US 11,986,961 B2
Machine learning model operation management system and machine learning model operation management method
Hitoshi Hasunuma, Kobe (JP); and Jun Yamaguchi, Kobe (JP)
Assigned to KAWASAKI JUKOGYO KABUSHIKI KAISHA, Kobe (JP)
Appl. No. 17/434,304
Filed by KAWASAKI JUKOGYO KABUSHIKI KAISHA, Kobe (JP)
PCT Filed Feb. 27, 2020, PCT No. PCT/JP2020/008150
§ 371(c)(1), (2) Date Aug. 26, 2021,
PCT Pub. No. WO2020/175643, PCT Pub. Date Sep. 3, 2020.
Claims priority of application No. 2019-037053 (JP), filed on Feb. 28, 2019.
Prior Publication US 2022/0152823 A1, May 19, 2022
Int. Cl. B25J 9/16 (2006.01); G05B 13/02 (2006.01); G05B 19/042 (2006.01); G06F 21/10 (2013.01); G06F 21/62 (2013.01); G06N 3/084 (2023.01)
CPC B25J 9/1656 (2013.01) [B25J 9/163 (2013.01); G05B 13/0265 (2013.01); G06F 21/10 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A machine learning model operation management system comprising:
a model building server that receives training data which is data for making a machine learning model to train an operation of a robot, and builds a trained machine learning model based on the training data; and
an operation server that receives operation information generated by a robot controller when the trained machine learning model operates for determining the operation of the robot, the trained machine learning model stored in the robot controller, wherein
the robot controller transmits externally an inquiry as to whether the robot controller has permission to use the trained machine learning model stored in the robot controller, the inquiry including model identification information that uniquely identifies the trained machine learning model, and if the robot controller has the permission to use, the robot controller makes the trained machine learning model available for use by the robot controller, and
the operation server stores the operation information in association with information that uniquely identifies the trained machine learning model.