Machine learning device, servo control device, servo control system and machine learning method

1. A servo control system comprising:

a machine learning device including:

a label acquisition unit for acquiring, as a label, a current command that drives a control target of a servo control device in a state in which disturbance is suppressed,

an input data acquisition unit for acquiring, as input data, a velocity of the control target driven based on the current command in the state in which disturbance is suppressed,

a learning unit for constructing a learning model for estimating a current to drive the control target from the velocity of the control target, by way of performing supervised learning with a group of the label and the input data as training data, and

a function output unit for generating a function for estimating a current to drive the control target from the velocity of the control target, based on the learning model constructed by the learning unit, and for outputting the function to the servo control device; and

the servo control device for outputting the current command and the velocity of the control target to the machine learning device,

wherein the servo control device includes: an estimated current generation unit for estimating a current to drive the control target from the velocity of the control target, based on the function outputted from the function output unit a subtraction unit for obtaining a difference between the current command and an estimated current that was estimated by the estimated current generation unit and a disturbance compensating unit for generating a compensated current for compensating disturbance based on the difference,

wherein the servo control device includes a current control unit that constitutes a current feedback loop, and an amplifier, wherein the current command to drive the control target is applied to the control target via the current control unit and the amplifier, and

wherein output of the amplifier is the current command acquired by the label acquisition unit.

2. A servo control system comprising:

a machine learning device including:

a label acquisition unit for acquiring, as a label, a current command that drives a control target of a servo control device in a state in which disturbance is suppressed,

an input data acquisition unit for acquiring, as input data, a velocity of the control target driven based on the current command in the state in which disturbance is suppressed,

a learning unit for constructing a learning model for estimating a current to drive the control target from the velocity of the control target, by way of performing supervised learning with a group of the label and the input data as training data, and

a function output unit for generating a function for estimating a current to drive the control target from the velocity of the control target, based on the learning model constructed by the learning unit, and for outputting the function to the servo control device; and

the servo control device for outputting the current command and the velocity of the control target to the machine learning device,

wherein the servo control device includes: an estimated current generation unit for estimating a current to drive the control target from the velocity of the control target, based on the function outputted from the function output unit a subtraction unit for obtaining a difference between the current command and an estimated current that was estimated by the estimated current generation unit and a disturbance compensating unit for generating a compensated current for compensating disturbance based on the difference,

wherein the control target includes a motor and a load that moves by way of the motor, and

wherein the velocity acquired by the input data acquisition unit is a rotation velocity of the motor and a velocity of the load.

3. A servo control device comprising:

a machine learning device including:

a label acquisition unit for acquiring, as a label, a current command that drives a control target of a servo control device in a state in which disturbance is suppressed,

an input data acquisition unit for acquiring, as input data, a velocity of the control target driven based on the current command in the state in which disturbance is suppressed,

a learning unit for constructing a learning model for estimating a current to drive the control target from the velocity of the control target, by way of performing supervised learning with a group of the label and the input data as training data, and

a function output unit for generating a function for estimating a current to drive the control target from the velocity of the control target, based on the learning model constructed by the learning unit, and for outputting the function;

an estimated current generation unit for estimating the current to drive the control target from the velocity of the control target, based on the function outputted from the function output unit;

a subtraction unit for obtaining a difference between the current command and the estimated current that was estimated by the estimated current generation unit;

a disturbance compensating unit for generating a compensated current for compensating disturbance based on the difference; and

a current control unit that constitutes a current feedback loop, and an amplifier,

wherein the current command to drive the control target is applied to the control target via the current control unit and the amplifier, and

wherein output of the amplifier is the current command acquired by the label acquisition unit.

4. A servo control device comprising:

a machine learning device including:

a label acquisition unit for acquiring, as a label, a current command that drives a control target of a servo control device in a state in which disturbance is suppressed,

an input data acquisition unit for acquiring, as input data, a velocity of the control target driven based on the current command in the state in which disturbance is suppressed,

a learning unit for constructing a learning model for estimating a current to drive the control target from the velocity of the control target, by way of performing supervised learning with a group of the label and the input data as training data, and

a function output unit for generating a function for estimating a current to drive the control target from the velocity of the control target, based on the learning model constructed by the learning unit, and for outputting the function;

an estimated current generation unit for estimating the current to drive the control target from the velocity of the control target, based on the function outputted from the function output unit;

a subtraction unit for obtaining a difference between the current command and the estimated current that was estimated by the estimated current generation unit; and

a disturbance compensating unit for generating a compensated current for compensating disturbance based on the difference;

wherein the control target includes a motor and a load that moves by way of the motor, and

wherein the velocity acquired by the input data acquisition unit is a rotation velocity of the motor and a velocity of the load.

5. A servo control method, comprising:

via a machine learning device:

acquiring, as a label, a current command to drive a control target of a servo control device in a state in which disturbance is suppressed;

acquiring, as input data, a velocity of the control target driven based on the current command in the state in which disturbance is suppressed;

constructing a learning model for estimating the current to drive the control target from the velocity of the control target, by way of performing supervised learning with a group of the label and the input data as training data; and

generating a function for estimating a current to drive the control target from the velocity of the control target, based on the learning model, and

via a servo control device:

estimating a current to drive the control target from the velocity of the control target, based on the function generated;

obtaining a difference between the current command and the current estimated; and

generating a compensated current for compensating disturbance based on the difference; and

applying the current command to the control target via a current control unit and an amplifier,

wherein output of the amplifier is the current command that is acquired in said acquiring.

6. A servo control method, comprising:

via a machine learning device:

acquiring, as a label, a current command to drive a control target of a servo control device in a state in which disturbance is suppressed;

acquiring, as input data, a velocity of the control target driven based on the current command in the state in which disturbance is suppressed;

constructing a learning model for estimating the current to drive the control target from the velocity of the control target, by way of performing supervised learning with a group of the label and the input data as training data; and

generating a function for estimating a current to drive the control target from the velocity of the control target, based on the learning model, and

via a servo control device:

estimating a current to drive the control target from the velocity of the control target, based on the function generated;

obtaining a difference between the current command and the current estimated; and

generating a compensated current for compensating disturbance based on the difference;

wherein the control target includes a motor and a load that moves by way of the motor, and

wherein the velocity acquired by the input data acquisition unit is a rotation velocity of the motor and a velocity of the load.