US 12,469,002 B2
Apparatus for automating inventory and automatic inventory system and method
Dinesh C. Verma, New Castle, NY (US); and Wayne B. Riley, Raleigh, NC (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Dec. 28, 2020, as Appl. No. 17/134,723.
Prior Publication US 2022/0207462 A1, Jun. 30, 2022
Int. Cl. G06Q 10/087 (2023.01); B25J 9/16 (2006.01); B25J 13/08 (2006.01); B25J 19/02 (2006.01); B65G 1/02 (2006.01); G06N 20/00 (2019.01); G06V 20/10 (2022.01); H04N 23/695 (2023.01)
CPC G06Q 10/087 (2013.01) [B25J 9/162 (2013.01); B25J 13/08 (2013.01); B25J 19/023 (2013.01); B65G 1/02 (2013.01); G06N 20/00 (2019.01); G06V 20/10 (2022.01); H04N 23/695 (2023.01)] 18 Claims
OG exemplary drawing
 
1. An apparatus for automating inventory procedures for products stored on shelves in a closed environment comprising:
a mobile mechanical device having at least one movable appendage, the at least one movable appendage including at least one camera and at least one additional sensory modality not being a camera;
a positioning system configured to use the at least one camera and at least one additional sensory modality to position the at least one movable appendage to take camera images of the products on the shelves in the closed environment from a plurality of different perspectives; and
a computer processing system configured to perform:
moving the mobile mechanical device to a location in the closed environment;
positioning the at least one movable appendage and using the at least one additional sensory modality to adjust the position of the at least one movable appendage to take camera images of the products on the shelves at the location from a plurality of different perspectives;
predicting context of the mobile mechanical device from an artificial intelligence context model;
selecting a type artificial intelligence model based on the predicted context;
using the selected type artificial intelligence model to predict the products on the shelf based on the camera images from the at least one camera,
selecting a counter artificial intelligence model based on the predicted products; and
using the selected counter artificial intelligence model to count the predicted products on the shelf based on the camera images from the at least one camera.