Disney Research unveiled a new very lifelike Audio-Animatronic robot with human-like gaze, designed to give the “illusion of life.”
Disney Research unveiled a new very lifelike Audio-Animatronic robot with human-like gaze, designed to give the “illusion of life.”
Disney Research just unveiled a lifelike robot designed to mimic human actions as close to lifelike as possible. The robot can mimic actions such as head tilting, blinking, and rapid eye movement.
This whole development is part of Disney’s goal to create Audio-Animatronics that can interact in a lifelike manner by using the “illusion of life.”
Audio-Animatronics are Disney’s copyrighted animatronics systems used throughout Disney licensed areas. They’re built to move and make noise, but most are fixed to a support.
Disney’s been moving into more lifelike and mobile AAs for a while now, starting with a flipping stunt robot, and now with their lifelike facial features and movements.
Disney Research’s “Realistic and Interactive Robot Gaze” study said the full development was authored by people on Disney’s Research and Imagineering team, as well as with two researchers from the California Institute of Technology and the University of Illinois at Urbana-Champaign.
Through the report, Disney Research explained the details and workings behind the new technology, with a YouTube video having been published later with footage of the robot’s lifelike functions.
This project follows some similar work on advanced Audio-Animatronics, but focuses primarily on “technical implementation” of human gaze. The robot’s designed to add a level of realistic characteristics, specifically for its eyes and head movement.
According to the report, the highly detailed responses were copied from complex responses in humans. They further by explaining that the project workers analyzed human responses to several interactions and movements.
All of these movements and changes occur on the robot which has a human face (with no outer skin), consisting of eyelids, eyes, a nose, and a mouth.
The robot interacts with humans through a camera attaches to its chest which analyzes input from people.
Through that input, the robot can then respond through either tilting its head, blinking, moving its eyes rapidly (like humans do), or breathing.
Copying actions from humans allows this robot to look human-like without verbalizing. The combination of this technology with Disney’s advanced speech systems in other Audio-Animatronics would allow realistic movements combined with lifelike speech.
The study states that “given the importance of gaze in social interactions as well as its ability to communicate states and shape perceptions, it is apparent that gaze can function as a significant tool for an interactive robot character.”
That quote from the study essentially states that in order to create a more useful interactive robot character, the use of humanlike interactions, as well as gaze, create a more perfect combination.
For example, no human maintains perfect eye contact throughout a conversation. With that knowledge, Disney’s robot doesn’t either.
Sign up for our newsletter to get bite-sized news about all five of our categories.
The specific implementation of eye gaze and movement is a part of Disney Research’s goal to create robots that can interact successfully with people in a way that Disney calls “the illusion of life.”
In this study, the “illusion of life,” consists of “robot gaze, animation, and show.” That combination would allow this specific backbone for AAs to serve as whatever character Disney needs it to be.
The illusion of life also consists of different emotional states for the robot as well, since Disney has different behavioral states consisting of “read, glace, engage, and acknowledge.”
All of these different parts add up to what Disney hopes can become a lifelike robot that creates “the illusion of life.”
Sign up for our newsletter to get bite-sized newsletters!
Sign up for our newsletter to get bite-sized newsletters!
Statural.com is an independently owned and operated news website, providing useful information across five major categories.
© 2020 Statural. All rights reserved