In an article for the MIT Technology Review, Jamie Condliffe writes on the U.S. Department of Defense’s (DoD) use of manned aircraft to launch drone swarms in October, 2016 (2017). A swarm of 103 Perdix drones, themselves developed at MIT, were launched by three F/A-18 fighter aircraft and were able to autonomously execute pre-planned missions without outside intervention. Even more importantly, the drones were able to self-regulate their formation and mission to account for losses, something critical over the battlefield.
Because the drones are designed to avoid a centralized, decision-making hub, they are able to monitor each other and adjust their flight parameters to account for the addition, subtraction, or movement of other drones in the swarm. Extending and combining the unmanned network, a larger unmanned vehicle could launch these smaller drones, which in turn could conduct multiple missions, depending on how they were equipped, and self-heal the group to provide maximum probability of success. This significantly reduces the number of personnel required to manage complex missions, minimizes human exposure to high-risk combat environments, and provides a less intrusive surveillance ability. Tested releases up Mach .6 easily put these in the performance envelope of any aircraft, including unmanned launch vehicles.
This self-controlling concept is a mimicry of what occurs in nature every day, and is gaining traction outside the military in civil applications. Intel has demonstrated swarm technology capabilities in several instances, culminating in the halftime show of Super Bowl LI this year (Intel, 2017). Disney cooperated with Intel and also demonstrated the capabilities of swarms in its holiday light show “Starbright Holidays” (Barrett, 2016). These two cases are obvious entertainment applications and may very well presage the future of aerial displays, augmenting or even supplanting riskier fireworks. Self-regulating drone formations are particularly promising in environments where multiple sensors (e.g., cameras) need to be deployed and coverage maintained even with inevitable losses due to, say, maintenance. For example, news or law enforcement coverage of large demonstrations could be easily done from a variety of vantage points and deployment time would be limited only by how long it takes to program the system profile. Multiple generic profiles could even be pre-prepared to further reduce deployment times. Environmental applications, such as atmospheric measurement, agriculture, and survey are other areas where this type of technology could excel. Mining and other enclosed area survey/mapping applications could benefit from multiple drones providing data input while adjusting on the fly to changing conditions. Hazardous areas such as a mine collapses or gaseous release, to radioactive zones in a faulty reactor, could be quickly and safely surveyed by a highly adaptable swarm of drones.
The future of drones may very well be in swarms rather than the inefficient use of one controller to one drone. Generating more data reliably and safely (and maybe even more cheaply) is going to change how industry, government, and military users think about unmanned aerial systems. Truly, the limits to their capabilities are only in the sensors that can be put on them and the imaginations of everyone involved.
Barrett, B. (2016, November 16). Disney’s Latest Attraction? 300 Drones Flying in Formation. Retrieved from https://www.wired.com/2016/11/disneys-latest-attraction-300-drones-flying-formation/
Condliffe, J. (2017, January 19). Watch a swarm of 100 drones being dropped from fighter jets. Retrieved from https://www.technologyreview.com/s/603337/a-100-drone-swarm-dropped-from-jets-plans-its-own-moves/
Intel (2017). Intel Drones Light up the Sky. Retrieved from http://www.intel.com/content/www/us/en/technology-innovation/aerial-technology-light-show.html