Skip directly to content

Disney Research and Robotic Interaction

on Thu, 05/30/2013 - 22:04

The Walt Disney Company has always been at the forefront of innovation. Founded by legendary animator Walt Disney in 1923, the company rapidly revolutionized the cartoon arts with a string of “firsts” during it early years, including the first cartoon with fully synchronized sound in 1928, the first full-color cartoon in 1932 and the first animated feature film in 1937. Disneyland, meanwhile, became the first modern theme park in 1955. Not content to rest on its laurels, the Walt Disney Company made a number of strategic acquisitions in recent years, including Pixar Animation Studios, Marvel Comics and Lucasfilm. Pixar had an effect beyond the media giant’s bottom line, however, resulting in the formation of Disney Research shortly after the purchase, a division inspired by Pixar’s work in computer graphics.

“Disney Research was launched in 2008 as an informal network of research labs that collaborate closely with academic institutions such as Carnegie Mellon University and the Swiss Federal Institute of Technology Zürich,” the company’s website states. “We’re able to combine the best of academia and industry—we work on a broad range of commercially important challenges, we view publication as a principal mechanism for quality control, we encourage engagement with the global research community, and our research has applications that are experienced by millions of people. We’re honoring Walt Disney’s legacy of innovation by researching novel technologies and deploying them on a global scale.”

Disney Research has a facility in Pittsburgh located at Carnegie Mellon University, where it conducts research into everything from radio and antennas to sports visualization. Its relationship with CMU, meanwhile, also enables Disney Research to branch into the field of robotics, where it has made significant strides with motion capture technology. According to the Pittsburgh Post-Gazette, that research has recently paid dividends with the “teaching” of a robot to receive objects handed to it, not simply by placing its own hand out but recognizing the movements of its human counterpart and reacting in a similar fashion.

“I was interested in making the robot more physically interactive instead of working on emotional interaction,” senior research scientist Katsu Yamane told the Post-Gazette in May 2013. “I wanted to see the robot act human-like.” To accomplish this feat, Yamane and Marcel Revfi, an exchange student from the Karlsruhe Institute of Technology, created a database of motion-capture footage featuring two people interacting with each other, and then programming the information into a robot. When faced with its own interaction with someone, the robot searches the database to determine not only how the human will react, but how it should as well.

The research conducted by Katsu Yamane and Marcel Revfi was presented at the 2013 IEEE International Conference on Robotics and Automation, and the resulting project essay has been nominated for the “Best Cognitive Robotics Paper.” It may seem like a small accomplishment, simply teaching a robot how to accept an objects from someone else, but it is an important step nonetheless when it comes to future human-robot interaction—paving the way for the futuristic world often envisioned in science fiction classics of the past and keeping with the Walt Disney Company’s history of innovation as well.

Anthony Letizia (May 30, 2013)

Follow Geek Pittsburgh: Facebook - Twitter - RSS Feed