Provided is a machine learning device configured to perform machine learning related to optimization of a compensation value of a compensation generation unit with respect to a servo control device configured to control a servo motor configured to drive an axis of a machine tool, a robot, or an industrial machine, and that includes at least one feedback loop, a compensation generation unit configured to generate a compensation value to be applied to the feedback loop, and an abnormality detection unit configured to detect an abnormal operation of the servo motor, wherein, during a machine learning operation, when the abnormality detection unit detects an abnormality, the compensation from the compensation generation unit is stopped and the machine learning device continues optimization of the compensation value generated by the compensation generation unit.

To determine a driving parameter for driving a motor having an unknown circuit constant in a simple way and to shorten time required for the determination. A parameter determination supporting device comprises: initial parameter determination means that determines an initial parameter for a trial run based on specification information about a motor driver and output specification information about a motor; program generation means that generates a trial run program based on the output specification information about the motor, the trial run program being used in a trial run for acquiring data necessary for adjusting a parameter for determining an output from the motor; automatic measurement means that automatically measures operation information determined by driving the motor by applying the initial parameter and following the trial run program; estimation means that estimates the circuit constant of the motor based on the operation information; and calculation means that calculates an optimum parameter conforming to the output specification of the motor based on the circuit constant.

A servo movement control method, device, and terminal device are provided. The method includes: controlling an output shaft of the servo to rotate according to a first motion instruction; detecting whether a second motion instruction is received within a first preset time period, and re-planning a second target motion curve to a second target end position from a corresponding target position on a first target motion curve when receiving the second motion instruction; and controlling the output shaft to rotate from an actual position when receiving the second motion instruction to the second target end position according to the second target motion curve. When the second motion instruction is received, the servo is controlled to rotate from the target position to the second target end position according to the second motion instruction, so that the servo is switched from the first motion instruction to the second motion instruction smoothly.

Provided is a machine learning device configured to perform machine learning related to optimization of a compensation value of a compensation generation unit with respect to a servo control device configured to control a servo motor configured to drive an axis of a machine tool, a robot, or an industrial machine, and that includes at least one feedback loop, a compensation generation unit configured to generate a compensation value to be applied to the feedback loop, and an abnormality detection unit configured to detect an abnormal operation of the servo motor, wherein, during a machine learning operation, when the abnormality detection unit detects an abnormality, the compensation from the compensation generation unit is stopped and the machine learning device continues optimization of the compensation value generated by the compensation generation unit.

A controller that performs, for one or more axes of a machine, position control by taking friction into consideration includes a data acquisition unit acquiring at least a position command and a position feedback and a compensation torque estimation unit estimating coefficients of a friction model used when the position control is performed, on the basis of a position deviation which is a difference between the position command and the position feedback.

The present disclosure relates to servo control technology, which provides a method, device, and terminal device for servo movement smoothing. The method includes: obtaining a starting position and a control command for a rotation of an output shaft the servo; determining an ending position and a rotation time for the rotation of the output shaft in accordance with the control command; constructing a movement curve of the output shaft based on the starting position, the ending position, and the rotation time; and controlling the output shaft to rotate from the starting position to the ending position in accordance with the movement curve. The above-mentioned method smooths the movement of the servo by constructing a simple linear function, which greatly reduces the calculation amount in comparison with the technical solution using the cubic Bessel formula, and is capable of reducing the requirements for the hardware performance of servos.

A servo control system includes a position command generator, a position detector for a feed axis, a positional deviation obtainer for calculating a positional deviation, a position control loop, a band limiting filter for attenuating a high frequency component of the positional deviation, a dynamic characteristic compensation element for advancing a phase, a learning controller including the band limiting filter and the dynamic characteristic compensation element, a sine wave sweep input unit for applying a sine wave sweep to the position control loop, a frequency characteristic calculator for estimating the gain and phase of position control loop input and output signals, and a learning control characteristic evaluation function calculator for calculating an evaluation function, which indicates a position control characteristic with the learning controller based on a frequency characteristic by actual measurement and a frequency characteristic of the learning controller.

A servo movement control method, device, and terminal device are provided. The method includes: controlling an output shaft of the servo to rotate according to a first motion instruction; detecting whether a second motion instruction is received within a first preset time period, and re-planning a second ideal motion curve to a second ideal end position from a corresponding ideal position on a first ideal motion curve when receiving the second motion instruction; and controlling the output shaft to rotate from an actual position when receiving the second motion instruction to the second ideal end position according to the second ideal motion curve. When the second motion instruction is received, the servo is controlled to rotate from the ideal position to the second ideal end position according to the second motion instruction, so that the servo is switched from the first motion instruction to the second motion instruction smoothly.

To determine a driving parameter for driving a motor having an unknown circuit constant in a simple way and to shorten time required for the determination. A parameter determination supporting device comprises: initial parameter determination means that determines an initial parameter for a trial run based on specification information about a motor driver and output specification information about a motor; program generation means that generates a trial run program based on the output specification information about the motor, the trial run program being used in a trial run for acquiring data necessary for adjusting a parameter for determining an output from the motor; automatic measurement means that automatically measures operation information determined by driving the motor by applying the initial parameter and following the trial run program; estimation means that estimates the circuit constant of the motor based on the operation information; and calculation means that calculates an optimum parameter conforming to the output specification of the motor based on the circuit constant.

An automatic parameter adjustment device capable of quantitatively determine the degree of importance of a processing time and a processing accuracy, and effectively adjusting a parameter in accordance with a processing condition. The adjustment device includes: a parameter changing part which changes a control parameter; a test program executing part which transmits a test program to a numerical controller for executing the test program; an execution result obtaining part which obtains an execution result of the test program; a storing part which stores the execution result and the parameter corresponding thereto; a weighting part which determines weighting coefficients of the processing time and processing accuracy as evaluation criteria based on input or setting by an operator; and a parameter extracting part which evaluates the execution result based on the weighted evaluation criteria, and extracts an optimum parameter from the storing part based on the evaluated execution result.