Walt Disney Imagineering Shares How Robotic Olaf Learned to Walk on a Boat and More Using NVIDIA Technology

Walt Disney Imagineering sent their self-walking Olaf on a field trip to NVIDIA GTC, the world’s largest AI conference, where the jovial snowman joined NVIDIA CEO Jensen Huang onstage at his keynote.

Olaf Closes Out NVIDIA GTC Keynote

NVIDIA CEO Jensen Huang closed out his nearly two-hour March 2026 keynote address with a segment on physical AI and robotics, highlighting one particular case study: the self-walking animatronic Olaf coming to both World of Frozen lands.

NVIDIA worked with Disney Research on their proprietary Kamino simulator, a GPU-based physics solver that trains robots via thousands of parallel environments, allowing the robot to learn through deep reinforcement in simulation. Olaf was used to demonstrate how this translates to navigating real-world obstacles and terrain, with examples showing a collision with a cart and walking through snow.

Olaf popped up onstage to interact and walk around with Huang, showing how he could easily navigate the environment and physically and verbally respond to Huang’s exclamations.

Olaf in World of Frozen at Disneyland Paris

With Disney Adventure World set to open soon at Disneyland Paris, we’ve been seeing more footage of the self-walking animatronic Olaf, who will make his debut in World of Frozen. One of these recent reveals has been his surprising ability to navigate the moving boat used in the Snowflower Festival show.

Kyle Laughlin, SVP, Research & Development and Technology & Engineering at Walt Disney Imagineering, talked with Disney Experiences about what they’ve been doing to get Olaf “show-ready” by March 29, and revealed that deep reinforcement learning through Kamino has been crucial to the process:

Olaf will be debuting at World of Frozen as part of the daily Celebration in Arendelle show, which takes place on a boat in the lagoon. This meant he had to learn to balance on an unstable surface. Through deep reinforcement learning, in just a matter of hours, he earned his sea legs. … Deep reinforcement learning through simulation allowed Olaf to learn to stand and walk in challenging environments, like on a boat, in a fraction of the time it would take a human child.

Laughlin also shared that Kamino is allowing Imagineering to develop characters at an accelerated rate of both production and quality:

The speed at which we’re able to create new characters — and get them in front of our guests — is unprecedented. We are working to bring more emotive, expressive, and surprising characters to guests at our parks and ships around the world.

You can watch the full NVIDIA GTC keynote here (the Olaf segment starts at 2:37:39):

For the latest Disney Parks news and info, follow WDW News Today on Twitter, Facebook, and Instagram.