The Navy Littorally Has a Drone Problem

October 25, 2016

For special access to experts and other members of the national security community, check out the new War on the Rocks membership.

The idea of non-state actors attacking from the sea is not new. We merely have to look at attacks by Somali pirates or Al Qaeda’s attack on USS Cole for historical precedent. Similarly, there are centuries of precedent for land-based attacks on ships. But, drone technological advances that are already on the horizon change the situation by diversifying both the types of threats and weaponry and also places them in the hands of non-state actors.

Defending ships worth hundreds of millions of dollars against thousand dollar drones creates a perilous economic game and changes the risk calculus of the Navy’s presence mission. We were recently given an early indication of how this might happen. Houthi rebels in Yemen reportedly fired missiles at the USS Mason multiple times in recent weeks. These conventional attacks by an unconventional force did not successfully hit their target but merit our attention as this scenario may repeat itself in the not-too distant future with unmanned aerial vehicles mounted with explosives.

As discussed by Ulrike Franke, the booby-trapped IED that recently wounded French special operators in Syria showed that low-cost, commercially available, payload-carrying drones are now being weaponized. T.X. Hammes gave a bracing assessment on the “democratization of airpower,” discussing how technologies available to non-state groups could change the character of warfare.  Mark Jacobsen argued that the most dangerous insurgent operated drones will be custom-built rather than purchased “off of the shelf” and adapted from there. As 3D printing technologies advance and become more affordable, they explain, non-state actors will be capable of producing their own deadly drones. However, Hammes and Jacobson attacked the problem largely from the perspective of operations on or over land.  The U.S. Navy must also face this challenge.

How might this threat to our maritime assets manifest itself? The littorals will become evermore dangerous.

Looking further into the future, machine learning could allow these drones to identify and select specific targets with little human involvement, which would be critical on long distance raids over the horizon. Jacobsen explained how his team created drones capable of flying up to 80 miles with a two-pound payload or 40 miles with a four-pound payload. This sort of technology would make land-based attacks on naval ships near the coastline possible.

If non-flagged, or deceptively flagged, seemingly benign vessels sail into territorial waters, they can narrow the gap between the drone launch point and their final target. One can imagine a scenario where multiple non-state ships, similar to the “little blue men” of the Chinese Maritime Militia, close with U.S. warships under the protected veil of a civilian craft before launching waves of drones.

If these launching platforms can get close enough to get their drones within range of U.S. forces, their flying IEDs used en masse would present a serious problem for U.S. vessels. U.S. Navy ships are equipped with either the Ship Self-Defense System (SSDS) or Aegis Combat System to defend against airborne missile attacks. Neither system is designed to defeat small, maneuverable drones. That leaves the close-in weapon system (CIWS) as the lone defender of the ship. CIWS identifies, tracks, and neutralizes close airborne contacts that have penetrated the ship’s long-range defenses. If the swarm of attacking drones were numerous enough, it is likely that a few could slip past the CIWS defense bubble. And a few is all it would take to disable a ship or seriously interrupt operations. For instance, drones could be “trained” to hunt and kill — with shrapnel or explosively formed penetrators — the radar, flight deck, bridge, and communications antennae. Even if only one of these targets were hit, the ship would be taken out of the game.

The Navy is in the process of developing lasers to protect against drone attacks. However, these lasers are not likely to enter the fleet until 2020 and have only proven their effectiveness against slow moving, single drones, rather than maneuverable swarms. Current methods are not viable for the immediate threat of non-state weaponized drones. If non-state actors are able to present a clear danger to U.S. ships operating in littoral waters, then the U.S. Navy will have to change its decision calculus. The Navy already faces a shortage of ships and a tight budget. On top of this, a viable drone risk to ships operating close enough to a coast could make presence missions unpalatable.


Ensign Richard Kuzma is a Master in Public Policy Candidate at Harvard Kennedy School studying unmanned and autonomous systems. He will join the Surface Fleet after graduation. The views expressed here are not those of the US. Navy or any part of the U.S. government.

Photo: U.S. Navy

We have retired our comments section, but if you want to talk to other members of the natsec community about War on the Rocks articles, the War Hall is the place for you. Check out our membership at!

18 thoughts on “The Navy Littorally Has a Drone Problem

  1. Why limit to the threat to unmanned aircraft though? What about unmanned surface craft and unmanned underwater craft?
    Unmanned underwater craft have been used for years in the commercial sector for years by the oil and gas industry for exploration, by film crews and have been tested by the US Navy for reconnaissance. Small unmanned patrol craft are already in service to protect port facilities. A number of special operations units have demonstrated the use of jet skis for maritime operations, it wouldn’t be a technological stretch to modify a jet ski for remote control and if it can carry 1-2 people with their gear, then it would be able to carry a decent size explosives payload.
    Imagine this scenario, the U.S. Navy is operating in the Mediterranean as we do now, out on standard patrol, one of the cruisers is coming into port and is near an area with moderate commercial/tourist traffic, where it would be normal to see jet skis and small watercraft. A group of jet-skis starts to veer towards the cruiser. They don’t look out of the ordinary (They’re unmanned, but they have mannequins packed with explosives). A dozen of them start to converge on the cruiser, each packed with over 300lbs of high explosives. Even if the crew were able to spot them in time, what could they really do? Small arms fire from the deck, isn’t going to be much of a deterrent and any of the other weapons systems onboard are either anti-aircraft or anti-ship.

  2. It’s time for everyone to cancel their Popular Mechanics subscriptions…or at least stop letting the articles leak onto WoR.

    The “I” in IED is for “improvised” — improvising anything leans heavily on the KISS principle, since improvised complexity is often another way of saying “unreliable”. Moreover, improvised solutions usually don’t break new ground in capabilities — they’re crude copies of more effective weapons, whether that’s land mines, anti-tank warheads, or cruise missiles. That doesn’t make them ineffective, but they’re not as effective as something properly built for the purpose. Moreover, they’re not something that doesn’t exist somewhere else. There’s no Tony Stark hidden in a cave somewhere, building unknown new weapons with scrap metal and chewing gum.

    A booby-trapped UAV is simple. A barrage of small UAVs is fairly simple. A coordinated swarm of small UAVs precisely maneuvering to identify vulnerable points on their target and disable them with warheads requiring precise alignment (as EFPs do) is not…in fact, that’s something we haven’t built yet. So let’s pull the threat back into the land of reality. Serious threats to ships will still be anti-ship missiles, suicide boats, or something else with enough explosives to make a large smoking hole in the side.

    As far as countermeasures, KISS applies as well. Worried about a small UAV with a hand grenade-sized payload? Rig overhead nets…likely more effective than upsetting the locals by spraying gunfire around the ship, or irradiating the crew trying to jam the controls. Rigging anti-torpedo nets when moored used to be SOP…no reason a similar countermeasure wouldn’t work against small boats, jet skis, and submersibles. It’s more effort, but good force protection usually is.

  3. I was inclined to lean with Warlock…
    Is this substantially more dangerous than an RPG?, a Claymore only weighs 3.5 lbs, etc

    Prodrone already sells a copter that totes 44 lbs for 30 minutes. That covers line of sight.
    44 lbs could be substantially lethal without the expertise needed for winged flight.

    Once you go to a winged platform, weight goes up substantially for given power and the skills are in every modeling magazine for the last 50 years.

    Maybe a CIWS in 5.56?

    1. We’re missing the largest threat of all, launch unarmed drone swarms numbering in the dozens or hundreds at a ship. Pepper the swarm waves with small numbers of armed drones. The ship commander has no idea of the non-lethality or lethality of his incoming salvos of identified threat platforms. He is obligated to try to swat every one of them. Exhaust the magazines in the ship and follow-in with waves of heavy hitters on a defenseless ship assuming no non-kinetic options are deployed shipboard.

      “Alone and unafraid” indeed.

      This changes the littoral approach calculus enormously. Combine that with the well-known mechanical neuroses of the Little Crappy Ship or Little Career Slayer (LCS) and it is a recipe for disaster.

      Per drones, we are at the Wright brothers stage of unmanned exploitation in small platforms, we will look back ten years from now and be amazed at the advances. Quads are not the aerodynamic wunderkind that UAVs and helicopters (a combination of prop and rotor demanding sophisticated technology and TTP to remain aloft). Quads are slow but quads are dead simple; they are the airborne equivalent of Chechen RPGs versus Russian armor in Grozny in 1995.

      A cost competition even first world hyper-powers stand to lose.

      But then again global meddling comes at a cost.

    2. You probably could, but hopefully you’d use better references. This report is certainly not a serious engineering or operational assessment. The student team made some rather sloppy assumptions, particularly with respect to the “invisibility” of a rather complex trap, the performance of their “RC UAV”, and ship countertactics. They raised a good point (but not a new one) that air defense radars have trouble with low-speed targets. But take their results at face value, and assume that 2-3 missiles get through. A nuisance for certain, and a very bad day for anyone near the impact point, but even a Harpy missile has only a 70-lb warhead. (Maybe against LCS….) This is a survivable attack. Increasing the number of UAVs simply makes the launch sites easier to detect and interdict.

      There’s a reason we’re not replacing magazines of anti-ship missiles with small drones — the missiles, even considering defenses, are more lethal.

  4. Timely article; the hybrid aspect example at the beginning of the article highlights existing and potential future hybrid adversaries use of available technology in an area that really hasn’t been expanded upon that much: naval warfare.

    AS a discussion driver, the article identified “what may be” if/when technology catches up to the premise; especially given the state of UAS/drone technology specific to command and control aspects (right now it’s one drone, one operator). Enabling a swarm of drones is not as easy as it sounds as noted by a commenter above. The issue of command and control internal to the swarm that has yet to be made into a capability outside some academic institutions (MIT, Naval Postgraduate School as examples) and at least one entertainment organization (Cirque du Soleil) and both had/have a lot of money, very specific environments, and experienced people to do it.

    One commenter noted above that UAS are not the only threat and USVs and UUVs can be as well. True, but then USV and UUV technology investment, capabilities, and usability lag far behind UAS. Not to say they won’t eventually be a threat, there are still in the S&T realm for the most part; untethered versions are in test by the Navy.

    Overall, the issue related to commercial drones isn’t that I can get a swarm of them flying and looking to do damage, it’s the issue of available spectrum. Right now, the bulk of commercial drones tend to self-jam because they all operate in the same frequency band (think, very similar and in smart phone bands). A classic example occur during the 2015 World Drone Day in Las Vegas when a record setting 100 drones were to have taken off yet only about 5 got up because they all jammed each other. Yes, technology will eventually address this, but the frequency spectrum is not something you can “grow” because it is what it is and you can’t make more. Therefore, launching dozens or hundreds at a ship are not currently possible and a large-scale swarm is very visible since such a swarm would have to be launched in small groups or enmass (providing a nice radar return event with composite construction0, a lot of noise, and your launch location is nicely identified to target.

    The larger issue is detection, tracking, and identification of these things. The quads have a significant restraint in flight time based on available battery power and what the quad is carrying. Think 20 minutes on a windless, non-humid day that includes launch and recovery (granted, a one-way mission would provide more flight time). Fixed wings have more time if they don’t use batteries and acoustic sensors can pick up battery and gas operated UAS because of the decibel out puts of the blades/propellers.

    Countermeasures in existence that work for detection and counter-UAS are in the EW arena and rightly so; interrupt the C2 signal and many of the drones simply fly back to their point of origin. Other sensors on the open market can identify where the UAS started from and who is operating it. EW isn’t the only defense as Warlock notes above and innovation isn’t just the area of the hybrid adversary–American troops are pretty good at that as well.

    AS for the 2012 paper, its a good qualitative and quantitative reference if somewhat dated. Since then there has been a myriad of efforts across DOD, academia, and other nations to address swarm capability and defense. I just came of a test developing a Concept of Employment for an offensive swarm armed with electronic attack payloads and, prior to that, development of TTP for detect, track, and ID of UAS related to how counter-UAS aspects apply to informing the HQs responsible for area air defense can streamline their engagement process–after all, detection is hard but getting less hard, its the tracking and positive identification that the UAS or swarm of UAS is a threat. We all know what happens when positive ID is not acquired.

    Does this change the littoral approach calculus? Perhaps and that requires further review while defenses are in development. Luckily, detection and defending technologies for counter-UAS are a growth industry and everyone these days seems to have a counter-UAS organization.

    1. The Navy’s LOCUST program shows that drone swarms are already viable weapons.

      And you can forget about jamming them. Optical communications and other techniques (especially autonomy) will mean there is nothing to jam.

      I have yet to see anything that looks like a viable defence against a swarm of drones.

      1. A counter-drone swarm could be just as deadly to the small cheap drones, and cost equally as little to field.

        Small cheap drones must necessarily be either non-threatening (small or ineffective payloads that at most are a nuisance to a modern warship), be slow and therefore easy to shoot down, and have short range. Once you start building bigger, more lethal payloads, add speed (as in supersonic) and have long range, you no longer have “cheap drone swarms”.

        There is nothing particularly challenging in taking out small, slow swarms of drones long before they get close to warships. Already most of our warships have both manned attack helicopters as well as unmanned drone choppers to send out. Drone swarms are going to be visible on radar. And once an adversary sends out a single drone swarm, they just gave away the launch points which are then subject to counterattack.

        And of course drone swarms are tailor-made for laser weapons, which are about to enter the fleet. A single laser designator can detect and target literally thousands of drones in less time than it takes to blink an eye. The same exact principle is used in LIDAR, which uses a single laser range finder to map out millions of physical surface data points in a blink of an eye. This technology has been in use for mapping and remost sensing for decades – if you don’t believe DARPA isn’t already far ahead of this, you’re not thinking rationally.

        This is just the same old battle of offense vs. defense that will never end. Weapons beget counter-weapons. We’ll win that battle every time.

  5. The issue generalizes, nor is it exclusively an airborne threat. The last three yeas I have been a mentor for the local high school First Robotics organization and the commercial capability out there that is dirt cheap – at least by military standards – is simply eye watering. A raspberry pi computer, more capable than most computers that run he weapons that are actually in the field today is – wait for it – $35. A GPS module capable of providing more accuracy to that computer than we had in a Block 30 F-16 INS, is $30. A nine df accelerometer, magnetometer, gyro chip is $ 100, and these are for the brand name units that have been tested. Cheap Chinese knock-offs are available much cheaper, as little as 10% of the ‘brand name’ units, albeit 20% of them will be defective and will have to be discarded.

    The most complicated technology of WWII, the naval torpedo, could be put together by most high school robotics clubs today (less the explosive warhead), complete with terminal homing, for only a few thousand bucks worth of batteries and parts. I am not exaggerating.

  6. The cheapest interim solution is skeet shooting by the whole watch. A little practice will make an amatuer adequate to the job. Today commercial drones are quite fragile.

    That will buy you time for coupling your EW and acoustics (and visual?) to lighter fast guns. But magazines will need to be deep.

    Programable autopilots are still relatively expensive, but cut the EW cord to the operator. A level of sophistication is required unless the strike area is preplanned (there is that littoral challenge again, maneuver), as well as transfer to onboard sensors for final course.

    All very doable, and very counterable today.

    Bring in the sophistication of a state actor and this all gets much more challenging.

    Or go to winged craft, but then we are getting into sizes that might be tracked and shot, though again, throwing 30mm at these is expensive.

    1. Skeet shooting is not viable.

      A small lethal drone like Switchblade comes in at over 90mph and has a target area the size of a baseball. Anyone out skeet shooting is simply going to be an easy target.

      1. Or for five grand get something that will take off and land at the push of a button and have FLIR capability out to twelve kilometers while carrying a 2kg payload. Something that could fly in at night, making skeet shooting REAL problematic, and then attack a soft target like an aircraft or helicopter whose fuel and ammo could add secondary explosions to the mix.

        No, those pooh-poohing the threat are whistling past the graveyard. Moderate cost drones (defined roughly as costing the same as a used Toyota Hi-Lux pickup mounting a 12.5 mm machine gun, are EXTREMELY capable and keeping them from being exported (or God forbid used domestically against military and civilian airfields and similar inherently soft targets) will not be easy.

        1. Neither of you are making a good case case. David’s January article on the LOCUST project makes assumptions on air vehicle performance, controllability and tracking at low altitudes, and lethality that don’t hold up. While perhaps not intentionally, his comments imply that these smaller platforms retain the performance of their larger brethren — that a LOCUST, based on a UAV dropped through a dropsonde port — has similar performance to an IAI Harpy (a 300-lb, 110-kt platform for which the speed increases with every article I read!) It’s just not true. Small size limits seeker size (detection range and accuracy), it limits the powerplant (platform speed and range), it limits the size of aerodynamic surfaces (controllability and flight path accuracy), and it limits warhead size (lethality). The picture of a cloud of small UAVs moving precisely to strike vulnerable antennas, weapons sensors, and exposed controls on a ship just doesn’t hold up for a ship at rest or steaming slowly, let alone for a warship maneuvering and employing countermeasures (chaff rockets may prove to be as much physical countermeasure to small UAVs as they would disrupting targeting). Throwing a cloud of hand grenades — or even small bombs — at a small warship will cause some damage, but would require lots of luck to cripple it, let alone destroy it. Those issues can be fixed, but pretty soon you’ve built yourself an anti-ship missile, which is likely to be bigger and more expensive.

          The same thing applies spending $5k to essentially do what a barrage of cheap rockets or decently-trained mortar crew could do just as well. And again, assuming not even the simplest passive protective measures, like revetments with overhead nets.

          It’s all certainly possible — very possible. What it’s not, in comparison with currently available alternatives, is cheap or easy. And the assumption that an autonomous or remotely-guided/piloted vehicle is immune from limitations that apply to other weapons just doesn’t fly (or swim, or crawl). And that’s not just whistling. :)

          1. Though in a sense, the only target that matters is a plane on a carrier flight deck. If 10,000 small drones can score a few hit on that, they will pay for themselves ten times over. More likely they would get several, and bring all flight operations to a halt. Killing any exposed personnel would also be brutally cost-effective.

            And in the process, they will exhaust more than their own value in defensive munitions.