Advances in Artificial Intelligence: Micro-Drone Swarms - BSQ Research

Advances in Artificial Intelligence: Micro-Drone Swarms

Micro-drone swarms that are capable of advanced behaviors such as collective decision-making, evasive formation flying, and adaption to constituent loss, were demonstrated by the US Department of Defense in Oct 2016. This is a remarkable advance in the use of automation and artificial intelligence for military applications.

According to a news release published by the Department of Defense, the Strategic Capabilities Office of the Pentagon, partnering with Naval Air Systems Command, successfully demonstrated one of the world’s largest micro-drone swarms. The test was conducted at China Lake, California, and consisted of 103 Perdix drones launched from three F/A-18 Super Hornets. It took place in October 2016 and was documented on Sunday’s CBS News program “60 Minutes”.

The idea of a swarm of micro-drones which can communicate with each other and are capable of intelligent behaviour was pioneered by a group of engineering students at MIT in 2011. Their ideas were subsequently modified for military use at MIT”s Lincoln Laboratory, under the direction of the Pentagon’s Strategic Capabilities Office, in 2013. The first feasibility test was conducted in 2014, and the 2016 test was the first demonstration of advanced and intelligent behaviour.

Each Perdix drone is 17 cm in length with a wingspan of 30 cm, weigh 290 grams, and can reach a speed of 110 km/hour. They have two sets of wings which are straddled by a 3D printed plastic body containing a lithium battery and a small camera. Propulsion is provided by a 6.5 cm propeller at the rear. Each individual drone does not have a separate control, but together they share a collective, distributed intelligence. They travel in a leaderless swarm, and can change their formation to dodge air defenses, and can recover from the loss of individual members. They adapt to changes in their numbers and maintain constant communication with other members at all times.

“Due to the complex nature of combat, Perdix are not pre-programmed synchronized individuals, they are a collective organism, sharing one distributed brain for decision-making and adapting to each other like swarms in nature,” said SCO Director William Roper. “Because every Perdix communicates and collaborates with every other Perdix, the swarm has no leader and can gracefully adapt to drones entering or exiting the team.”

The current generation of micro-drone swarms are built for reconnaissance and surveillance. However, plans have been made to construct swarms of larger drones with offensive capabilities. And, although the Pentagon’s current plans are to use micro-drone swarms as part of a human-machine battlefield network, it is not hard to envision a future where swarms of weaponised drones will carry out battle missions completely on their own.

Leave a Comment

Your email address will not be published. Required fields are marked *