A remote-operated aircraft flown into a target while the operator watches not only isn’t something new, but it’s already been done in volume and found wanting. An apparent attempt to assassinate the President of Venezuela by such means last month didn’t work.
Some guy scraped up after an attempted murder-by-drone.
A drone that can fly at 200mph for three kilometers, operated over an ECM-hardened control link, and delivering a 5+ lb armor-piercing explosive warhead has existed for ovr fifty years now. The AT-3 Sagger came, made its impact, and we all moved on. Visually flying your remote-controlled aircraft into a target is called Manual Command Line of Sight, and it’s not used for anything important (more discussion here). Small autonomous systems will change the battlefield, but flying a much smaller payload onto a target over a less secure control channel than primitive missiles is not going to be how it happens. Hit probability and kill probability are low, with high vulnerability to electronic countermeasures. Taping a grenade to your kid’s toy helicopter won’t going to give you the edge you need to win on tomorrow’s battlefield. The characteristics of such a line-of-sight weapon compare poorly to the currently dominant line of sight personal weapon, the rifle.
Anyway, what will change? The important factors are availability & control.
Availability
Small unmanned systems will be extremely “available”, which is to say they’ll be relatively cheap and there will be a lot of them. Much of the anxiety over drone warfare results from the not always well-articulated concern that they will also be available to new categories of users compared to previous novel or analogous weapons. The guided missiles mentioned above are also relatively cheap, especially compared to their intended targets. However, they are not cheap to, say, the person reading this, even disregarding the risk of criminal liability.
Typical merchant of guided missiles
As we’ll see below, militarily-useful UAS will be deceptively complex. Design and manufacture will require sufficient economies of scale to prevent clandestine production or use of revolutionary weapons in well-policed territories. Again, the illuminating comparison to existing weapons. A drone analogue to the rifle (line of sight guided onto target) and the mortar (man portable, attacks from above at increased range: e.g. the IRA and Syrian rebels for improvised examples) might be within reach of an individual or small organization. But we already live in such a world. These weapons did (past tense, it’s already happened) increase the amount of destruction one person or a small group can inflict. But their use hinges on social and political factors — your local police department would probably be in deep trouble against two hostile squads of riflemen, but the threat just isn’t there.
At most, the availability of combat drones may provide some marginal improvements to civilian armament, but this effect will be different in degree and kind to military equipment. Another analogy, to night vision: the night and thermal optics available to the average citizen do provide some new capability, mostly relevant to a small group of enthusiasts. But this pales in comparison to the ubiquity and quality of the devices in actual fighting units.
Anyway, the most visible military benefit of these things will be that there will be a lot of them, and they’ll be relatively expendable — more like munitions than end items. The high availability of small UAS will make humans relatively more valuable, since UAS will functionally extend the capability of soldiers and important decisions will continue to be made by humans (indeed, the ratio of important to unimportant decisions will likely increase as trivial decisions are automated). Compare the military value of an individual soldier or small unit in a pre-industrial, early 20th century, and modern army — small UAS will continue this trend. But now it’s time to talk about control and coordination.
Control Systems
The small UAS revolution will depend mostly on improvements to the control systems of these devices. This will include not just making human interfaces more intuitive, but implementing relatively complex behaviors — such as swarming — in a manner transparent to a single or small number of human operators.
Consider a soldier now operating an RQ-11 Raven, a small fixed-wing reconnaissance drone. He operates the Raven over a two-way low-bandwidth line of sight command channel, and the Raven sends back high-bandwidth video to the operator. What’s going to set apart future systems from the current state of the art will be the ability to integrate multiple systems transparently.
For instance, say we want to really exploit the availability of SUAS to the hilt and have multiple vehicles per operator — effectively increasing his (unit’s) field of view. Multiple real-time video feeds may be fine for a headquarters, but probably not at the suqad or platoon. At some point between sensor data and interface output, then, discrimination must occur (preferably with a manual override) regarding what the operator needs to see.
Another example: the Switchblade is a small UAS. It has an explosive payload about equivalent to a 40mm grenade round, and is flown onto its target by an operator over a low-bandwidth command signal based on high-bandwidth video from the munition. (This is a substantial improvement over MCLOS because the operator only needs to consider the position of the target, not both the target and the munition.)
This device most closely resembles hobbyist kamikaze drones, but exceeds them in image resolution and response to control input. Still, something like this could be within reach of a lone engineer. The cost, size, and lethality of this weapon compare poorly to an autoloading rifle and a backpack full of loaded magazines, the chief advantage being standoff. For attacking a single, positively-identified target this works well, but for mass killing it does not. I will examine this problem more in another post.
Again, while thinking about what new capabilities small UAS will bring to the battlefield, we need to think about how to avoid overloading the operator while increasing his capability. I’m going to make some illustrations of what I’m talking about to make this clearer. These graphics are somewhat simplified. Line thickness corresponds to signal bandwidth; green is imagery, purple are command links, and red are munition paths.
A single operator using a Switchblade-type kamikaze munition.
How a small unit coordinates organic UAS using current technology.
Future small UAS. Increased automation including operator-transparent cross-platform coordination, filtered imagery throughput, fewer human operators.
Target prioritization to maximize disruption given supply constraints.
The need to harden all these control signals against electronic warfare via encryption, antenna nulls, etc. will further remove military systems from their civilian counterparts.
Airspace Control
The inherent usefulness of flight in maintaining lines of sight and freedom of movement means most small robots will fly, although not all of them; ground drones will be especially common in urban areas. The usefulness of flight and the ubiquity of small UAS will effectively increase the height of ground units as battlefield features. This may decrease the usefulness of helicopters, who will have to choose between flying high and exposing themselves to anti-air weapons or flying low and risking collision. Airspace deconfliction of the one SUAS per platoon or company currently deployed already fails routinely; this problem will get far worse. If the ability to establish, modify, and propagate airspace Restricted Operating Zones is too difficult, units will have to establish standard zones assumed to accompany their reported positions at all times.
The enemy will be doing the same thing under similar constraints, meaning fights for local air superiority will become a problem for ground tactical units.
Conclusion
Although small unmanned aerial systems might provide some marginal novel capability to non-state actors, the complexity and specialization of military-use SUAS will separate these devices from their civilian counterparts. Conceptualizations of the future use of these devices as weapons and enablers should look past preliminary, unsophisticated examples and focus on control, coordination, and availability in order to identify new capabilities rather than remain stuck in the framework of current guided munitions and remote surveillance platforms.
Inky
It’s curious to watch what’s happening. In many ways this development mirrors the emergence of aircraft as a serious force on the battlefield, and likewise, defensive capability lags behind offensive.
And while this is the case, the civil UAV models will continue to pose significant threat. The problem with them is that within the limits of the conventional AA defense, UAV’s would quickly saturate and overwhelm the system. Conventional AA is not prepared to deal with threats that might have the same cost as one missile. More than that, neither radars that can reliably detect small to mircro UAV’s, nor missiles to hit them don’t exist and even if they did they would probably be saturated as well. So… AoE? Some sort of drone disablement system, most likely jamming. I think something like that is [already developed](http://www.thedrive.com/aerial/11505/the-7-most-significant-anti-drone-weapons) however their effectiveness is quite limited by necessity of line-of-sight, inability to engage more that one target at once, and, most importantly, reliance on MK1 eyeball for detection. And they are all supposed to work against unmodified factory drones, that will land when asked in a stern voice. Other systems include lazorz, that are quite expensive and most definitely not man-portable. EMP generators are pretty expensive and inherently prone to blue on blue.
So far looks like sword:shield – 1:0.
P.S. [Oldie but goodie](https://youtu.be/HipTO_7mUOw) but slightly off-topic because not a military one. Whaddya say?
sfoil
It’s very important to think about exactly what capabilities (would) make UAVs more dangerous than mortar rounds, rifle bullets, and ATGMs. Artillery radars and point-defense systems can definitely detect incoming 60mm mortar rounds, which are baseball-sized. Once such a round is in flight only a kinetic kill will stop it from hitting whatever it’s falling towards. If my UAV is the same size as a mortar round but won’t hit its target unless it can maintain a high-bandwidth, easily-characterizable video link back to an operator all the way in, I don’t even need to shoot it down. I can jam the command link (or I make the adversary pump tons of power into transmitting commands in order to break through jamming, which makes him easier to find). If existing defense systems can point a Vulcan cannon at a mortar round (and they can) then they’re going to have an even easier time pointing a beam at it — instead of power, ammo feeds, recoil-absorbing mechanisms, safety no-fire zones I just need power and an antenna on a gimbal. AFAIK, these factors were behind the failure of the only known drone-swarm attack: https://www.cnbc.com/2018/01/11/swarm-of-armed-diy-drones-attacks-russian-military-base-in-syria.html
Also, I can keep hanging mortar rounds while the first one is in flight, whereas if I’m flying a basic drone into the target I have to have an operator for every drone in the air — now it’s actually harder to saturate defenses. This is why control systems are so important: in order to achieve saturation like you’re talking about, I need to be able to launch a bunch of drones and either have them know exactly where they’re going (hard, and it gets harder the smaller the drone’s payload is) or know when they need to hand themselves off for manual guidance or somehow have one guy flying a lot of these things at once if they aren’t completely autonomous.
You can also imagine synergies. Instead of trying to saturate a defense by flying drones into a target, fly one drone overhead and have it spot (possibly with a laser target designator) the mortar rounds you launch. This is something that can be done right now, and in fact is done by sufficiently non-skittish commanders.
The slaughterbots video: what’s the range and loiter time on those things? Probably not long unless there’s a miniature nuclear reactor in there. How far away does it need to be from a target before obtaining positive ID (its sensor aperture doesn’t look any bigger than your cell phone camera). Both of these things suggest you need to be pretty close to the target — have actionable intel — to fire. Are they easier or harder to manufacture than nerve gas? Ammonium nitrate? If they’re harder, then they’re not what you want for mass killing. The real lure of the video is the phantasm of flawless surgical strikes. I will say that whoever made it is thinking straight about one thing: there’s no command link to jam, it’s fully autonomous — although there are four induction motors operating on a standard frequency and geometry. Will this increase the importance of electronic warfare? Sure (it’s a growth industry). The little guy might not have that, but you can already walk up to the little guy in the grocery store and shoot him. Is dumping a planeload or driving a truckload of something like this onto a target more effective than an equivalent weight in bombs? Maybe, a little.
sfoil
I’d also like to point out something about the relationship to aircraft: as soon as the first guy went up in a balloon, he realized he had something really special because he could see everything. An in fact, aerial recon changed the character of war very quickly. The first pilots also intuitively felt like they could, finally, just fly over the enemy’s army to his capitol and hurl bombs at their defenseless target, winning instantly. Outrageous claims were made — there is incontrovertible evidence that fears of mutual destruction predate Hiroshima by decades. It didn’t work like that, and a hundred years later we’re still hearing them go on that this time, they really mean it. Yet despite their weapons being more powerful, ever more precise, and their enemies more helpless than ever — it hasn’t come to pass. I’m a skeptic, obviously.
Inky
Oh, and enable Markdown in comments, please?