JPH1189291A - Acceleration/deceleration control method of motor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress vibration without changing control rules with a conventional controller, by performing control by an acceleration time that is equal to or more than the shortest time being derived from the response time of a machine mechanism that is interlocked to the rotary shaft of a motor, and a deceleration time that is determined with reference to the natural frequency of the machine mechanism. SOLUTION: When a vibration amplitude stops in a process where it gradually increases in negative direction within a period Tβ of 2/4-3/4 of a period T that is determined by the natural frequency of a robot, the polarity of vibration (b) when deflection on deceleration returns differs from that of natural vibration (a). The amplitude of actual measurement synthetic vibration (C) after the stop becomes the difference between the vibration amplitude where the deflection returns and the natural vibration in the attenuation process, thus reducing the vibration on stop. An acceleration time Ta is set to the shortest response time or more of a robot, and the robot can be subjected to trapezoid or S-control and can be accelerated to a specific speed during the acceleration time Ta. Vibration being generated due to the inertia of a drive force transfer system when the robot stops is determined due to the control at a deceleration stage.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motor acceleration / deceleration control method, and more particularly to a motor acceleration / deceleration control method suitable for realizing vibration suppression when a multi-axis, articulated robot or the like is stopped.

[0002]

2. Description of the Related Art In recent years, industrial robots have been used to increase the efficiency of product processing and assembly operations. In the past, robots performed simple tasks, but recently,
Robots are becoming more multi-axis and multi-joint, and are now performing complex tasks.

However, when the robot is operated, vibrations are generated due to inertia of the driving force transmission system and the arm mechanism of the robot, and there is a problem that it takes much time to set the arm at a predetermined position. .

Therefore, conventionally, the following various methods have been proposed to suppress the vibration.

(1) Speed and current loops and gain are adjusted to set and operate a gain without vibration, or H-infinity control is used (Japanese Patent Laid-Open No. 5-397907).
Japanese Patent Application Laid-Open No. 7-79585) and the use of a neural network (Japanese Patent Application Laid-Open No. 6-35525) stabilize the control system to suppress vibration.

(2) By smoothing a command signal for driving a robot arm or the like in a teaching / reproducing robot, a change in acceleration is smoothed to suppress generation of vibration (Japanese Patent Laid-Open No. 4-362707).

(3) The deceleration pattern is formed into a curved shape for suppressing the vibration of the work (Japanese Patent Application Laid-Open No. 4-47415).

(4) The acceleration value when the sign of the acceleration difference is inverted on the acceleration diagram is set to 0 (JP-A-63-14).
1109).

[0009]

However, in the above-mentioned method (1), it is difficult to incorporate these control methods into an existing apparatus composed of a motor and an amplifier, which may lead to an increase in manufacturing cost. In addition, the method (2) cannot cope with rapid acceleration / deceleration, and may lead to an increase in operation time.

A common feature of the above methods (1) to (4) is that it is not possible to suppress vibration at the time of stoppage without increasing the acceleration / deceleration time using an existing servo amplifier. That is.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and a motor acceleration / deceleration control method capable of suppressing vibration without changing a control law with an existing robot controller or servo controller. The purpose is to provide.

[0012]

A motor acceleration / deceleration control method according to the present invention is directed to a motor acceleration / deceleration control method for a mechanical mechanism having a rotating or linearly moving shaft interlockingly connected to a rotating shaft of the motor. Is set to be equal to or longer than the shortest response time derived from the response speed of the mechanical mechanism, and the deceleration time is determined based on the natural frequency of the mechanical mechanism.

Accordingly, if the acceleration time is shorter than the shortest response time, the acceleration time ends before the mechanical mechanism starts accelerating, so that the mechanical mechanism can perform desired acceleration control. Can not. Therefore, by setting the acceleration time to be equal to or longer than the shortest response time, it is possible to cause the mechanical mechanism to perform desired acceleration control, and the mechanical mechanism is accelerated to a predetermined speed during the acceleration time.

On the other hand, a mechanical mechanism, such as a robot, having a rotating or straight-moving axis interlockingly connected to the rotation axis of the motor, when the operation is stopped, is caused by the inertia of the driving force transmission system of the robot and the arm mechanism. It is known that vibration occurs. How this vibration can be suppressed is determined by how to control the deceleration stage.
That is, the suppression of the vibration is highly dependent on the deceleration plan.

It has been found that the vibration is effectively suppressed by setting the deceleration time on the basis of the natural frequency of the mechanical mechanism. Specifically, the deceleration time is
It is desirable to set the period to be longer than about two quarters and shorter than about three quarters of the period determined by the natural frequency of the mechanical mechanism. Further, the deceleration time may be a time obtained by adding a time for n cycles to the set time.

In the above method, acceleration may be performed by trapezoidal control during acceleration, and deceleration may be performed by S-shaped control during deceleration. By performing trapezoidal control during acceleration, power consumption can be efficiently reduced, and by performing S-shaped control during deceleration, vibration suppression during stoppage can be made more effective.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment in which a motor acceleration / deceleration control method according to the present invention is applied to, for example, a stepping motor acceleration / deceleration control device attached to, for example, a six-axis robot (hereinafter simply referred to as a robot). Examples (hereinafter simply referred to as an acceleration / deceleration control device according to an embodiment) are shown in FIGS.
8B.

The acceleration / deceleration control device 10 according to the present embodiment
As shown in FIG. 1, motors for the six axes of the robot (X-axis motor 12x, Y-axis motor 12y, Z-axis motor 1
2z, U-axis motor 12u, V-axis motor 12v, and W-axis motor 12w), and driver groups (X-axis driver 14x, Y-axis driver 14y, Z
Axis driver 14z, U-axis driver 14u, V-axis driver 14v, and W-axis driver 14w), and a controller group (X-axis controller 1) for controlling each of the driver groups.
6x, Y axis controller 16y, Z axis controller 16
z, U axis controller 16u, V axis controller 16v
And a W-axis controller 16w), a program ROM 18 in which programs for acceleration / deceleration control and the like are registered, a data ROM 20 in which preset data and the like are stored, external data and data created by each controller, and the like. 22, an input / output port 26 for inputting / outputting data to / from external devices (another computer, input device (keyboard, mouse, etc.) 24, a network, etc.), and a CPU (CPU) for controlling these various circuits. (Control device and logical operation device) 28. In the following description, the CPU 28 in the acceleration / deceleration control device 10 is referred to as a host CPU 28 to distinguish it from the CPU 50 (see FIG. 2) in each controller 16.

The above-mentioned various circuits exchange data between the respective circuits via a data bus 30 derived from the host CPU 28, and further include an address bus and control bus (both not shown) derived from the host CPU 28. Each is controlled by the host CPU.

The acceleration / deceleration program registered in the program ROM 18 has a multitasking program configuration, and the acceleration / deceleration program is used to execute acceleration / deceleration processing for a six-axis motor group (to be precise, a signal for each controller). Process) is apparently performed at the same time.

Here, the configuration of one controller 16, for example, the X-axis controller 16x will be described with reference to FIG. The remaining controllers (16y, 16z,
16u, 16v and 16w) have a similar configuration.

As shown in FIG. 2, the X-axis controller 16x generates a drive pulse P for the X-axis motor 12x at a timing obtained by an acceleration / deceleration calculation, and outputs the drive pulse P to the X-axis motor 12x. A program ROM 40 in which (drive pulse processing program) is registered; a data ROM 42 in which necessary data and the like are stored in advance; a RAM 44 in which data from the outside and data created in the controller 16 are stored; Equipment (host CPU
28, limit switch 46, X-axis driver 14x, etc.)
It has an input / output port 48 for inputting and outputting data to and from the CPU, and a CPU (control device and logical operation device) 50 for controlling these various circuits.

The above-mentioned various circuits exchange data between the respective circuits via a data bus 52 derived from the CPU 50 and further via an address bus and a control bus (neither shown) derived from the CPU 50. CPU 50 each
It is configured to be controlled by.

The drive pulse processing means (drive pulse processing program) as software has three forms. The first form is, as shown in FIG. 3A, an acceleration time Ta.
The trapezoidal control is performed together with the deceleration time Tr and the acceleration time Ta and the deceleration time Tr as shown in FIG. 3B.
In the third embodiment, trapezoidal control is performed with an acceleration time Ta, and deceleration time Tr is performed, as shown in FIG. 3C.
Performs S-shaped control.

Selection of these modes is performed, for example, by the operator inputting a number indicating a mode selection instruction through the input device 24 (see FIG. 1). The input number is temporarily processed by the host CPU 28, and is output to the controllers (16u to 16z) as number data indicating the form.

The driving pulse processing means is provided in
As shown in the figure, the form number reading means 60 for reading form number data Dn from the host CPU 28, and the host CP
Based on the input of the load signal Sa from the U28, the acceleration time reading means 62 for reading out information Da on the acceleration time Ta from the teaching data Dt sent from the host CPU 28, and corresponds to the read acceleration time Ta. An acceleration counter 64 that counts the number of reference clocks and outputs an acceleration time end pulse Pa when the counting is completed;
First acceleration pulse rate creation means 68 for creating a pulse rate Rta required for trapezoidal control for acceleration from the read acceleration time Ta and sequentially developing the pulse rate Rta in an acceleration table 66; A second acceleration pulse rate creation means 70 for creating a pulse rate Rsa required for S-curve control for acceleration from the acceleration time Ta and sequentially developing the pulse rate Rsa in the acceleration table 66; Among the pulse rate creation means 68 and 70, an acceleration selector 72 for selecting the acceleration pulse rate creation means 68 or 70 corresponding to the form number read by the form number reading means 60, and the acceleration counter 64 When the reference pulse number corresponding to the preset constant speed time Tc is counted based on the input of the acceleration time end pulse Pa from The deceleration for reading information on the deceleration time Tr from the teaching data Dt sent from the host CPU 28 based on the input of the load signal Sa from the host CPU 28 based on the constant speed counter 74 that outputs the regular deceleration start pulse Ps. The time reading means 76 counts the number of reference clocks corresponding to the read deceleration time Tr, and when the counting is completed, the deceleration time end pulse Pr
And a first deceleration pulse rate for generating a pulse rate Rtr necessary for trapezoidal control for deceleration from the read deceleration time Tr and sequentially developing the pulse rate in the deceleration table 80. Means 82 and a pulse rate Rsr required for S-shaped control for deceleration from the read deceleration time Tr, and a deceleration table 80
Second deceleration pulse rate creating means 8 for sequentially developing the pulse rate
4, and among these first and second deceleration pulse rate generating means 82 and 84, the deceleration pulse rate generating means 82 or 84 corresponding to the form number read by the form number reading means 60 is selected. And a selector 86 for deceleration.

The driving pulse processing means includes a start pulse P from the host CPU 28 in addition to the various means.
an acceleration pulse output means 88 which is started based on the input of st and reads out the pulse rate Rta or Rsa sequentially from the acceleration table 66 and outputs the drive pulse P at a timing corresponding to the pulse rate Rta or Rsa; Constant speed pulse output means 90 which is started based on the input of the acceleration time end pulse Pa from the counter 64, reads out a constant speed pulse rate (fixed), and outputs a drive pulse P at a timing corresponding to the pulse rate. Is started directly by the input of the normal deceleration start pulse Ps from the constant speed counter 74 or indirectly by the input of the limit pulse Pm from the limit switch 46, and sequentially from the deceleration table 80 to the pulse rate Rtr or Rsr is read and each pulse rate Rtr or Rsr
And a counter 94 which counts the number of reference clocks corresponding to the input pulse rate and outputs a count reset signal Sr when the counting is completed. Having.

The processing by the acceleration pulse output means 88 is performed by the acceleration time end pulse Pa from the acceleration counter 64.
Or the limit pulse Pm from the limit switch 46, the processing is forcibly terminated, and the processing by the constant speed pulse output means 90 is performed by the constant speed counter 74.
Forcibly terminates based on the input of the normal deceleration start pulse Ps from the controller or the input of the limit pulse Pm from the limit switch 46.
Is forcibly terminated based on the input of the deceleration time end pulse Pr from the deceleration counter 78.

Next, the processing operation of the drive pulse processing means (drive pulse processing program) according to the present embodiment will be described with reference to the flowcharts of FIGS.

First, in step S1 of FIG. 5, the form number data Dn sent from the host CPU 28 is read through the form number reading means 60.

Next, in step S2, the acceleration pulse rate creating means (68 or 70) and the deceleration pulse rate creating means (68 or 70) corresponding to the read form number are passed through the acceleration selector 72 and the deceleration selector 86. 8
2 or 84) is selected. For example, when the form number indicating the first form is read, since both acceleration and deceleration are trapezoidal controls, the first and second selectors 72 and 86 determine the first form.
When the acceleration pulse rate creation means 68 and the first deceleration pulse rate creation means 82 are selected and the form number indicating the second form is read, both acceleration and deceleration are set to S.
Since the character control is used, the second acceleration pulse rate generation means 70 and the second deceleration pulse rate generation means 84 are selected by the selectors 72 and 86.

When the form number indicating the third form is read, the acceleration is trapezoidal control and the deceleration is S-shaped control. Is selected, and the second deceleration pulse rate creating means 84 is selected by the deceleration selector 86.

Next, in step S3, the host CP
It is determined whether or not the load signal Sa from the U28 has been input, and the step S3 is repeated until the load signal Sa is input. That is, it waits for the input of the load signal Sa.

When the load signal Sa is input, the process proceeds to the next step S4, in which the acceleration time Ta and the acceleration data Ta out of the teaching data Dt transmitted from the host CPU 28 through the acceleration time reading means 62 and the deceleration time reading means 76. Information on the deceleration time Tr is read out.

Next, in step S5, of the first and second acceleration pulse rate generating means 68 and 70,
From the read acceleration time Ta, the pulse rate Rta or Rsa required for trapezoidal control or S-shaped control for acceleration is sequentially created through the acceleration pulse rate creation means 68 or 70 selected by the acceleration selector 72. Then, the created pulse rate Rta or Rsa is stored in the acceleration table 66 in chronological order to complete the acceleration table 66.

Next, in step S6, of the first and second deceleration pulse rate generating means 82 and 84,
From the read-out deceleration time Tr, the pulse rate Rtr or Rsr required for trapezoidal control or S-shaped control for deceleration is sequentially generated from the read deceleration time Tr through the deceleration pulse rate generation means 82 or 84 selected by the deceleration selector 86. Then, the created pulse rate Rtr or Rsr is stored in the deceleration table 80 in chronological order to complete the deceleration table 80.

Next, in step S7, an initial value "0" is stored in an index register i used as a read index of a record for the acceleration table 66, and the index register i is initialized.

Next, in step S8, the host CP
It is determined whether or not the start pulse Pst from U28 has been input, and until the start pulse Pst is input,
Step S8 is repeated. That is, the start pulse Ps
It waits for the input of t.

When the start pulse Pst is input, the process proceeds to the next step S9, and the reference clock number corresponding to the read acceleration time Ta is sent to the acceleration counter 64 via the acceleration time reading means 62. Store,
Further, it outputs an enable signal to the acceleration counter 64. The acceleration counter 64 counts (decrements) the reference clock number based on the input of the enable signal. When the count value becomes 0, the acceleration counter 64 outputs an acceleration time end pulse Pa.

Next, in step S10, the acceleration table 6 is output through the acceleration pulse output means 88.
6, the pulse rate Rta or Rsa is read from the record (i-record) indicated by the value of the index register i.

Next, in step S11, the read reference clock number corresponding to the pulse rate Rta or Rsa is stored in the counter 94 through the acceleration pulse output means 88. Outputs an enable signal. The counter 94 counts (decrements) the reference clock number based on the input of the enable signal. When the count value becomes 0, the counter 94 outputs a count reset signal Sr.

Next, in step S12, a pulse output subroutine is entered. In the pulse output subroutine, as shown in the flowchart of FIG. 7, first, in step S101, one drive pulse P is applied to the X-axis motor 12 via the input / output port 48 and the X-axis driver 14x.
Output to x. Then, in step S102, it is determined whether or not the counter 94 has output the count reset signal Sr, and the step S102 is repeated until the count reset signal Sr is input. That is, it waits for the input of the count reset signal Sr.

The pulse output subroutine ends when the count reset signal Sr is input. That is, this pulse output subroutine performs a process of delaying by a time corresponding to the pulse rate after outputting one drive pulse P.

When the pulse output subroutine is completed, the process proceeds to the next step S13 shown in FIG. In this step S13, it is determined whether or not the limit pulse Pm from the limit switch 46 has been input.
If it is determined that the limit pulse Pm has not been input, the process proceeds to the next step S14, where the limit pulse Pm
When it is determined that is input, the process proceeds to a process after step S21 in FIG. 6 described later, that is, a pulse output process for deceleration.

In step S14, it is determined whether or not the acceleration time end pulse Pa has been input from the acceleration counter 64. If it is determined that the acceleration time end pulse Pa has not been input, the process proceeds to step S15, in which the value of the index register i is updated by +1 to indicate the next record of the acceleration table 66, and then the process proceeds to step S10. Then, the process returns to step S10 and thereafter.

By repeating the processing of steps S10 to S15, trapezoidal control or S-shaped control for acceleration is performed on the X-axis motor 12x.

On the other hand, if it is determined in step S14 that the acceleration time end pulse Pa has been input, the process proceeds to step S16 in FIG. 6 and is set in advance through the constant speed pulse output means 90. The reference clock number corresponding to the constant speed time (fixed) is stored in the constant speed counter 74, and an enable signal is output to the constant speed counter 74 to start counting by the constant speed counter 74.

Next, in step S17, a reference clock number corresponding to the constant speed pulse rate (fixed) is counted by the counter 94 through the constant speed pulse output means 90.
And outputs an enable signal to the counter 94 to start counting by the counter 94.

Next, in step S18, a pulse output subroutine is entered. In this pulse output subroutine, a process is performed in which one drive pulse P is output and then delayed by a time corresponding to the pulse rate for fixed speed (fixed).

When the pulse output subroutine ends, the flow advances to the next step S19. In this step S19, it is determined whether or not the limit pulse Pm from the limit switch 46 has been input. When it is determined that the limit pulse Pm has not been input, the process proceeds to the next step S20, and when it is determined that the limit pulse Pm has been input, the process after step S21 described later, that is, for deceleration, Proceed to pulse output processing.

In step S20, it is determined whether or not the regular deceleration start pulse Ps from the constant speed counter 74 has been input. Regular deceleration start pulse Ps
If it is determined that has not been input, the process returns to step S17, and the processes after step S17 are repeated. By repeating the processing of steps S17 to S20, control for constant speed is performed on the X-axis motor 12x.

If it is determined in step S13 or step S19 that the limit pulse Pm has been input, or if it is determined in step S20 that the normal deceleration start pulse Ps has been input, Proceeding to step S21, the read-out deceleration time Tr is output through the deceleration time reading means 76.
Is stored in the deceleration counter 78, and an enable signal is output to the deceleration counter 78. The deceleration counter 78 counts (decrements) the reference clock number based on the input of the enable signal. When the count value becomes zero, the deceleration counter 78 outputs a deceleration time end pulse Pr.

Next, in step S22, an initial value "0" is stored in an index register j used as a read index of a record for the deceleration table 80, and the index register j is initialized.

Next, in step S23, the deceleration table 8 is output through the deceleration pulse output means 92.
From among the 0 records, the pulse rate Rtr or Rsr is read from the record (j record) indicated by the value of the index register j.

Next, at step S24, the read reference clock number corresponding to the read pulse rate Rtr or Rsr is stored in the counter 94 through the deceleration pulse output means 92, and the counter 94 is enabled. A signal is output to start counting by the counter 94.

Next, in step S25, a pulse output subroutine is entered. In this pulse output subroutine, a process of outputting one drive pulse P and delaying by a time corresponding to the pulse rate Rtr is performed.

When the pulse output subroutine ends, the flow advances to the next step S26. In this step S26, it is determined whether or not the deceleration time end pulse Pr from the deceleration counter 78 has been input. If it is determined that the deceleration time end pulse Pr has not been input, the process proceeds to step S27, where the value of the index register j is updated by +1 to indicate the next record of the deceleration table 80, and then the process proceeds to step S23. Then, the process returns to step S23 and thereafter.

By repeating the processing of steps S23 to S27, trapezoidal control or S-shaped control for deceleration is performed on the X-axis motor 12x.

On the other hand, if it is determined in step S26 that the deceleration time end pulse Pr has been input, the process proceeds to the next step S28, where it is determined whether a program end request has been issued. This determination is made when the power is turned off.
This is performed depending on whether or not an end request interrupt such as an interrupt has occurred.

If there is no program end request, the flow returns to step S7 in FIG. 5 to wait for the next start request from the host CPU 28. If there is a program end request, this drive pulse processing program ends. Will be.

In the acceleration / deceleration control device 10 according to this embodiment, the acceleration time Ta is set to be equal to or longer than the shortest response time derived from the response speed of the robot, and the deceleration time Tr is determined based on the natural frequency of the robot. .

That is, if the acceleration time Ta is set to be shorter than the shortest response time, the acceleration time Ta ends before the robot starts accelerating, and the robot cannot perform desired trapezoidal control. . Therefore, by setting the acceleration time Ta to be equal to or longer than the shortest response time, it is possible to cause the robot to perform desired trapezoidal control or S-shaped control, and the robot accelerates to a predetermined speed during the acceleration time Ta. Is done.

On the other hand, it is known that when the robot stops operating, vibration is generated due to the inertia of the driving force transmission system and the arm mechanism of the robot. How this vibration can be suppressed is determined by how to control the deceleration stage. That is, the suppression of the vibration is highly dependent on the deceleration plan.

By setting the deceleration time Tr based on the natural frequency of the robot, the vibration can be effectively suppressed. Specifically, the deceleration time Tr is
It is desirable to set the period to be longer than about two quarters and shorter than about three quarters of the period T determined by the natural frequency of the robot. In this case, the time (nT + (2T / 4) obtained by adding the time (nT) for n cycles to the set time (2T / 4 to 3T / 4)
~ 3T / 4)).

Here, the deceleration time Tr is set to a time shorter than about one-fourth of the period T determined by the natural frequency of the robot (0 to T).
/ 4) (or a time obtained by adding a time corresponding to n cycles to the time (nT + (0 to T / 4))) (comparative example:
The vibration shown in broken line a in FIG. 3A) and the deceleration time Tr are longer than about two-quarters of the period T determined by the natural frequency of the robot.
A time shorter than about three quarters (2T / 4 to 3T / 4) (or a time obtained by adding a time corresponding to n cycles to the time (nT + (2
T / 4 to 3T / 4)))) (Example: FIG. 3
(See the solid line b of A)) was confirmed by a simulation based on the first embodiment. The results are shown in FIGS. 8A to 9B.

As shown in FIGS. 8A and 8B, the natural vibration of the robot from the deceleration start time t0 (indicated by a two-dot chain line a)
Is assumed to gradually decrease, a period Tα of a quarter of the period T determined by the natural frequency of the robot is a process in which the vibration amplitude gradually increases in the positive direction, as in the comparative example. When stopped within this period Tα, the polarity of the vibration (indicated by a broken line b) when the deflection at the time of deceleration (the deflection of the arm or the like generated during the deceleration) returns is the same as the polarity of the natural vibration. As a result, the synthetic vibration actually measured after the stop (solid line c)
(Indicated by) is the sum of the amplitude of the vibration in which the deflection returns and the amplitude of the natural vibration in the damping process, and it can be seen that the vibration at the time of stoppage becomes very large. Therefore, a period Tα of a quarter of the cycle can be defined as a danger stop time zone.

On the other hand, as shown in FIGS. 9A and 9B, during the period Tβ of 2/4 to 3/4 of the period T determined by the natural frequency of the robot, the vibration amplitude gradually increases in the negative direction. As in the embodiment, when the vehicle stops within this period Tβ, the vibration at the time when the deflection at the time of deceleration returns (shown by a broken line b)
Is different from the polarity of the natural vibration (indicated by the two-dot chain line a). As a result, the amplitude of the composite vibration (indicated by the solid line c) actually measured after the stoppage is reduced to the amplitude of the vibration in which the deflection returns. It can be seen that the amplitude of a certain natural vibration has been subtracted, and the vibration at the time of stopping is reduced. Therefore, a period Tβ of two-fourths to three-fourths of the cycle can be defined as a safe stop time zone.

Therefore, the natural frequency of the robot is measured,
At the time of the deceleration control, the deceleration end time point te is a period (2T / 4 to 3
T / 4) (or a period obtained by adding a cycle to the period n times (n
By selecting the deceleration time Tr so as to be included in T + (2T / 4 to 3T / 4))), it is possible to efficiently suppress the vibration generated when deceleration is stopped.

As described above, since the deceleration time Tr is set to a relatively long time (nT + (2T / 4 to 3T / 4)), the entire acceleration / deceleration time (acceleration time Ta + constant speed time Tc) is set.
Although the acceleration time Ta may be longer, the acceleration time Ta is set to be equal to or longer than the shortest response time derived from the response speed of the robot. It is possible to suppress the time from becoming long (see FIG. 3A).

By the way, when the second mode is selected, as shown in FIG. 3B, the S-shaped control is performed for both the acceleration time Ta and the deceleration time Tr. This is more effective in suppressing vibration.
Therefore, in the second embodiment, it is possible to improve the suppression of vibration during deceleration stop.

Also, in the second embodiment, as described above, the deceleration end time te is a period (2T / 4 to 2/4) of the period T determined by the natural frequency of the robot.
3T / 4) (or a period (nT + (2T / 4 to 3T / 4)) obtained by adding a cycle to the period n times), so that the deceleration time Tr is selected. The generated vibration can be suppressed more efficiently.

Next, in the third embodiment, as shown in FIG. 3C, trapezoidal control is performed during the acceleration time Ta, and S-shaped control is performed during the deceleration time Tr.

In general, trapezoidal control requires less motor current and instantaneous current during acceleration than S-shaped control, and S-shaped control is more effective in suppressing vibration than trapezoidal control. Therefore, in the third embodiment, it is possible to efficiently reduce power consumption, use a motor having a small capacity, and further improve vibration suppression during deceleration stop.

Next, the operation of the acceleration / deceleration control device 10 according to the present embodiment until the acceleration time Ta and the deceleration time Tr are determined will be described with reference to the flowchart of FIG.

First, in step S201, the form of the acceleration / deceleration control is selected. First, the first mode, that is, the mode in which trapezoidal control is performed with the acceleration time Ta and the deceleration time Tr, is selected.

Next, in step S202, an acceleration time Ta is set. As the acceleration time Ta, a time longer than the shortest response time derived from the response speed of the robot is selected. Specifically, teaching data indicating, for example, a movement of 10 cm is supplied to the acceleration / deceleration control device 10 in advance, and the time from when the start button is pressed until the robot actually starts moving is electrically counted by a counter. Is performed a plurality of times, and a time longer than the shortest response time is adopted as the acceleration time Ta.

Next, in step S203, the deceleration time Tr is temporarily determined. In this case, a normally used deceleration time Tr is selected.

Next, in step S204, the robot is actually operated using the acceleration / deceleration control device 10, and the natural frequency of the decelerating robot (the natural frequency determined by the load capacity) is measured (FIG. 9). 9B).

Next, in step S205, the multiplier n for the period T determined by the measured natural frequency is set to 1
And

Next, in step S206, the deceleration time Tr is set to a period (nT + 2T / 4) obtained by adding two quarters of the period to n periods.

Next, in step S207, the robot is actually operated using the acceleration / deceleration control device 10 to measure the attenuation state of the vibration of the robot after the deceleration stop.

Next, in step S208, it is determined whether or not the damping state satisfies a prescribed condition (for example, the vibration amplitude when 0.5 second has elapsed from the time of deceleration stop is 1 mm or less). If the damping state does not satisfy the specified condition, the process proceeds to step S209, and the deceleration time Tr is set to the time α.
The time is extended by a minute to obtain a new deceleration time Tr. The time α is a time sufficiently shorter than a quarter period of the cycle, and for example, a time period of 1/40 of the cycle can be selected.

Next, in step S210, the n
It is determined whether or not the time R added to the cycle is equal to or more than a quarter of the cycle. That is, the deceleration time Tr is a period obtained by adding a period of 3/4 of the period to the n period (nT + 3T / 4).
Is determined.

When the deceleration time Tr is n cycles, it is 3/4 of the cycle.
If the time (nT + 3T / 4) obtained by adding the period has not exceeded, the process returns to step S207 again, and the operation after step S207 is performed.

On the other hand, when the deceleration time Tr becomes n cycles,
If the time obtained by adding the period of 3 / min (nT + 3T / 4) is exceeded, the multiplier n for the cycle is updated by +1 (step S211), and in the next step S212, the multiplier n is equal to or more than the specified upper limit value m. Is determined.
The upper limit value m can be selected in consideration of the man-hour for setting the deceleration time Tr. For example, 4 is selected. The same applies to the time α added in step S209, and the resolution can be selected in consideration of the man-hour for setting the deceleration time Tr.

If the multiplier n is less than the upper limit value m, the process returns to step S206 again, and the operations after step S206 are repeated. If the multiplier n is equal to or more than the upper limit value m, the process returns to step S201 again, and The form of the acceleration / deceleration control is selected, and the operation after step S201 is repeated.
In the step S201, the second mode is selected in the second mode, that is, the mode in which the S-shaped control is performed for both the acceleration time Ta and the deceleration time Tr, and the third mode is the third mode.
That is, trapezoid control is performed during the acceleration time Ta, and the deceleration time Tr
Is selected for the S-shaped control. The order of the selection can be determined appropriately.

Then, in step S208,
If it is determined that the damping state satisfies the prescribed condition, the process proceeds to step S213, where the deceleration time Tr is determined, and in the next step S214, the acceleration time Ta is reset if necessary. This corresponds to step S21
If the deceleration time Tr determined in 3 is a relatively long time, the entire acceleration / deceleration time becomes long, and there is a possibility that drive control for the robot becomes slow. To prevent this, the acceleration time Ta is reset so that the entire acceleration / deceleration time does not substantially increase. Also in this resetting, a time longer than the shortest response time derived from the response speed of the robot is selected.

[0088]

EXAMPLE 1 As shown in FIG. 11, a 6-axis portal robot 100 was set on a table 102, and a V-axis was operated to activate a tool tip 10.
4 moves along a relatively long path along the X-axis direction, the vibration of the tool tip 104, in particular, the deceleration end point te
0.5 second after the vibration amplitude A was measured using the laser displacement meter 106. The control mode of acceleration / deceleration with respect to the V axis used at this time is as follows, and the payload is 80 kgf in each case.

(1) First comparative example: As shown in FIG. 12A, both the acceleration time Ta and the deceleration time Tr are set to trapezoidal control.
The acceleration time Ta was set to 0.25 seconds, and the deceleration time Tr was set to the time (here, 0.25 seconds) obtained by adding a period less than one quarter of the period to the period of the natural frequency determined by the load capacity. .

(2) Second comparative example: As shown in FIG. 13A, both the acceleration time Ta and the deceleration time Tr are S-shaped,
The acceleration time Ta was set to 0.25 seconds, and the deceleration time Tr was set to the time (here, 0.25 seconds) obtained by adding a period less than one quarter of the period to the period of the natural frequency determined by the load capacity. .

(3) First embodiment: As shown in FIG. 14A, the acceleration time Ta is trapezoidal control, the deceleration time Tr is S-shaped control, the acceleration time Ta is 0.20 seconds, and the deceleration time Tr is portable. 2/4 of the period of the natural frequency determined by weight
The time obtained by adding the period of less than 3/4 (here, 0.3
0 second).

FIGS. 12B and 13 show the measurement results of the vibration amplitude A.
B and FIG. 14B. In the first comparative example, as shown in FIG. 12B, the vibration amplitude A is 3.9 mm, and the second
In the comparative example of FIG. 13, as shown in FIG.
Was 4.2 mm. On the other hand, in the first embodiment, FIG.
As shown in B, the vibration amplitude A is 0.64 mm, and it can be seen that the vibration amplitude A is much smaller than in the first and second comparative examples.

Embodiment 2 As shown in FIG. 15, a 6-axis portal robot 100 is set on a table 102, and the Y-axis is operated to activate a tool tip 10
When the tool 4 is moved along a relatively short path along the Y-axis direction, the vibration of the tool tip 104, particularly the deceleration end point te
0.5 second after the vibration amplitude A was measured using the laser displacement meter 106. The control mode of acceleration / deceleration with respect to the Y axis used at this time is as follows, and the payload is 80 kgf in each case.

(1) Third Comparative Example: As shown in FIG. 16A, both the acceleration time Ta and the deceleration time Tr are set to trapezoidal control.
The acceleration time Ta is 0.20 seconds, and the deceleration time Tr is a period obtained by adding a period less than a quarter of the period T to the period T of the natural frequency determined by the load capacity (here, 0.20 seconds). And

(2) Fourth comparative example: As shown in FIG. 17A, both the acceleration time Ta and the deceleration time Tr are controlled in an S-shape.
The acceleration time Ta is 0.20 seconds, and the deceleration time Tr is a period obtained by adding a period less than a quarter of the period T to the period T of the natural frequency determined by the load capacity (here, 0.20 seconds). And

(3) Second embodiment: As shown in FIG. 18A, the acceleration time Ta is trapezoidal, the deceleration time Tr is S-shaped, the acceleration time Ta is 0.16 seconds, and the deceleration time Tr is portable. The period (0.24 seconds in this case) was obtained by adding a period of 2/4 to less than 3/4 of the period T to the period T of the natural frequency determined by the weight.

FIGS. 16B and 17 show the measurement results of the vibration amplitude A.
B and FIG. 18B. In the third comparative example, the vibration amplitude A was 13.1 mm as shown in FIG. 16B, and in the fourth comparative example, the vibration amplitude A was 4.0 mm as shown in FIG. 17B. . On the other hand, in the second embodiment, as shown in FIG. 18B, the vibration amplitude A is 0.5 mm, and the vibration amplitude A is very small as compared with the third and fourth comparative examples. I understand.

Acceleration / deceleration control device 10 according to this embodiment
Has shown an example applied to an acceleration / deceleration control device of a stepping motor attached to a robot, for example.
The present invention can also be applied to an acceleration / deceleration control device for a DC servomotor or an AC servomotor.

It should be noted that the motor acceleration / deceleration control method according to the present invention is not limited to the above-described embodiment, but may employ various configurations without departing from the spirit of the present invention.

[0100]

As described above, according to the motor acceleration / deceleration control method according to the present invention, the motor acceleration / deceleration control method for a mechanical mechanism having a rotating or straight-moving shaft operatively connected to the rotating shaft of the motor is provided. , The acceleration time is set to be equal to or longer than the shortest response time derived from the response speed of the mechanical mechanism, and the deceleration time is determined based on the natural frequency of the mechanical mechanism.

Therefore, the vibration can be suppressed without changing the control law with the existing robot controller or servo controller.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an acceleration / deceleration control device according to the present embodiment.

FIG. 2 is a block diagram showing a configuration of a controller incorporated in the acceleration / deceleration control device according to the embodiment.

FIG. 3A is a speed sequence diagram showing acceleration / deceleration control according to a first mode of the drive pulse processing means, and FIG. 3B is a speed sequence diagram showing acceleration / deceleration control according to the second mode of the drive pulse processing means. FIG. 3C is a speed sequence diagram showing acceleration / deceleration control according to the third embodiment of the drive pulse processing means.

FIG. 4 is a functional block diagram showing a drive pulse processing unit according to the embodiment.

FIG. 5 is a flowchart (part 1) illustrating a processing operation of a drive pulse processing unit according to the present embodiment;

FIG. 6 is a flowchart (part 2) illustrating a processing operation of the drive pulse processing means according to the present embodiment.

FIG. 7 is a flowchart illustrating a processing operation of a pulse output subroutine.

FIG. 8A is a speed sequence diagram in a comparative example, and FIG. 8B is a waveform diagram showing a damping state of vibration after deceleration stop in the comparative example.

FIG. 9A is a speed sequence diagram in the embodiment, and FIG. 9B is a waveform diagram showing a damping state of vibration after deceleration stop in the embodiment.

FIG. 10 is a flowchart showing an operation until the acceleration time and the deceleration time are determined in the acceleration / deceleration control device according to the present embodiment.

FIG. 11 is an explanatory diagram showing an example of an experimental apparatus for measuring a vibration amplitude after deceleration stop in Example 1.

FIG. 12A is a speed sequence diagram showing acceleration / deceleration control in a first comparative example, and FIG. 12B is a waveform diagram showing a vibration state in the first comparative example.

FIG. 13A is a speed sequence diagram showing acceleration / deceleration control in a second comparative example, and FIG. 13B is a waveform diagram showing a vibration state in the second comparative example.

FIG. 14A is a speed sequence diagram showing acceleration / deceleration control in the first embodiment, and FIG. 14B is a waveform diagram showing a vibration state in the first embodiment.

FIG. 15 is an explanatory diagram showing an example of an experimental apparatus for measuring the vibration amplitude after deceleration stop in Example 2.

FIG. 16A is a speed sequence diagram showing acceleration / deceleration control in a third comparative example, and FIG. 16B is a waveform diagram showing a vibration state in the third comparative example.

FIG. 17A is a speed sequence diagram showing acceleration / deceleration control in a fourth comparative example, and FIG. 17B is a waveform diagram showing a vibration state in the fourth comparative example.

FIG. 18A is a speed sequence diagram showing acceleration / deceleration control in the second embodiment, and FIG. 18B is a waveform diagram showing a vibration state in the second embodiment.

[Explanation of symbols]

10. Acceleration / deceleration control device 12u-12
z ... motor 14u-14z ... driver 16u-16
z: Controller 60: Form number reading means 62: Acceleration time reading means 64: Acceleration counter 66: Acceleration table 68, 70: First and second acceleration pulse rate creation means 72: Acceleration selector 74: Constant speed Counter 76: deceleration time reading means 78: deceleration counter 80: deceleration tables 82, 84: first and second deceleration pulse rate creation means 86: deceleration selector 88: acceleration pulse output means 90: constant speed Pulse output means 92: deceleration pulse output means

Claims (4)

[Claims] 1. A method for controlling the acceleration and deceleration of a motor in a mechanical mechanism having a rotating or straight-moving axis operatively connected to a rotating shaft of the motor, wherein an acceleration time is equal to or longer than a shortest response time derived from a response speed of the mechanical mechanism. The acceleration / deceleration control method for a motor, wherein the deceleration time is determined based on a natural frequency of the mechanical mechanism. 2. The motor acceleration / deceleration control method according to claim 1, wherein the deceleration time is longer than about two-fourths and shorter than about three-fourths of a period of a natural frequency of the mechanical mechanism. A motor acceleration / deceleration control method characterized by being set. 3. The motor acceleration / deceleration control method according to claim 2, wherein the deceleration time is a time obtained by adding a time for n cycles to the set time. 4. The motor acceleration / deceleration control method according to claim 1, wherein acceleration is performed by trapezoidal control during acceleration, and deceleration is performed by S-shaped control during deceleration. Acceleration / deceleration control method.