I particularly like one of the definitions for the word “swarm” that can be found in the Merriam Webster dictionary: “A large number of animate or inanimate things massed together and usually in motion.”
Swarms of individual animals, such as birds, can attain incredible mass, involving thousands moving in lockstep with each other. Coordinated motion of this size and complexity implies some kind of behavioral structure, and as with just about anything behavioral, structures have rules that are followed by every individual. Rules of course can be mathematically modeled and Wikipedia’s entry on “Swarm Behavior” captures this point: “From the perspective of the mathematical modeler, it is an emergent behavior arising from simple rules that are followed by individuals and does not involve any central coordination.”
Did you catch the part about not involving any central coordination? This sets swarming apart from centrally coordinated, large scale behavior such as an orchestra that plays under the direction of a conductor.
With apologies in advance to any purist of English language usage, I am going to use the word “swarm,” in the context defined above, to apply to all natural and non-natural examples of individual things that come together for some collective benefit following rules instinctually embedded in each individual. Bees and other animate things swarm for protection or some other advantage that they cannot alone achieve. We call this “instinctual” behavior because an animal is born with the rules embedded in its brain. Something, such as the presence of a predator, triggers the instinct to come together and the rules of behavior, such as turn right when the bee on your left does, ensure than every individual bee turns right on cue.
This brings up another important point about swarming. It is tactical, not strategic. What I mean by this is that swarming occurs in response to an immediate threat or need, and is not planned in advance as a strategy, although this is changing with the kind of optimization applications described above. Something triggers an instinctual response on the part of many individual entities that is designed to meet the specific nature of the threat involved. When confronted with a predator, fish swarm in a massive balls of individuals, offering each individual a greater chance for survival.
The effort humans have expended on the observation and understanding of swarming leans disproportionally to military developments. We see that with military applications, swarming has long been a tactical practice. Again from Wikipedia, “military swarming is primarily tactical, sometimes operational and rarely strategic, and is a complement to other efforts rather than a replacement for them. Swarming is a logical extension of network-centric warfare….”
In 1998, a seminal paper was published in the Proceedings Magazine of the U.S. Naval Institute, entitled “Network-Centric Warfare – Its Origins and Future.” Long before Facebook and other social networks as we know them today existed, the authors of this paper argued that the basis of war was evolving to reflect the way information technology was changing how people did business, and interacted with each other. War planning was moving away from “platform” centric strategy and tactics which were defined by massive weapons systems and top-down direction and control to achieve the desired strategy of the commander. You did what you were ordered to do even if the situation on the ground changed.
Reflecting on the fact that people using mobile phones, and equipped with powerful personal tools such as PCs connected to email and internet service providers, were taking on more control over their purchasing and socialization choices as members of growing networks of like minded people, the authors foresaw that network-centric warfare would enable “…forces to organize from the bottom up—or to self-synchronize—to meet the commander’s intent.” Self-synchronization; isn’t that a characteristic of swarming?
That changing view of warfighting was penned in 1998. Let’s fast forward to 2014. Connected, smart devices for use in military applications have revolutionized war fighting. Advances in wireless networks have had an equally compelling impact on how offensive and defensive tactics can adapt to the changing situation on the ground during battle.
Further, connected smart “micro” devices interacting with each other in a larger “macro” device (an intra-device network application) can in turn allow the macro devices to communicate and cooperate with each other (an inter-device network application). Within these devices and their network applications, programmers can embed rules of behavior (instincts) that within each macro device triggers a response to a predetermined stimulus, and then embed rules of cooperation between macro devices to organize in a manner that can neutralize the threat and optimize the chances of survival of the greatest number of macros. Just like fish when they swarm.
It should come then, as no surprise, that the U.S. Navy demonstrated a swarming defense system this year that brought together a number of autonomous, unmanned, smart, connected devices (small gun boats) that could respond to a direct threat. This system also had the benefit of being cheap because it repurposed systems developed by NASA (National Aeronautics and Space Administration) for the use in space probes.
This past August, on the James River in Virginia, 13 small autonomous boats simulated a patrol during which a potential enemy vessel was detected. This triggered a defensive response in five of the boats, which changed course toward the target and coordinated positions among themselves. Movement of the swarm of boats to the target was self-regulated following rules of engagement embedded in each of the boats onboard control systems. They were self coordinating but reflexively cooperative, reflecting the primary characteristics of swarming.
In the past three years we have seen a surge of reports about autonomous, smart, connected devices that can organize themselves into cooperative swarms that improve the chances of success in defined missions, such as disaster searches. One of the more popular TED talks, recorded in 2012 that now has over 3 million views and subtitles in 31 languages, was by Vijay Kumar, entitled “Robots That Fly…and Cooperate.” Kumar is the Deputy Dean for Education in the School of Engineering and Applied Science at the University of Pennsylvania, and headed up the GRASP (General Robotics, Automation, Sensing and Perception) Lab as it experimented with cooperative autonomous drones.
This brings me back to the title of this column: Swarming is how you say “social networking” in M2M. You are going to hear more about this because the number of people investigating and innovating ways to make autonomous, smart, mobile connected devices swarm cooperatively continues to increase. Who are these people? You can meet many of them at the annual SwarmFest, an event that began in 1998 (that was a really seminal year as you saw above) under the auspices of the Swarm Development Group, formed to foster the interdisciplinary development of complex adaptive systems.
It is not inconceivable that some day in the future, the swarm you see on the horizon coming toward you will be connected devices, not birds. Speaking of bird swarms, have you seen Hitchcock’s master work, The Birds? Pleasant dreams!
Tim Lindner is senior business consultant with a software company and a regular contributor to Connected World. He can be reached at firstname.lastname@example.org[button link="https://connectedworld.com/subscribe-connected-world/" color="default" size="small" target="_self" title="" gradient_colors="," gradient_hover_colors="," border_width="1px" border_color="" text_color="" shadow="yes" animation_type="0" animation_direction="down" animation_speed="0.1"]Subscribe Now[/button] Gain access to Connected World magazine departments, features, and this month’s cover story!