Machine learning device, machine learning method, and machine learning program

A machine learning method of a machine learning device that learns an action of a driving source device in a transport device continuously transporting at least two transported objects along a transport path, wherein the driving source device is a motor or a clutch that switches transmission of power of the motor, the machine learning method comprising:

• acquiring position information of the at least two transported objects on the transport path on the basis of a result of detection by a sensor provided in the transport path;
• calculating a reward on the basis of a time interval between the two transported objects or a distance between the two objects using the position information acquired, according to a predetermined rule;
• learning an action by calculating an action value in reinforcement learning on the basis of the position information acquired and the reward calculated;
• and
• generating and outputting control information, wherein the control information includes at least one of a control signal, a control current, and a frequency with which the motor and/or the clutch is operated, that causes the driving source device to perform an action determined on the basis of a learning result.

A non-transitory recording medium storing a computer readable machine learning program running on a machine learning device that learns an action of a driving source device in a transport device continuously transporting at least two transported objects along a transport path, wherein the driving source device is a motor or a clutch that switches transmission of power of the motor, wherein

• a hardware processor of the machine learning device is caused to execute:
• acquiring position information of the at least two transported objects on the transport path on the basis of a result of detection by a sensor provided in the transport path;
• calculating a reward on the basis of a time interval between the two transported objects or a distance between the two objects using the position information acquired, according to a predetermined rule;
• learning an action by calculating an action value in reinforcement learning on the basis of the position information acquired and the reward calculated;
• and
• generating and outputting control information, wherein the control information includes at least one of a control signal, a control current, and a frequency with which the motor and/or the clutch is operated, that causes the driving source device to perform an action determined on the basis of a learning result.