US 12,406,228 B2
Methods and autonomous robots for taking inventory in a structure
Christophe Maurice Thibaut, Noyelles les Seclin (FR); Nassima Medjenah, Roubaix (FR); Dorian Ibert, Roubaix (FR); and Karl Bogaert, Roubaix (FR)
Assigned to OVH, Roubaix (FR)
Filed by OVH, Roubaix (FR)
Filed on Aug. 25, 2023, as Appl. No. 18/238,318.
Claims priority of application No. 22306282 (EP), filed on Aug. 30, 2022.
Prior Publication US 2024/0070611 A1, Feb. 29, 2024
Int. Cl. G06Q 10/087 (2023.01); G05D 1/00 (2024.01); G06T 7/73 (2017.01); H04N 23/90 (2023.01)
CPC G06Q 10/087 (2013.01) [G05D 1/0246 (2013.01); G06T 7/73 (2017.01); H04N 23/90 (2023.01)] 13 Claims
OG exemplary drawing
 
1. A method for taking inventory of a plurality of objects within a datacenter, the method comprising:
causing the autonomous mobile robot to navigate through at least a portion of the datacenter;
while navigating through the portion of the datacenter, causing at least one camera of the autonomous mobile robot to acquire a plurality of positioning images of at least one rack of the datacenter at a first image resolution;
determining that at least one positioning image of the plurality of positioning images contains an image of a predetermined landmark, wherein the predetermined landmark is a plurality of electronic equipment module lights of the at least one rack;
in response to determining that the at least one positioning image of the plurality of positioning images contains the image of the predetermined landmark, causing the autonomous mobile robot to navigate to a predetermined data collection position, the predetermined data collection position being based at least in part on a position of the autonomous mobile robot during acquisition of the at least one positioning image;
causing the at least one camera of the autonomous mobile robot to acquire at least one inventory image at a second image resolution in the predetermined data collection position, the second image resolution being greater than the first image resolution; and
extracting, from the at least one inventory image, a plurality of inventory labels, each inventory label relating to a corresponding object of the plurality of objects within the datacenter, wherein extracting the plurality of inventory labels further comprises extracting a corresponding inventory label identifying a corresponding electronic equipment module of the at least one rack.