Control system, factory system, learning system, estimation model generation method and actuator state estimation method

摘要

The control system 1 includes a factory system 3 configured to control actuators 11, 12, 13, 14, 15, and 16, and a learning system 2 configured to extract information for machine learning associated with actuators 11, 12, 13, 14, 15, and 16 from factory system 3 via network NW and generate an estimation model of the condition of the actuators 11, 12, 13, 14, 15, and 16 by machine learning using the information. The factory system 3 includes a plurality of nodes X including a robot controller 4 configured to control the actuators 11, 12, 13, 14, 15, and 16 and a host controller 5, and a communication management device 7 connected to the learning system 2 via the network NW. The communication management device 7 includes the data acquisition unit 511 that acquires information for machine learning from one of the nodes X other than the communication management device 7, and a data construction unit 512 that constructs a data including the information acquired by the data acquisition unit 511 and makes the data readable from the learning system 2.