Niantic Spatial Developed Spatial AI for Robot Navigation
The company uses 3D mapping technology to enable centimeter-level navigation for industrial and delivery robots.
Updated on Sept. 27, 2026 in Robotics
Live Poll
Do you trust artificial intelligence to safely navigate shared spaces like streets and offices?
Niantic Spatial, which rebranded after selling its gaming unit, has developed a Visual Positioning System (VPS) that builds 3D and 4D models of physical spaces to guide autonomous machines. The technology allows robots to navigate industrial sites and urban environments with centimeter-level accuracy.
Why it matters
By mapping the physical world, Niantic Spatial aims to bridge the gap between AI software and complex real-world environments where humans and machines must operate safely together. This focus on physical space reflects an effort to capture utility in settings that represent the majority of global economic activity.
The system generates 3D models in one hour from 10-15 minutes of footage captured via 360-degree cameras. This data powers the Visual Positioning System, which achieves centimeter-level accuracy by comparing live camera imagery to pre-built spatial maps.
The players
Niantic Spatial
A technology company focused on spatial AI and 3D mapping that pivoted from a gaming business to industrial automation tools.
Cho In-hee
The CEO of Niantic Spatial who assumed leadership in March 2026 to direct the firm's transition toward spatial AI for physical work environments.
The details
Niantic Spatial utilizes 360-degree cameras to capture and reconstruct physical environments, mapping pathways and equipment into 3D and 4D structures. The Visual Positioning System, or VPS — a navigation technology that determines location by comparing live camera feeds to a stored 3D map — then uses these models to calculate a robot's precise location and orientation. This allows for navigation in complex environments like factories where standard GPS signals are often inaccurate.
Timeline
Cho In-hee joined Niantic Spatial as CEO in March 2026.
CEO Cho In-hee discussed spatial AI strategies during a meeting in Seoul on September 21, 2026.
The Tech Race
Niantic Spatial is scaling its mapping architecture beyond its initial deployment with the Singapore Land Authority to target broader industrial sectors. This move positions the company to compete for infrastructure-level AI contracts, shifting its focus toward high-value, site-specific autonomy.
The platform is currently being deployed by entities like Coro Robotics for delivery operations and ExxonMobil for machinery navigation. The technology requires 360-degree camera footage of an environment to create the necessary spatial models before robots can operate in those specific zones.
The takeaway
Niantic Spatial is betting that the transition to robot-human collaboration relies on precise, fast-generated spatial maps. Watch for upcoming announcements regarding the company's formal partnerships with South Korean industrial site operators as a key indicator of its adoption trajectory.
Further reading
For more information on the evolution of automated navigation, visit our dedicated Robotics page.
Source note: This article includes information reported by 조선일보.
Live Poll
Do you trust artificial intelligence to safely navigate shared spaces like streets and offices?
