Tag Archives: Georgia Tech

Robot turns musicians into three-armed drummers

Georgia Tech researchers have created a wearable robotic limb that allows drummers to play with three arms.

Georgia Tech researchers have created a wearable robotic limb that allows drummers to play with three arms. The two-foot long smart arm attaches to a musician’s shoulder, and responds to human gestures and the music it hears.

The robotic arm is ’smart’ for a few reasons. First, it knows what to play by listening to the music in the room. It improvises based on the beat and rhythm. For example, if the musician plays slowly, the arm slows the tempo. If the drummer speeds up, it plays faster.


Beyond that, the extra limb knows where it’s located at all times, where the drums are, as well as the direction and proximity of the human arms. When the robot approaches an instrument, it employs its built-in accelerometers to sense the distance and proximity. On-board motors ensure the stick is always parallel to the playing surface, enabling it to rise, lower or twist to ensure solid contact with the drum or cymbal. The arm moves naturally with intuitive gestures because it was programmed using human motion capture technology.

“If you have a robotic device that is part of your body, it’s a completely different feeling from working alongside a regular robot,” explains Georgia Tech professor and project supervisor Gil Weinberg. “The machine learns how your body moves and can augment and complement your activity. It becomes a part of you.”

Gil Weinberg with Tyler White playing drums with a robotic arm

The team is currently exploring the use of an EEG headband, so future robotic arms could read a drummer’s brain waves to detect when they think about changing tempo or position. But why stop at music? Looking ahead, the researchers also hope to expand their efforts into the healthcare and industrial fields.

“Imagine if doctors could use a third arm to bring them tools, supplies or even participate in surgeries. Technicians could use an extra hand to help with repairs and experiments,” Weinberg adds.

You can see their prototype in action below, and read all about it here. Rock on!

[Images: Georgia Tech]

Control your own swarm of robots with a swipe of your finger

Using a tablet and a red beam of light, researchers have created a system that enables people to control an army of robots with the swipe of a finger.

Have you always wanted to control your own swarm of tiny robots? Thanks to Georgia Tech researchers, not only is it possible, it’s as easy as swiping your finger across a tablet. While commanding an army of Terminator-like machines may be cool and all, these bots were designed to work unison to accomplish a common objective throughout industrial and agricultural settings, as well as in disaster recovery missions.


As the team reveals, leading the mob of robots is pretty straightforward, only calling for a mobile device and a red beam of light. An operator taps the tablet to control where a beam of light appears on a floor. The swarm then rolls toward the illumination, constantly communicating with each other and deciding how to evenly cover the lit area. When the user swipes the tablet to drag the light across the floor, the robots follow. If the operator puts two fingers in different locations on the tablet, the machines will split up to cover both areas and repeat the process.

“It’s not possible for a person to control a thousand or a million robots by individually programming each one where to go,” explained Magnus Egerstedt, Schlumberger Professor in Georgia Tech’s School of Electrical and Computer Engineering. “Instead, the operator controls an area that needs to be explored. Then the robots work together to determine the best ways to accomplish the job.”

Professor Egerstedt envisions various use cases for such bots, including tsunami-ravaged regions. For instance, the robots could search for survivors, dividing themselves into equal sections. If the help of machines were suddenly required in a new area, a single person could instantly redeploy them. Another prime example is employing the robots for agriculture. Given that the technology is simple to use, any farmer could now streamline the crop-checking process without having to physically walk down each field.

Impressively, what sets the Georgia Tech algorithm apart from others is that the robots can change their minds. In other words, if a user sends them to one area, that operator can quickly change their path with just a swipe to send them somewhere else. Egerstedt adds, “The field of swarm robotics gets difficult when you expect teams of robots to be as dynamic and adaptive as humans. People can quickly adapt to changing circumstances, make new decisions and act. Robots typically can’t. It’s hard for them to talk and form plans when everything is changing around them.”

What’s more, the tablet-based control system is geared towards everyone — even those without robotics experience. Now, if only we can deploy a similar army of bots to clean the house…. Intrigued? Read more from the Georgia Tech team here, as well as head over to its official page to keep up-to-date with the latest projects from the GritsLab.

A cosplay cyberpunk build

Cyberpunk films and novels are often set in post-industrial dystopias characterized by extraordinary cultural ferment and the use of technology in ways never anticipated by its original creators.

As William Gibson noted in Burning Chrome, “the street finds its own uses for things.” 

Although Gibson wrote those words way back in 1981, they more than aptly describe the cyberpunk build designed by the very talented mel ell, a cos-player and graduate student at Georgia Tech in Atlanta.

According to the Adafruit crew, mel ell admires the cyberpunk and fantasy genre styles of various artists in comics and games. Indeed, she used the genre art as a basis for the design of her slick outfit.

“The process was a lot of fun and took approximately three months of on-and-off planning and building. The assembly is made from over 60 parts designed in Solidworks and sewn/cut/glued/laser-cut/heat-formed using various techniques,” mel ell explained.

“The costume includes color changing LEDs on the spine and front that are controlled by an [Atmel-based] Arduino board and onboard RGB controllers. It is powered by 16 AA batteries, 1 LiPo rechargeable battery, two 2032 coin cells and one 9-volt battery. In total there’s more than 70 LEDs on the entire costume and over 60 parts.”