Disney Imagineering reveals a robotic Olaf character
The Frozen character uses reinforcement learning to reproduce animator-designed motion for theme park guests.
Walt Disney Imagineering revealed a robotic version of Olaf from Frozen, the latest of its expressive character robots. The team used deep reinforcement learning in simulation to teach the robot to imitate motion created by animators, and gave it an articulated mouth and eyes and a removable carrot nose. Disney said the work draws on the Newton simulation framework developed with Nvidia and Google DeepMind, and that it aims to bring more such characters to its parks and ships. It is a prominent example of entertainment robots built specifically to delight and interact with the public.
Source: The Walt Disney Company
This archive entry summarises reporting published elsewhere; follow the source link for the full story. The summary was drafted with AI from that reporting.