Swarm Weapons: Demonstrating a Swarm Intelligent Algorithm for Parallel Attack

By Sean M. Williams

Editor’s note: The following article contains the conclusions reached from research on the feasibility and implication of swarming weapons for use in the future operating environment. For a copy of the full monograph, please contact the author directly at sean.williams.3.sw@gmail.com. Unless otherwise annotated, images are author’s original work.

Estimated Reading Time: 7 minutes

Swarm weapons are feasible within the next decade and those swarm weapons could have a significant impact on the battlefield providing operational advantages through employment concepts like swarm parallel warfare, cooperative survival, pulsing, and adaptive area reconnaissance. To some extent, swarms are already starting to emerge on a small scale. Today, academic and military institutions are continually launching smaller scale swarms with up to 100-agents. Even a swarm of this scale could have an impact on the battlefield, but the real change and potential will come as the size of these swarms start to increase over the next decade. As swarms measured in the hundreds start to emerge, the true potential of swarm weapons becomes possible. By integrating existing technology with these new and larger swarms, their potential becomes apparent. A swarm weapon is more than just a quantitative advantage; it is the ability for a weapon to adapt to the changing environment through emergence.

Emergence allows the actions of a single agent at the micro scale to shape the swarm’s overall behavior at the macro scale. In effect, a swarm is capable of battlefield-wide adaptation. The combination of macro and micro adaptation allows the swarm weapon to advantageously shape the battlefield. Integrating the human and swarm together into a team achieves a new level of capability that is impossible otherwise. The swarm offers a search capacity, but the human offers the advantage of reason, analysis, and critical thinking. By integrating the strengths of both, the human-swarm teaming concept is truly more capable than either is alone. The question now is how you could use these capabilities of pulsing, parallel warfare, area adaptive reconnaissance, and cooperative survival on the battlefield.

Concept of Employment #1: Swarm Breach
The first concept of employment to discuss is the swarm breach because it is feasible even today with 100 weapons using the algorithm proposed here. The most likely use of a swarm breach would be a shaping operation to enable the maneuver of another unit. In the land and maritime domain, it might be the breach of an enemy defensive line then holding that breach point open while the force designated as the main effort exploits the breach. In the air domain, this would be very similar, but likely against surface-to-air missile systems with a follow-on force exploiting the breach by penetrating the enemy support area to interdict targets. The advantage of a swarm breach over conventional weapons is that the adaptive behavior of the swarm would allow the swarm to hold the breach open, integrate swarm reserves, and respond to enemy reinforcements.


Concept of Employment #2: Swarm Area Defense
The second concept of employment to consider is the swarm area defense. Field Manual 3-90-1, Offense and Defense defines area defense as a defensive task that concentrates on denying enemy access to designated terrain for a specific time. In this case, the swarm’s search area would coincide with the unit’s defense area. As threats enter the defense area, the swarm would detect threats, relay imagery and position to a human for confirmation, and then attack. In this case, multiple swarms can act in a layered defense with a swarm out further providing an outer perimeter defense. The advantage of swarm area defense is that the swarm’s adaptive area reconnaissance capability combined with pulsing allows the swarm to adapt and overcome both enemy breaches and frontal attacks.


Concept of Employment #3: Swarm Wide Area ISR
The third proposed method to employ the swarm weapon would be a wide area surveillance network in a contested environment. The US military has typically dominated the air, land, sea, and space domain, but the proliferation of advanced weapons is challenging that dominance especially in near-peer contested environments. Many of the intelligence platforms used today will not be survivable against a near-peer competitor, thus requiring military leaders rethink how they gather intelligence. Swarm weapons offer a potential solution that can surge into a contested environment to gather intelligence using the same concept proposed for the parallel attack. In this case, rather than attack the target, the drone only captures imagery to either store onboard for exploitation after it returns to base or it could be offboarded using beyond line of sight communication. In effect, the swarm can find both threats and targets in a search area using radio frequency and visual sensors. The adaptive wide area network allows the network to adapt to losses of drone assets and communication degradation. Cooperative survival increases the swarm’s resilience and allows the swarm to search a wide area for targets while surviving in a contested environment. The advantage of this form of employment is that it meets a critical ISR requirement in a contested environment while maintaining survivable systems.


Concept of Employment #4: Swarm Parallel Warfare
The final concept proposed here is parallel warfare, which is arguably the most difficult and contentious but also provides a significant advantage. This concept is feasible within the next ten years. Additionally, the use of swarm weapons and their adaptive emergent behavior makes the weapon more efficient and lethal. Rather than having to deploy 1000 non-swarming weapons, a swarm can accomplish more with less. Modeling showed the swarm was able to achieve the same effects using only 800 drones. This form of selective targeting makes it possible to implement John Warden’s theory of operational paralysis by simultaneously attacking the enemy’s vulnerable redundancies.

Each of these capabilities offers an operational advantage on the battlefield, but there are still gaps in existing research to address. First, research efforts must find a technological solution to the swarm communications network. This network should emphasize the use of the emergent command and control model because of the advantages in swarm range especially when considering the wide area surveillance and parallel attack concept. The range of the emergent network would be critical to either one of these concepts. Next, researchers need to analyze the human-swarm teaming concept in more detail looking at long-range communications architectures. Because parallel warfare would require these weapons operate beyond the enemy front-line forces, beyond line of sight communications with the swarm would be critical. Finally, an often-overlooked issue with swarm weapons is the logistical requirements of supporting swarm weapons. The results of this research effort showed that, although expendable, swarm weapons have the potential to also be survivable. Research efforts should analyze whether it is technologically feasible to reuse swarm weapons and what impact would that have on logistics. As swarm weapons develop, researchers must address each of these issues as early as possible.

In conclusion, swarm weapons have begun their transition from science fiction to reality. As states progress along this path over the next decade, the question they must ask is what impact they will have on the battlefield. Although they will not dramatically change our concept of war, they have the potential to significantly shape the battlefield to one’s advantage. Through emergence, swarm weapons can adapt at both the macro and micro scale. The advantage they provide is more than just a quantitative advantage. Swarm weapons can adapt and selectively target in a way that they shape the battlefield. This provides more than just an interdiction capability and it is more than just an air capability. Through swarm breach, wide area surveillance, parallel warfare, and area defense, they have the potential to shape and affect all domains.

Sean Williams is currently a student at the School for Advanced Military Studies at Fort Leavenworth, Kansas.

The views expressed are those of the author and do not necessarily reflect the official policy or position of the Department of the Air Force or the US government.

Leave a Reply