Drift correction for industrial augmented reality applications

• Main Authors: James Aaron Crews, Joshua Sanders Friesenhahn, Trevor Duncan Schleiss
• Format: Patent
• Language: English
• Published: 16.11.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

This application relates to a method for updating node locations in a real-world, mapped environment, the method comprising: tracking locations of a mobile device as a user of the mobile device moves through the mapped environment, capturing, by a camera of the mobile device, images of the mapped environment as the user moves through the mapped environment, identifying a specific object within a field of view of the camera, wherein identifying the specific object includes predicting, by a machine learning (ML) model processing one or more of the captured images, an object type, and identifying a node, within a map database representing the mapped environment, that corresponds to the object type and using a tracked location of the mobile device to update or confirm a location of the node stored in the map database.