Drone Swarms as Air Defence Systems

C8XEBY Starlings Sturnus vulgarus arriving at night time roost near Gretna Scotland December

In a major instance of failure of air defence systems, the Saudi Arabia failed to protect its Abqaiq and Khurais oil facilities when Houthi rebels launched 18 drones and seven cruise missiles against the two facilities on 14 September 2019. While the drones struck Abqaiq, the  cruise missiles fell short and did not hit the facility though four cruise missiles did strike the installations at Khurais.

 The air defences deployed to protect the Abqaiq facility’s air defences reportedly included the American-made Patriot system, Oerlikon GDF 35mm cannons equipped with the Skyguard radar and Shahine, a version of France’s Crotale missile system. Various reasons for failure of the air defences have been offered – ranging from ‘the best air defences systems also fail’ by the US Defence Secretary to the speed and angle of the drones and missiles that prevented the air defences from engaging them. What was clear was that the Saudi air defences did not engage the drones and failed in their mission1.

There have been previous instances of drone swarms being used for offensive missions like the repeat attacks on the Russian base in Khmeimim, Syria. A mix of anti-aircraft guns, missiles and electronic warfare systems were used by the Russians to either shot down or neutralise the drones underlining the basic fact that effective  air defence systems are not based on one weapon capable of stopping everything2. An air defence system has diverse, layered and integrated air and missile defence systems to counter varied threats. Unless a layered, interconnected system is fielded, there will be gaps in the air defences that can be exploited by an adversary. One of the more effective means of exploiting these ‘gaps’ is the use of drones – and drone swarms.

Drone are one of the most rapidly proliferating technologies which have found varied military applications. Use of drones for armed strikes and reconnaissance have been a routine practice now but what is now getting increasing attention is use of drone swarms  i.e. “multiple unmanned platforms and/or weapons deployed to accomplish a shared objective, with the platforms and/or weapons autonomously altering their behaviour based on communication with one another” for military missions. The cases mentioned above are just two of the instances when drone swarms have been used in recent times to carry out offensive missions. The use of drones (and drone swarms) has been subject of many a study with most studies focussed on their use in offensive missions and methods to counter them, but like most technological advancements it needs to be understood is that drones, and drone swarms, can be used in defensive missions as well. 

At the rate of technical advancement, the  “quadcopter” drones and their successors are likely to have the capability to fly autonomously—at much higher altitudes, with longer flights—and be capable of complex formation manoeuvres by 2025, if not sooner. This opens up the possibility of the use of drone swarms in defensive missions as a large number of expendable, and cheap, systems operating in sync can be used very effectively to deny the freedom of action to an adversary. 

The use of swarms has been categorised as the fourth doctrinal form of  conflict by John Arquilla and David Ronfeldt inSwarming and the Future of Conflict3. After Melee, Massing and Manoeuvre, Arquilla and Ronfeldt believe that the future conflict will see the use of large numbers of dispersed individuals or small groups coordinating together and fighting as a coherent whole and this will the age of Swarming. 

Using a large number of individual entities, swarm warfare will combine the highly decentralized nature of melee combat with the mobility of manoeuvre but with a higher degree of organization and cohesion, allowing a large number of individual elements to fight collectively. While a number of these entities be controlled centrally, some will be operating independently with a large number operating autonomously. Artificial Intelligence (AI) will be central to this form of warfare and  it will have a higher organizational and communication requirements than manoeuvre warfare, since the number of simultaneously manoeuvring and fighting individual elements will be significantly larger. 

In this form of warfare, unmanned systems will play an important role,  replacing the  manned systems and carrying out the full spectrum of combat missions.  The use of low-cost uninhabited systems will enable a large numbers of forces to be fielded with the AI and information technology allowing them to fight in a coordinated manner. The systems, even when dispersed over large distance, will fight together in a coherent and coordinated manner due to networking and automation. The ‘forces’ will have the ability to be dispersed and yet retain the capability to converge at the desired point(s) at the given time. This makes the force(s) difficult to be countered as they carry out their missions against the adversary. Defensive missions will obviously be integral to such use of drones and drone swarms.

U S Navy has already demonstrated the ability of a swarm of autonomous uninhabited surface vessels to intercept and surround an unknown and potentially hostile vessel and it is not difficult to use the same methodology and employ drone swarm to counter hostile airborne threats4

Studies and trials are already being carried out to use a swarm to take out a hostile swarm. Even keeping in mind the economics of this approach, as long as the counter- swarm is cheaper and/or more effective than the enemy swarm, it would be relatively a low-cost way to defend against enemy swarm attacks. Work on this was carried out by the United States Naval Postgraduate School in 2012 wherein ‘swarm-on-swarm’ warfare tactics was tested with a 50-on-50 aerial swarm fight.5 The key to developing such a ‘counter-swarm’ is having the best algorithms to enable better coordination and faster reaction times, rather than simply the best platforms. 

Going further, the same approach can be used to counter single/multiple threats by drones swarming and ‘closing’ the skies. A swarm of thousand, or thousands of simple drones networked to operate in sync can be used to form a dome over a high-value target. The drones, once launched, would position themselves at pre-determined positions, making an impenetrable lattice of inter-connected drone. Any incoming missile, regardless of its speed or manoeuvrability, would be intercepted by this swarm dome, with the hostile missile hitting  multiple drones. This would act like the balloon barrages using thick, impenetrable wires fixed to the balloons to “area deny” low-level flying aircraft in its simple form. The same drones could also serve effectively as air mines, colliding with or exploding in the vicinity of incoming bombers. To form such a dome, simpler, less-advanced drones can be used as all they need to do after launch is to position themselves at pre-determined positions.

Using their inherent mobility and manoeuvrability, the swarm can also be launched as and when required to intercept any hostile incoming aerial target, akin to aimed fire anti-aircraft fire. This would be like aimed fire by an AA gun or missile system. An upscale method  can be by using intelligent swarms with ‘seek and destroy’ capability. The swarm can either position itself at a desired position ( and shift to new location if required), wait for aircraft to collide with them or can home on  to an  aircraft in a kamikaze-like kill. Algorithms already exist that can programme  a drone with “see-and-avoid” ability. This has been demonstrated  at the Massachusetts Institute of Technology (MIT)  with proven autonomous software logic. In the MIT study, a an open-source stereovision algorithm was used to enable a “drone to detect objects and build a full map of its surroundings in real time . . . at 120 frames per second.” Extending the logic, the algorithm can be reversed to detect objects in space and not avoid them, thus resulting in an impact6

Drone swarms will have a significantly longer loiter time. Better and longer sustaining batteries coupled with more efficient aerodynamics and lighter components will allow the swarms to operate over longer ranges. The ‘swarm’ that attacked the Russian base in was launched from the village of Muwazarra in southern Idlib province, about 80 kilometres away. The swarm was made up of crude fixed-wing drone resembling an oversize toy aircraft and appeared to have been cheaply assembled from a type of plywood. With better drones, the swarms will be able to strike at much longer ranges or loiter and remain effective as a shield over the designated area for longer durations. 

How much of a defence will such a shield provide is a valid question to be asked as the efficacy of the ‘air defence’ provided by the swarm will depend on the degree of damage this dome or shield can cause. The basic question that whether an impact with the drone(s) will cause any significant damage to the hostile airborne object- aircraft or a missile. The common assumption is of the damage caused something akin to the damage caused by a bird hit but it may not be a correct analogy as recent incidents have shown. 

Firstly, the birds instinctively attempt a last-ditch manoeuvre to avoid approaching aircraft but it will not be so in case with the swarm as the drones would be programmed not to avoid the aircraft but to have an impact. This considerably raises the probability  of an impact as compared to the possibility of a bird hit. Secondly, with a dense swarm forming the dome, the hostile aircraft will have an impact with more than one drone thus increasing the degree of damage that would be caused. 

The impact of drones with an aircraft or a ballistic/ cruise missile is as yet thought to be of marginal import with minimal damage, not something dissimilar to a bird hit, but recent incidents have shown that such presupposition may not be correct. As per one research, the birds “behave like fluids” at impact, with “the disintegration and the flowing of the bird absorb[ing] energy, which decreases the impact forces.”7   Due to this, the damage caused is not always significant. On the other hand, drones are different and are a  “non-deformable impactor . . . creates a localized strain field in the target material with high peak forces, which supports . . . material failure”,8 thus causing greater damage in case of a hit. The  material composition of the drone and the potential for higher relative airspeed of impact also contribute to this. The case of a r RQ-7 impact with a C-130 in Afghanistan resulting in a ruptured fuel tank with damage to a wing spar and the wing box is just one example9.

The use of drones in a swarm also contributes to higher hit probability. 

The use of swarm to form a dome or a shield is not the only method to counter the hostile aircraft. They can be used in vicinity of adversary’s air bases to prevent the use of runways. The simplest method would be to attack the runway and other infrastructure to deny its use but a more innovative method is to deploy the swarm near the airbase  to form a net or a shield at either ends of the runway to prevent take off and landings. The swarm can remain dispersed in the area near the base and converge only when the base is being used. This would an area denial mission and be an offensive use of the swarm but as Churchill said, “the great defence against aerial menace is to attack the enemy’s aircraft as near as possible to their point of departure.”10

The other uses in air defence role could be the employment of  swarms of uninhabited air vehicles to provide early detection of any threat including from enemy swarms. Vasile Sandru and Marius Radulescu of Henri Coanda Air Force Academy, Brasov, Romania in their paper The Use of UAVs during actions of Integrated Air Defence Systems have discussed the use and integration of UAVs in the surveillance network of an Integrated Air Defence System (IADS)11. Some applications can be of use in areas with poor radar coverage due to terrain considerations.

Swarms are relatively a new field of study in our context but is important that while we discuss their potential use, the possibilities of their use in air defence mission are also studied and exploited.


1. Michael Safi and Julian Borger, How did oil attack breach Saudi defences and what will happen next?, The Guardian, 19 Sep 2019 accessed at https://www.theguardian.com/world/2019/sep/19/how-did-attack-breach-saudi-defences-and-what-will-happen-next and Kareem Fahim and Kareem Fahim, Saudi Arabia oil output takes major hit after apparent drone attacks claimed by Yemen rebels, Washington Post, 15 September 2019, accessed at https://www.washingtonpost.com/world/drone-attacks-on-saudi-oil-facilities-spark-explosions-and-fires/2019/09/14/b6fab6d0-d6b9-11e9-ab26-e6dbebac45d3_story.html

2. David Reed, A swarm of armed drones attacked a Russian military base in Syria, CNBC, 11 January 2018 accessed at https://www.cnbc.com/2018/01/11/swarm-of-armed-diy-drones-attacks-russian-military-base-in-syria.html

3. Arquilla, John and David Ronfeldt, Swarming and the Future of Conflict, RAND Corporation, Santa Monica, California, 2000, https://www.rand.org/pubs/documented_briefings/DB311.html. 4. Leslie F. Hauck and Dr. John P. Geis II, Air Mines : Countering the Drone Threat to Aircraft, Air & Space Power Journal , Spring 2017 pp 26-40

5. Paul Scharre, Robotics on the Battlefield Part II: The Coming Swarm, Center for a New American Security, October 2014, https://s3.amazonaws.com/files.cnas. org/documents/CNAS_TheComingSwarm_Scharre.pdf?mtime=201609060820596

6. Mary Grady, MIT Drone Avoids Obstacles Autonomously, AV Web, 4 November 2015, http:// www.avweb.com/avwebflash/news/MIT-Drone-Avoids-Obstacles-Autonomously-225143-1.html

7. Alexander Radi, Potential Damage Assessment of a Mid-Air Collision with a Small UAV, Monash University, Australia: Civil Aviation Safety Authority, 6 December 2013

8. ibid.

9. ibid.

10.  Joint Publication 3-01, Countering Air and Missile Threats, 23 March 2012, IV-1, http://www.dtic .mil/doctrine/new_pubs/jp3_01.pdf.

11. Vasile Sandru and Marius Radulescu, The Use of UAVs during actions of Integrated Air Defence Systems, Review of Air Force Academy, No 3(30), Henri Coanda Air Force Academy, Brasov, Romania, 2015