Introducing NavFoM: The World’s First Cross-Embodiment, Cross-Task Navigation Foundation Model
A Galbot presenter introduces NavFoM, a navigation model for the Galbot dog, and describes how it handles tracking, obstacles and delivery routes.
In this talk
- 0:05
NavFoM’s scope
NavFoM is presented as a foundation model designed to work across different robot embodiments and navigation tasks.
- 0:22
Perception and movement
The Galbot dog uses 360-degree vision to respond to its surroundings in real time. It is described navigating through varied environments.
- 0:46
Tracking people safely
The dog avoids sudden obstacles such as pedestrians while following a target, including when nearby distractions resemble that target. If the target leaves view, it searches and reacquires them. When crossing roads, it follows closely with safety as a priority.
- 1:23
Autonomous delivery routes
For a delivery task, the dog plans its route and moves autonomously through crosswalks, over bridges and up stairs to reach the destination.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.