DARPA’s Vision of Future War — Swarms of Missiles and Drones

Robert Beckhusen
War Is Boring
Published in
5 min readMar 31, 2015


SoSITE project wants to overwhelm air defense systems


One of the most important jobs for an air force is suppressing enemy air defenses. It means hacking, jamming or otherwise blowing up radars and anti-aircraft missile sites — often during the opening stages of a war.

Now a new project from the Defense Advanced Research Projects Agency envisions a near-future strategy using a cargo plane converted into a missile and drone-packing mothership — all commanded by an F-35 stealth fighter.

In short, it’s about how to build a communications network between the mothership, drones and the fighter jet — which work together to destroy an enemy air defense site before it can shoot back.

The agency calls it the Systems of Systems Integration Technology and Experimentation program. It’s a dry and a mouthful, but mainly refers to the software that controls the whole network.

Having the missiles, planes and drones is one thing. It’s another thing to get them to communicate with each other without overwhelming the F-35’s pilot — and without being hacked.

It’s about distributing “air warfare capabilities across a large number of interoperable manned and unmanned platforms,” DARPA stated in a March 30 press release. “The vision is to integrate new technologies and airborne systems with existing systems faster and at lower cost than near-peer adversaries can counter them.”

Here’s a video showing how the agency sees this in action. An F-35 flies toward a hostile radar and surface-to-air missile site, followed by a “mission truck akin to a modified C-130” that’s loaded with missiles and drones.

SoSITe concept. Darpa video

The C-130 functions like a mothership, and launches several largely-autonomous drones toward the target. The drones close in, jam the enemy radar and transmit targeting data to the F-35.

Like a middleman, the fighter pilot’s job is to look at what the drones see, confirm they’re looking at an enemy target — and then order the mothership to unleash its missiles. Computer algorithms sort the drones’ data so the pilot only sees what he or she needs to see.

“The mission truck launches a swarm of small low-cost cruise missiles, or LCCMs, that speed toward the enemy radar target,” the DARPA video narrated. “While each missile has a relatively small warhead, collectively they can have a tremendous impact.”

There’s little the air defense site can do about it.

In the video, the missile launchers destroy several incoming LCCMs … but there’s too many. Like a swarm of mosquitoes, you can’t swat them all. The other idea is that it costs less to blow up the radar with a swarm of cheap, tiny missiles than it costs the enemy to try — in vain — to shoot them down.

At top — the mothership launches its missiles. Above — the missile swarm takes casualties. DARPA video captures

This makes the tactic — in theory — asymmetrical, like a guerrilla army using a fusillade of cheap rocket-propelled grenades to destroy a big, expensive armored vehicle. The mothership in this scenario is also beyond the range of the missile launchers on the ground — that’s important, too.

The project is still in its conceptual phase. The agency wants to begin experiments in 2017 and scale it up to testing “integrated air-air and precision strike kill chains” in 2019.

But if the communications network is unreliable or overly-complicated, then it might not work. Then there’s the question of how to stop a hacker from breaking into the network.

DARPA wants the system to rely on modular, “open” software architectures. If the Air Force wants to upgrade any part of the system, engineers could simply upload new “apps” developed separately — rather than rebuilding the entire system from scratch.

But this makes the software easier to hack. For instance, a hacker might be able to “spoof” the software into uploading a piece of malware that it thinks is an upgrade.

“The more flexible an interface is, the more it may potentially be vulnerable to an attacker leveraging that flexibility to deny or access an interface,” the agency’s request for proposal stated.

What makes the concept work — in theory — is that the drones fly, track and jam largely on their own, while being cheap to produce. At the same time, they’re packing some pretty sophisticated sensors. Same goes for the missiles.

But having the fighter pilot control the drones would likely be way too much micro-management. Which means that if DARPA can’t come up with drones that are cheap, small and autonomous enough, then the whole system breaks down. It becomes too complicated, too expensive or both.

Plus, they have to be secure against hacking — and have crystal-clear datalinks.



Robert Beckhusen
War Is Boring

Editor at War Is Boring. Email: firstnamelastname (at) gmail.