Disney proposes new plans for using AI in its parks
In new patent applications published this week, Disney detailed plans for automating stage production tasks in live performances.
The application, "Methods for Alignment of Plan to Execution Using Multimodal Cues for Script-Based Performance Modeling," offers a title that would dissuade all but the most dedicated fan from continuing to read it. But I plowed through the verbiage, and this is what I found.
Disney describes a system that monitors actors' performances and automatically responds to them, using machine learning. The system could be used simply to ensure that performers stay on script. That was a supervisor's job back when I worked in the Magic Kingdom, and every once in a while you'd be subject to an "SOP" check to make sure that you were not spieling stuff that, well, you shouldn't.
But that's a lot of tech effort for a system this complex. In more interesting applications, Disney could use this monitoring system for "automatically responding to cues included in the script or other plan, such as initiating sound effects, visual effects, lighting effects, or execution of animatronic or other robotic actions," the application said.
Think of this as automating the functions now carried out by a tech team on the production board. The system could, in effect, listen for the cues in the script and then trigger whatever effect should come next. In a show with live actors and Audio Animatronics, it could provide a new way to allow the two to interact. Right now, the actor has to time their performance to match the preset timing of the animatronic. Disney's proposed system flips that, and programs the animatronic control to listen to and respond to the actor.
The proposed machine learning model for managing all this opens other opportunities, as well, as Disney looks to expand its use of AI to enable more and quicker deployment of attraction and entertainment concepts.
Another patent application filed this week details Disney's plans to use machine learning to with attraction maintenance. In "Visualization-Based Anomaly Detection for Predictive Maintenance," Disney proposes a system for looking at parts in an apparatus, then comparing them to what they should look like and flagging them for repair or replacement, if needed.
"Conventional solutions for anticipating apparatus malfunctions and failures have relied upon the deep knowledge base of highly experienced system engineers, which, due to the heavy reliance of those solutions on the expertise of particular individuals. are brittle and ultimately untenable. Consequently, there is a need in the art for a solution enabling system engineers of varying degrees of experience to readily diagnose the operational states of complex apparatuses, and to accurately anticipate situations in which those system are likely to malfunction or fail," the application said.
In other words, we need an AI-powered machine that can use its automated pattern-matching abilities to help less experienced maintenance people know when it's time to swap out parts, since so many of our experience maintenance techs have left or retired. Okay, I filled in that last part, but let's be real, that's the issue here.
Follow the links above to read the applications on the U.S. Patent Office's website.
Like Theme Park Insider? Sign up for more Disney and theme park news by subscribing to our free weekly newsletter, or upgrade to a paid subscription for exclusive monthly content.
Replies (2)
If they could use it to substantially cut down wait times, I would retract all my negativity towards AI and fully embrace the rule of our robot overlords.
This article has been archived and is no longer accepting comments.

Robert wrote: "Right now, the actor has to time their performance to match the preset timing of the animatronic. Disney's proposed system flips that, and programs the animatronic control to listen to and respond to the actor."
That's truly brilliant. Excellent use of AI.
Edit: What if Disney's proposed system programs the animatronic control to listen to and respond to the guests?