Swarming Terror

Swarming Terror

Zachary Kallenborn, Gary Ackerman, and Philipp C. Bleek

On 25 April 2022, the White House released a new  National Action Plan for countering domestic-unmanned aircraft systems.[1] As the White House implements the plan, and works with Congress to secure new authorities like the new bipartisan Drone Act of 2022, policy-makers need to consider how drone terrorist threats scale and evolve over time.[2]

Drone Swarm. Source: Marco Verch (CC BY 2.0)

Multi-Drone Swarms

Multi-drone terrorism represents an emerging terrorism threat, with a range of potential consequences including, at the high end, mass casualties. Although terrorists could quite easily acquire numerous drones, they face considerable challenges in obtaining and deploying the technology to control multiple drones at once. This is especially true for drone swarms in which multiple drones are integrated into a single weapon platform with inter-drone communication. The real difficulties involved with mounting a truly massive drone attack means that policy-makers must plan for a broad range of threats, and carefully balance the costs of defense systems against risks posed to particular targets.[3]

In a May 2017 speech, General Raymond Thomas (retired), former Commander of Special Operations Command, described how, during the 2016 Battle of Mosul, “At one point there were 12 ‘killer bees,’ if you will, right overhead and underneath our air superiority.”[4] These enemy “killer bees” were simple commercial quadcopters that ISIS had outfitted with explosives. The event portends a larger trend: terrorists using not only drones, but multiple drones at once, to collect intelligence, record propaganda, and carry out attacks.

Amplifying Effects

Drones could create harm through attacking vulnerable critical infrastructure facilities The 2019 Abqaiq-Khurais drone attack in Saudi Arabia caused the country to suspend over half the nation’s daily oil output, which was not fully restored until 31 October.[5][6] (The Iranian-sponsored non-state Houthi rebels took responsibility for the attack, though later evidence suggests Iran was behind the attack.) Multiple drones could even cause mass casualties. Imagine a group of drones flying above a concert venue, dropping bombs over participants. the drones could spray traditional chemical weapons agents or non-traditional but still quite harmful agents like opioids.[7]

Multi-drone attacks amplify the harm terrorists can cause with a single drone. At the most basic level, five drones can cause five times the damage of a single drone, all else being equal. But quantity has a quality all its own. Current counter-drone systems are poorly equipped to handle multiple drones at once.[8] Drone detectors may only be able to follow a single drone. Counter-drone systems like net guns or trained birds are far less effective against a cluster of drones, because they cannot defeat so many drones quickly enough. That’s a huge challenge in certain types of attack, like an attempted assassination: if the terrorist aims to kill a world leader, only a single drone needs to hit its mark.

Improvements in global positioning system (GPS) accuracy enable smart targeting, allowing drone guidance with accuracy within less than a meter. Employing facial recognition software on drones could conceivably give rise to what we term “mass discriminatory terrorism”: a terrorist attack against a large number of specific individuals with minimal extraneous casualties. Imagine a terror attack on Congress: a swarm of drones fly about, targeting only the terrorists’ political opponents.

Challenges to Terrorist Deployment

Acquiring a whole lot of drones is relatively easy: a terrorist can go on Amazon and order 50. But having 50 drones does not mean having 50 hands to fly them, especially given that most terrorist operational cells consist of only a handful of attackers. As an operator flies more drones, the difficulty grows quickly. The operator needs to ensure the drones do not collide, do not fly into a building, and achieve their overall objective. Missy Cummings and Paul Mitchell estimate that if a pilot attempts to fly more than about four drones flown simultaneously, the ability to achieve mission objectives plummets.[9]

Terrorists hoping to fly 50 drones have a couple of options: recruit or train more drone pilots or procure the technology to integrate multiple drones. Large, well-resourced terror groups may be able to train numerous drone pilots; but even then coordinating a dozen or more pilots presents a logistical hurdle. This would be a challenge for small groups and all but impossible for a lone wolf or small cell.  Integrating multiple drones can be achieved by having the drones communicate with each other, for example, sharing data regarding obstacles and maintaining a safe distance from other drones in the cluster. This would enable a single pilot to fly many more drones than would otherwise be possible. Then there are drone swarms that are fully autonomous and can guide themselves to preprogramed targets, avoiding obstacles and coordinating their flight patterns, without the need for a pilot. While possible—and indeed states have successfully demonstrated large drone swarms—these kinds of integrated multi-drone systems require capabilities that extend beyond those required for traditional drones, including sensor fusion and complex swarming algorithms.

In deciding to pursue multi-drone systems, terrorists will weigh the advantages against these challenges, and other limitations like drone payloads. Affordable commercial drones also tend to have limited payloads, in the range of a few pounds at most. That limits how much harm the drones can do: fewer explosives, smaller weapons, or smaller amounts of chemical or biological agent. Terrorists are likely to weigh this trade-off against alternative delivery systems. Why not just load a huge nitrogen fertilizer bomb into the back of a rental truck, as the Oklahoma City bombers did?

Policy Implications

So, terrorists could use multiple drones to cause harm, perhaps even mass harm, and defending against them is hard. What should policy makers do about it?

Homeland security, law enforcement, and national security agencies should develop indicators of terrorist group interest in multi-drone capabilities. Warning signs could include mass purchase of drones, interest in software intended to manage multiple drones at once, or experimenting with drones generally. This would likely require working with and building awareness among commercial producers and distributors of drone systems, so that those companies know when to report suspicious activity.

Policymakers also need to improve defenses for high-risk targets, like airports, critical infrastructure assets, and heads of state. A major component of this is technological. States need to build better counter-drone defenses that can detect and defeat multi-drone attacks. But the policy, legal, and strategic dimensions shouldn’t be neglected. In the United States, only certain federal government authorities can operate counter-drone systems, because systems like jammers can also have major negative effects in jamming otherwise good signals. This can be a big challenge when protecting private sector facilities like critical infrastructure. If the United States keeps the law in place, it should consider strategies that accommodate these requirements, such as building a national counter-drone network of remotely operated or autonomous systems.

Terrorists are increasingly use drones to take to the sky. Terrorists may increasingly use multiple drones to cause even greater harm. A plague of robotic locusts may be coming…

For a longer, more nuanced version of this argument, see the authors’ new study in Terrorism and Poliitical Violence,”A Plague of Locusts? A Preliminary Assessment of the Threat of Multi-Drone Terrorism.”

Endnotes

[1] “FACT SHEET: The Domestic Counter-Unmanned Aircraft Systems National Action Plan.” The White House. Washington, DC. 25 April 2022, https://mail.google.com/mail/u/0/#inbox/FMfcgzGpGnDfsbnvhPtqqNFKJzBzCrJH.

[2] S.3542 – Drone Act of 2022. 117th Congress (2021-2022), https://www.congress.gov/bill/117th-congress/senate-bill/3542/text.

[3] Zachary Kallenborn, Gary Ackerman, and Philipp C. Bleek, “A Plague of Locusts? A Preliminary Assessment of the Threat of Multi-Drone Terrorism.” Terrorism and Political Violence. 20 May 2022, https://www.tandfonline.com/doi/full/10.1080/09546553.2022.2061960.

[4] Cited in David B. Larter, “SOCOM commander: Armed ISIS drones were 2016’s ‘most daunting problem.’” Defense News. 19 May 2017, https://www.defensenews.com/digital-show-dailies/sofic/2017/05/16/socom-commander-armed-isis-drones-were-2016s-most-daunting-problem/.

[5] Ben Hubbard, Palko Karasz and Stanley Reed, “Two Major Saudi Oil Installations Hit by Drone Strike, and U.S. Blames Iran.” New York Times. 14 September 2019 (updated 15 September 2019), https://www.nytimes.com/2019/09/14/world/middleeast/saudi-arabia-refineries-drone-attack.html.

[6] Humeyra Pamuk, “Exclusive: U.S. probe of Saudi oil attack shows it came from north – report,” Reuters. 19 September 2019, https://www.reuters.com/article/us-saudi-aramco-attacks-iran-exclusive/exclusive-u-s-probe-of-saudi-oil-attack-shows-it-came-from-north-report-idUSKBN1YN299.

[7] See Derrick Tin, Zachary Kallenborn, Alexander Hart, Attila J. Hertelendy, and Gregory R. Ciottone, “Opioid Attack and the Implications for Counter-Terrorism Medicine.” Prehospital and Disaster Medicine. Vol. 36, no. 6. 11 October 2021: pp. 661–663, https://pubmed.ncbi.nlm.nih.gov/34629120/.

[8] For background, see Arthur Holland Michel, Counter-Drone Systems, 2^nd Edition. Annandale-on-Hudson: Center for the Study of the Drone at Bard College. December 2019.

[9] Mary L. Cummings and Paul J. Mitchell, “Predicting Controller Capacity in Supervisory Control of Multiple UAVs.” IEEE Transactions on Systems, Man, and Cybernetics. Vol. 38, no. 2, 2008, https://ieeexplore.ieee.org/ document/4445761.