Researchers at Georgia Tech's GRITS Lab are working with swarms of mini robots that communicate with one another to work effectively. The aim of the research is to create networks that can be controlled by inputting instructions to a single robot. Beginning with a leader, each robot communicates with its nearest neighbors until the instructions have been shared across the entire network. In an effort to create the most efficient "follow-the-leader" algorithms, the researchers are getting the robot swarm to play musical notes on a simulated piano.
The control algorithms take into account the number of robots and their individual locations, which are tracked by cameras stationed above them. Each of the Khepera robots also comes equipped with a number of ultrasonic range sensors and cameras that allow the robot to detect its proximity to others. A good algorithm leverages the robots' positioning to play the notes with a regular tempo and minimize the total distance traveled, while simultaneously preventing bumps and collisions with other members of the swarm.
It's similar to the work done by researchers at the University of Pennsylvania, who programmed a swarm of robot quadrotors to play the James Bond theme together.
Musical inspiration could pave the way towards practical applications of the swarms. One group at the lab is developing a mobile sensor network that could be deployed on Antarctica to study the effects of climate change. Another is using the robots as stand-ins for military convoys overseen by protective drones.
GRITS Lab's Khepera bots play a rendition of Beethoven's "Fur Elise" in the following video.