# US 2019/0317456 A1 · Adjusting device and adjusting method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0317456 A1
- Document ID: US20190317456A1
- Status: Published application
- Kind: A1
- Application number: 16/286,982
- Filed: 2019-02-27
- Published: 2019-10-17
- Earliest claimed: 2018-04-13
- Assignees: Fanuc Corporation
- Applicants: Fanuc Corporation
- Inventors: Ryoutarou TSUNEKI; Shunpei TANAKA
- Web page: https://patentgrep.com/patent/US20190317456A1

## CPC

- G05B 13/0265: {the criterion being a learning criterion}
- G05B 19/4155: characterised by program execution, i.e. part program or machine function execution, e.g. selection of a program
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)

## Abstract

To provide an adjusting device and an adjusting method for appropriately controlling the machine learning reduced in cost with respect to calculation load and learning period of time in the case where an evaluation program for machine learning is used separately from a machining program and the like. The present invention includes a feedback information acquiring part configured to acquire, from a control device, feedback information obtained when an evaluation program including various types of learning elements is executed in the control device, a determination part configured to determine which learning element the acquired feedback information corresponds to among the various types of learning elements, a feedback information transmitting part configured to transmit the acquired feedback information to a machine learning part corresponding to the learning element, a parameter setting information acquiring part configured to acquire control parameter setting information obtained through machine learning by use of the feedback information, and a parameter setting information transmitting part configured to transmit the acquired control parameter setting information to the control device.

## Claims

### Claim 1

An adjusting device configured to control a control device configured to control a motor configured to drive at least two axes and control a machine learning device having a plurality of machine learning parts configured to perform machine learning to the control device, the adjusting device comprising:
a startup command output part configured to output a startup command to start the machine learning part;
a feedback information acquiring part configured to acquire, from the control device, feedback information obtained when an evaluation program including various types of learning elements is executed in the control device;
a determination part configured to determine which learning element the acquired feedback information corresponds to among the various types of learning elements;
a feedback information transmitting part configured to transmit the acquired feedback information to the machine learning part corresponding to the learning element;
a parameter setting information acquiring part configured to acquire, from the machine learning part, control parameter setting information obtained through the machine learning by use of the feedback information; and
a parameter setting information transmitting part configured to transmit the acquired control parameter setting information to the control device.

### Claim 2 (depends on 1)

The adjusting device according to claim 1, the adjusting device comprising:
a storage part configured to store the evaluation program; and
an evaluation program output part configured to output the evaluation program to the control device.

### Claim 3 (depends on 1)

The adjusting device according to claim 1, the adjusting device comprising:
an evaluation program execution command output part configured to transmit an execution command to execute the evaluation program to the control device.

### Claim 4 (depends on 1)

The adjusting device according to claim 1, wherein
when the motor drives the at least two axes, the feedback information acquiring part acquires the feedback information from the control device, and transmits the acquired feedback information to the machine learning part corresponding to the feedback information.

### Claim 5 (depends on 1)

The adjusting device according to claim 1, wherein
the feedback information includes at least one of a position command, a position feedback, and a positional error.

### Claim 6 (depends on 5)

The adjusting device according to claim 5, wherein
the feedback information further includes at least one of a speed command, a speed feedback, and a speed error.

### Claim 7 (depends on 5)

The adjusting device according to claim 5, wherein
the feedback information further includes at least one of a current command, a current feedback, and a current error.

### Claim 8 (depends on 5)

The adjusting device according to claim 5, wherein
the feedback information further includes backlash compensation.

### Claim 9 (depends on 1)

The adjusting device according to claim 1, the adjusting device comprising:
a parameter correction information transmitting part configured to transmit control parameter correction information obtained from the machine learning part as action information to the control device, wherein
the machine learning is reinforcement learning, and
the feedback information acquiring part acquires the feedback information as state information from the control device.

### Claim 10 (depends on 1)

The adjusting device according to claim 1, wherein
the control device includes a plurality of position feedforward calculation parts, a plurality of speed feedforward calculation parts, a plurality of current feedforward calculation parts and a plurality of backlash compensating parts, and
in at least one of the plurality of position feedforward calculation parts, the plurality of speed feedforward calculation parts, the plurality of current feedforward calculation parts and the plurality of backlash compensating parts, a control parameter is set or updated by the control parameter setting information acquired from the machine learning part corresponding to the at least one.

### Claim 11 (depends on 1)

The adjusting device according to claim 1, wherein
the adjusting device includes the machine learning device.

### Claim 12 (depends on 1)

The adjusting device according to claim 1, wherein
the adjusting device is included in the machine learning device.

### Claim 13 (depends on 1)

The adjusting device according to claim 1, wherein
the control device includes the machine learning device.

### Claim 14 (depends on 1)

The adjusting device according to claim 1, wherein
the adjusting device is included in the control device, and further includes the machine learning device.

### Claim 15 (depends on 1)

The adjusting device according to claim 1, wherein
the adjusting device is included in the machine learning device, and the machine learning device is included in the control device.

### Claim 16

An adjusting method of an adjusting device for controlling a control device configured to control a motor configured to drive at least two axes and controlling a machine learning device having a plurality of machine learning parts configured to perform machine learning to the control device, the adjusting method comprising the steps of:
starting the machine learning part;
acquiring, from the control device, feedback information obtained when an evaluation program including various types of learning elements is executed in the control device;
determining which learning element the acquired feedback information corresponds to among the various types of learning elements;
transmitting the acquired feedback information to the machine learning part corresponding to the learning element;
acquiring, from the machine learning part, control parameter setting information obtained through the machine learning by use of the feedback information; and
transmitting the acquired control parameter setting information to the control device.
