Prometheus is said to have stolen fire from the Gods on Mount Olympus and given it to humankind so that people could thrive by benefiting from its power. Machines, augmenting humanity’s capabilities to such an enormous degree, could be considered the new “fire,” and Carnegie Mellon University Ph.D. candidate Madeline Gannon may have just invented a new way of controlling them.
During her time as a fellow at Autodesk’s Pier 9 AIR (Artists In Residence) workshop, Gannon developed a combination of motion capture technology and software “that translates human motions into instructions for the robot,” allowing her to control the movements of machines with her own bodily movements. In other words, instead of programming a given machine arm to “reach over there and pick up that block,” Gannon can simply make the reaching and grasping motions with her arm and hand, and have the machine mimic those movements.
With her new program, Quipt, Gannon has created “a gesture-based control software that gives industrial robots spatial awareness and spatial behaviors for interacting closely with people.” This is important because, not only does it allow machines to mirror human behavior, it also means easier, more intuitive methods of controlling machines for general purpose tasks. This can help bring them out of factories and into other areas of life, like photography, film production, construction — any other realm that would benefit from using powerful, intuitively controlled robotics.
In an effort to “stimulate interest in a field hampered by prohibitively high costs of entry,” Gannon has made her software open-source. Anybody who wants access to Quipt has it. And while movement-tracking software is well established, Gannon’s software is an improvement on what’s been done before because it allows for a wider range of movements. A single CNC machine arm, for example, could be outfitted with dozens of different tools (camera, saw, marker, etc.) and still be controlled using the same software.
Ultimately, Gannon says that she wants to “change industrial robots from adversaries into collaborators,” and use them to “amplify and extend human capabilities and creativity.” In other words, these new machines don’t necessarily need to replace human workers, they can instead help humans become super workers. Which is great, because we can only drink so much coffee.
What do you think about intuitively controlled industrial machines? Are they going to aid us like never before, or do you think the divide between machine and human should remain starkly defined? Let us know in the comments section below!
Image: Autodesk Pier 9 Workshop