US 12,393,196 B2
Systems and methods for training a robot to autonomously travel a route
Jean-Baptiste Passot, San Diego, CA (US); Andrew Smith, San Diego, CA (US); Botond Szatmary, San Diego, CA (US); Borja Ibarz Gabardos, San Diego, CA (US); Cody Griffin, San Diego, CA (US); Jaldert Rombouts, San Diego, CA (US); Oleg Sinyavskiy, San Diego, CA (US); and Eugene Izhikevich, San Diego, CA (US)
Assigned to Brain Corporation, San Diego, CA (US)
Filed by Brain Corporation, San Diego, CA (US)
Filed on Oct. 7, 2022, as Appl. No. 17/961,926.
Application 17/961,926 is a continuation of application No. 16/168,368, filed on Oct. 23, 2018, granted, now 11,467,602.
Application 16/168,368 is a continuation of application No. 15/152,425, filed on May 11, 2016, abandoned.
Prior Publication US 2023/0021778 A1, Jan. 26, 2023
Int. Cl. A47L 9/00 (2006.01); A47L 9/28 (2006.01); A47L 11/40 (2006.01); G05D 1/00 (2006.01)
CPC G05D 1/0274 (2013.01) [A47L 9/2826 (2013.01); A47L 9/2852 (2013.01); A47L 11/4011 (2013.01); A47L 11/4061 (2013.01); G05D 1/0088 (2013.01); G05D 1/0221 (2013.01); G05D 1/0246 (2013.01); A47L 2201/04 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A method for detecting an erroneous map, comprising:
producing a first map during navigation of a route by a robot using data from only a first sensor, the robot being an electro-mechanical machine configured for autonomous navigation;
evaluating a discrepancy between the first map produced by the first sensor and a second map stored in a memory of the robot, the second map being produced using data from a plurality of sensors of the robot;
determining the first map contains errors if the discrepancy between the first map and the second map is above a first threshold value, the discrepancy corresponding to a non-closed loop route in the first map produced due to an error in the first sensor; and
navigating using the second map while accounting for the discrepancy by the robot during the navigation with the second map,
wherein the evaluating of the discrepancy between the first map and the second map is based on—(i) detecting an initialization object at a start of the navigation of the route while the robot is at a first location; and (ii) detecting the initialization object at an end of the navigation of the route while the robot is at a different second location.