root sees them as a natural fit. Electronic warfare equipment and capabilities rarely have been available at such a low level, but those units will be in the forefront of any future fight, meaning that a squad’s ability to sense its environment and pass that information along or access a network node to bring in higher echelon fires will be critical to success.
Beyond sensing with the BEAM system, Root points to the work with the Lockheed Martin ASSAULT system that will give Marines at the squad level access to artificial intelligence and a kind of robot battle buddy.
“We want robots that have a bias towards action,” Root said.
The problem currently is that it’s hard for troops to trust robots to take independent action, such as serve a spotter such as a drone automatically checking out suspicious activity or an unseen corner of the terrain.
“But if we can train robots to estimate what the enemy is doing, then we can respond,” Root said.
And that goes beyond the squad leader level.
Root said that early experimentation was focused on giving the squad leader a menu of options and keeping that position best informed. But experimenters quickly discovered that squad leaders tend to have a handle on the area and their teammates.
That meant that the AI advantage could be pushed further down the ranks for even more basic tasks.
For example, Root said, the system could provide an audible or visual alert on the heads-up display if a squad member hadn’t checked an area for a while, or if the unit’s drone escort spots movement.
The heads-up display can also track eye movement to tell where the soldier or Marine is looking, see if they’re getting fatigued and to better keep them focused.
The BEAM system can tap into a whole local network to surveil the area, from vehicle-mounted to backpack mounted to drones accompanying the squad.
“Can I create my own electronic surveillance bubble?” Root said. That is a question the Marines of the 31st MEU will be testing.
By creating that bubble, the squad can then react better to threats that it otherwise wouldn’t have spotted, and even share that information throughout the squad, nearly instantaneously.
One node in the squad might detect the threat, then suddenly it shows up as a red dot on the map that all of the dispersed squad shares.
“Now everyone pivots to that threat,” he said. “You want to share that threat but you don’t want to have to get on the radio.”
Root said the Squad X precision engagement goal was met with a precision-guided 40 mm grenade. That munition was combined with a rocket assist and commercially-available components such as mobile phone technology to allow troops to hit targets up to 1 km away.
The standard 40 mm grenade launched by conventional squad weapons maximum range is about 400 meters.
That tech was turned over to the services for further development.
The project has also demonstrated AI capability that could be added to the Army’s Integrated Visual Augmentation System, or IVAS, helping better distinguish targets and spot electromagnetic spectrum threats as well as deciphering what information the user needed to see at the right time to avoid flooding them with too much data.