Patent Application
20090015188
The present invention relates to a method for suppressing variation in the speed of an actuator, in which variation in the rotational speed of an actuator output shaft caused by angle transmission error of a wave gear drive is suppressed in an actuator provided with the wave gear drive. The present invention also relates to a drive controller of an actuator in which this method for suppressing variation in speed is employed.
Well-known actuators include configurations in which the rate of output rotation of a motor is reduced by a wave gear drive, and a load-side member is held in a target position. The wave gear drive is provided with an annular, rigid internally-toothed gear, an annular, flexible externally-toothed gear, and a wave generator. In a typical wave gear drive, a state is created in which the flexible externally-toothed gear is bent into an elliptical shape by the wave generator, and the external tooth portions positioned on both ends of this elliptical shape in the major axial direction are engaged with corresponding internal tooth portions of the rigid internally-toothed gear. Once the motor causes the wave generator to rotate, the engagement positions of the two gears move in the circumferential direction, and relative rotation between the two gears is generated according to the difference 2n (where n is a positive integer) in the number of teeth between the two gears.
In general, the difference in the number of teeth between the two gears is two. The rigid internally-toothed gear is fixed, and the flexible externally-toothed gear rotates as an element for outputting reduced-speed rotation. The load-side member that is linked to the flexible externally-toothed gear is rotationally driven at a low speed. The reduction ratio i in this instance is represented by the equation below.
i=1/R=(Zc−Zf)/Zf
When, e.g., Zf=100 and Zc=102, the reduction ratio i is 1/50, and the output rotation moves in the opposite direction from direction of rotation of the motor.
The drive controller of an actuator having this configuration generally controls the positioning of the actuator using feedback control. Angle transmission error occurs in the wave gear drive in this instance. Error occurs between the target position and the actual position of the actuator output shaft (the output shaft of the wave gear drive) due to this angle transmission error. The positioning accuracy of the actuator provided with the wave gear drive could be improved if this error could be compensated for.
Methods for correcting the positioning error of an actuator are proposed in Patent Documents 1, 2, in which positioning error caused by angle transmission error of the wave gear drive is corrected, and positioning is accurately performed. Position accuracy is improved in Patent Document 1 by adding a degree of correction to the positioning feedback, and in Patent Document 2 by adding a degree of correction to the positioning command.
However, variations in the rotational speed of the actuator output shaft occur due to the effects of an error component in the wave gear drive in actuators provided with a wave gear drive. Such variations in rotational speed cannot be suppressed in conventional methods for correcting positioning error.
With the foregoing problems in view, it is an object of the present invention to propose a method for suppressing variation in the speed of an actuator that is capable of effectively suppressing variations in the rotation of an actuator output shaft caused by angle transmission error of a wave gear drive. It is also an object to propose a drive controller for an actuator in which this method is employed.
In order to achieve the aforementioned objects, in the present invention, a positioning error of an actuator output shaft caused by angle transmission error of a wave gear drive in an actuator provided with the wave gear drive is measured in advance; the speed of the actuator output shaft is computed in a speed-controlling feedback loop thereof using the result of correcting detected positioning information of the actuator output shaft using the positioning error; and this information is used as a speed feedback value. Alternatively, the result of using the detected positioning error to correct the speed computed from the detected positioning information is used as the speed feedback value.
In other words, in the method for suppressing variation in speed of an actuator of the present invention, an absolute position of one revolution of the motor rotating shaft is used as a reference to measure a positioning error of each rotational angle position of an actuator output shaft and create error correction data for the actuator output shaft for each of the rotational angle positions of the motor rotating shaft. During drive control, a rotational position of the motor rotating shaft is detected, the error correction data is used to determine an error-correction value allocated to the detected rotational position, and the speed of the motor rotating shaft is computed from a value obtained from using the error-correction value to correct the rotational position. Alternatively, the speed of the motor rotating shaft calculated from the rotational position is corrected on the basis of the error-correction value, and the determined speed is used as a speed feedback value for feedback control of the actuator output shaft.
In the present invention, the absolute position of one revolution of the motor rotating shaft is used as a reference to measure the positioning error of the actuator output shaft for at least one revolution of the actuator output shaft, and create error correction data.
The positioning error in the measured rotational positions are averaged in this instance, and error correction data representing the error-correction values for each of the rotational positions of one revolution of the motor rotating shaft can be created.
The error correction data can be made as a correction-pulse data string acting as error correction information for each of the rotational positions of one revolution of the motor rotating shaft, or can be made as a coefficient string for an approximation formula representing errors in each of the rotational positions of one revolution of the motor rotating shaft.
The present invention is also related to a drive controller of an actuator for suppressing variation in speed using the aforedescribed method, the drive controller being characterized by comprising a part for storing error correction data in which the error correction data is stored; a position detector attached to the motor rotating shaft; and a feedback-control part for performing feedback control so as to cause the actuator output shaft to arrive at a target position designated by position command information using, as a speed feedback value, a detection position of the motor rotating shaft detected by the position detector and a speed of the motor rotating shaft computed on the basis of the error correction data.
The feedback-control part may also perform feedback control so that the actuator output shaft arrives at a target speed designated by speed command information.
In the speed feedback values used for feedback control of the actuator in the present invention, a speed component caused by the angle transmission error of the wave gear drive is added to the speed computed from the detected position of the motor rotating shaft. Feedback is therefore used to control the speed, whereby variation in the rotational speed of the actuator output shaft caused by angle transmission error of the wave gear drive can be effectively suppressed.
An example of a drive controller of an actuator in which the method of the present invention is applied will be described below with reference to the drawings.
The drive controller 1 for controlling the driving of the actuator 2 comprises a feedback-control part 11 for performing feedback control so that the actuator 2 assumes a target position designated by a position command θr from a high-level device; and a correction-data storing part 12 in which is stored error correction data C representing position errors caused by angle transmission error of the wave gear drive 5. In the feedback-control part 11, a speed command θr is computed on the basis of the difference between the position command θr and a position feedback value θf from the position detector 6 in the position-controlling part 13 and is supplied to the speed-controlling part 14. In the speed-controlling part 14, a current command Ir is computed on the basis of the difference between the speed command ωr and a speed feedback value ωf supplied from a speed-computing part 15 and is supplied to a current-amplifying part 16. A motor-driving current is supplied to the motor 3 via the current-amplifying part 16.
In the speed-computing part 15, the position feedback value θf supplied from the position detector 6 is corrected on the basis of the correction data C, and the speed feedback value ωf is computed from the corrected value. The corrected value is determined by, e.g., adding or subtracting the correction data C relative to the position feedback value θf, and the speed feedback value ωf is computed from the corrected value. The equation for calculating the speed in this instance is as follows, where Δt is the sampling period.
ωf=((θf(t)+C(t))−(θf(t−1)+C(t−1)))/Δt
(Correction Data)
The correction data C of the present example is created as follows. First, the positioning error of the motor rotating shaft 4 in the actuator 2 is compressed by a factor of the reduction ratio of the wave gear drive 5 due to the reduction of the wave gear drive 5 connected to the motor rotating shaft. When the gear ratio of the wave gear drive 5 is, e.g., 50 or 100, the positioning error of the motor itself is compressed to 1/50 or 1/100, respectively. The positioning error of the actuator 2 is therefore primarily caused by angle transmission error of the wave gear drive 5, and therefore the positioning error of the actuator 2 in a single direction is determined by the angle transmission error of the wave gear drive 5.
Positioning is performed continuously in a single rotational direction, and the difference between the actual rotation angle and the intended rotation angle is determined for each position using reference positions. The single-direction positioning accuracy is the maximum of these values during course of one revolution.
The single-direction positioning accuracy of the actuator 2 in the present example, i.e., the positioning error of the actuator 2, is measured for one revolution of the actuator output shaft 7 on the basis of the absolute position of the motor rotating shaft 4. When the gear ratio of the wave gear drive 5 is, e.g., 1/50, the output shaft 7 rotates once for every 50 rotations of the motor rotating shaft 4.
The positioning error is measured for every, e.g., 3° of rotation of the motor rotating shaft 4 on the basis of the output of the position detector 6, as shown in
The measured data is substantially the same even when the angle transmission accuracy of the wave gear drive is used instead of the single-direction positioning accuracy.
It shall also be apparent that measurement may be performed not just for one revolution but for one or more revolutions of the actuator output shaft 7.
The error data for each measurement point, e.g., the error data for 120 locations, are averaged, and the correction data C for one revolution of the motor rotating shaft 4 is created as shown in the lowest row in the table in
The format of the created correction data C may be a correction pulse number for error correction for each of the rotational angle positions in one revolution of the motor rotating shaft 4. Correction pulse numbers measured for each, e.g., 3°, as above, can be put into a correspondence table for the allocations of rotational positions for every 3° of rotation of the motor rotating shaft 4, as shown in
Alternatively, the correction data for one revolution of the motor rotating shaft 4 may be expanded into a Fourier series and an approximation curve determined. The coefficients of the Fourier series represented by the approximation curve may be stored and saved as the correction data C. In this case, an initialization process may be performed when the driving current of the drive controller 1 is turned on, in which the stored coefficients are put into the approximation formula to calculate the correction data, and correction-pulse data strings are created as shown in
In the drive controller 1 of an actuator of the present example as described above, positioning error caused by angle transmission error of the wave gear drive 5 is measured in advance, and correction data C representing the positioning error in the rotational positions of the motor rotating shaft 4 is created. In the feedback control loop for controlling the motor, error correction values for the rotational positions are determined from the correction data C on the basis of the absolute position of the motor rotating shaft 4. The speed feedback value ωf is calculated while taking these error correction values into account, and the current command Ir for drive control of the motor is computed using this speed feedback value ωf. In the computed speed feedback value, the speed component caused by angle transmission error of the wave gear drive 5 is added to the rotational speed of the motor rotating shaft 4, and the speed is feedback controlled. Variations in the rotational speed of the actuator output shaft 7 caused by angle transmission error of the wave gear drive 5 can thereby be effectively suppressed.
Feedback control is performed on the basis of the position command θr from a high-level device in the present example, but feedback control may also be performed on the basis of the speed command ωr from a high-level device.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/JP2006/303984 | 3/2/2006 | WO | 00 | 8/12/2008 |
