possible, to perform the swarm mission successfully
during the demonstration.
“We want the swarm to do something during the
demonstration — fly-by, swarm control, turning
maneuvers, expanding and tracking how far they are
separated,” Mastroianni said.
The demonstration will be similar to a successful
one conducted last June on the James River in Virginia
involving unmanned surface vessels. During that test,
several unmanned watercraft used a technology called
Control Architecture for Robotic Agent Command and
Sensing (CARACaS) to form a picket around an asset
and protect it from an impending adversary.
In contrast, LOCUST UAVs all are configured iden-
tically and use the same radios, processing capability
and software. One UAV can be designated a “parent,”
with the others serving as subordinate “children,” with
roles adjustable as mission requirements would war-
rant. For example, if the swarm were to break into
smaller groups, other UAVs in those groups could
assume primary roles, according to Mastroianni
The relations among all UAVs in a swarm are deter-
mined by a series of mathematical equations, which
take location, air speed, altitude, maximum and mini-
mum turning radius and mission objective into consid-
eration, he said.
The surface mission would “remove our Sailors and
Marines from many dangerous situations,” Robert
Brizzolara, the head of the CARACaS
project and a colleague of Mastro-
ianni, said in a Navy press release
following the James River test.
Likewise, during the course of
the flight demonstration, the autonomous capabilities of the swarm
would enable them to make
changes in midair without operator intervention, and thus out of
harm’s way. The UAVs automatically could compensate, for example,
for any instances in which aircraft
malfunction, become disabled or
crash, or the scenario at the target
differs from what programmers
anticipated at the outset. The aircraft could change their aspect in
relation to a target or change the
target itself, if necessary.
“It’s more than just keeping 30
UAVs from running into each other
in the sky,” Mastroianni said. “It’s a
matter of here are contingencies
[and] here’s how you address it.”
Unmanned aircraft that carry
weapons systems would learn from observer aircraft
the best ways to approach targets simultaneously.
Every UAV would launch ordnance in a sequence
where the projectiles would hit their targets at the
same time, preserving the element of surprise.
“They’d have to ‘know’ their priority. If they’re after
a specific location or building, it would need to be programmed. But as autonomy develops, [the UAVs]
could say, ‘Find tanks,’ or whatever the case might be,”
To the Navy’s brain trust, projects like LOCUST and the
surface test of a year ago offer great potential to provide
better protection to Sailors and assets at minimal cost.
“It started because of CNO’s strong interest and desire
for offensive swarming capabilities,” Mastroianni said.
“CNO [Chief of Naval Operations ADM Jonathan W.
Greenert] made a phone call to CNR [then-Chief of
Naval Research RADM Matthew L. Klunder], and we
were in business.”
Networking unmanned platforms offers “a cost-
effective way to integrate many small, cheap and
autonomous capabilities that can significantly improve
our warfighting advantage,” Greenert said after the
2014 surface test.
For next June’s LOCUST demonstration, the 30
Coyotes will be launched from the high-speed vessel
Sea Fighter, based out of Naval Surface Warfare Center
Panama City, Fla.
The Coyote unmanned aerial vehicle (UAV) will be used as part of an Office of
Naval Research (ONR) demonstration of a concept known as LOCUST, or low-cost UAV swarming technology. A swarm of several dozen Coyote UAVs will be
tested at sea. A Coyote UAV is shown here on display at ONR's booth on the
exhibit floor of the 2015 Sea-Air-Space Exposition in National Harbor, Md.