JP3211406B2 - Vibration suppression control method and velocity pattern calculation method and apparatus using the same - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vibration control method,
In particular, the present invention relates to a vibration suppression control method suitable for suppressing vibration during positioning of a robot arm used for an automatic assembling machine or the like.

[0002]

2. Description of the Related Art FIG. 5 is a schematic configuration diagram of a general robot control system. As shown in the figure, a finger portion 5 for handling a workpiece is attached to a tip of a robot arm 4, and the robot arm 4 is driven to rotate and move up and down by a servomotor 3.
The servo motor 3 is driven from the controller 1 via the servo amplifier 2.

In the configuration described above, the operation of the robot arm 4 will now be described with reference to FIG.
This will be described according to the timing chart of FIG.

Now, the robot arm 4 is moved from position a to position c.
To be driven. In this case, the controller 1 gives a velocity command as shown in FIG. 6B to the servo motor 3 so that the robot arm 4 changes the position as shown in FIG. 6A.

In this case, ideally, the robot arm 4 together with the finger portion 5 accelerates from the time t1 to the time t2, has a constant speed from the time t2 to the time t4, and decelerates from the time t4 to the time t5. Position a while changing its speed
Change the rotation position from to the position c.

On the other hand, when the robot arm 4 is driven from the position a to the position b, the controller 1 similarly controls the servo motor 3 so that the robot arm 4 changes its position as shown in FIG. , A speed command as shown in FIG.

In this case, ideally, the robot arm 4 accelerates together with the finger portion 5 from time t1 to time t3, decelerates from time t3 to time t5, and constantly changes its speed from position a to position b. Change the rotation position until.

However, in practice, since inertial force and elastic force act on the robot arm 4 and the finger portion 5 driven by the servomotor 3, the acceleration, the end of the acceleration, the deceleration, and the end of the deceleration are different. In this case, vibration occurs in the robot arm 4 and the finger unit 5. As a result, as shown in FIGS. 7C and 7F, vibrations accompanying acceleration / deceleration occur in the finger portion 5 at the tip of the robot arm 4, and particularly when the positioning is stopped at the positions b and c. There is a problem that does not fit.

As described above, in a robot or the like, movable parts such as the robot arm 4 and the finger part 5 driven in response to an operation command from the controller 1 vibrate during acceleration / deceleration, at the start of movement, or at the time of stop. This is due to the inertial force or elastic force of the movable part, and becomes more conspicuous as the operating speed or the acceleration / deceleration speed is increased in order to increase the speed. This vibration is particularly problematic when positioning is stopped,
Since the next operation cannot be started until the vibration stops, the operation speed is substantially reduced.

Research for suppressing such vibration during positioning of the robot has been widely conducted in the form of vibration suppression control. For example, JP-A-63-314606 and JP-A-6-314606
The device shown in JP-A-3-314607 is an example of this, but in each case, after the vibration is generated at the tip of the robot arm, the signal of the acceleration sensor attached to the tip of the robot arm is fed back as disturbance torque, and the acceleration is reduced. It is intended to control the vibration by controlling.

[0011]

Since the conventional vibration suppression control method for a robot or the like is performed by the feedback control as described above, the vibration is actually generated before the vibration is started. Even if the time required for the vibration to converge can be reduced, it cannot be reduced to zero. In addition, since control information for converging the vibration is grasped from the vibrating state, vibration control cannot be performed unless vibration occurs, and it cannot be said that vibration control is complete. In addition, in order to follow the vibration and suppress the vibration, a CPU capable of high-speed calculation is required, and there is a problem that the vibration control system is expensive and complicated.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems of the prior art. In particular, when the robot arm is moved for a short distance by successively performing an accelerating operation and a subsequent decelerating operation of the robot arm, the position of the tip of the robot arm or the like is determined. By learning the vibration associated with the robot in advance, adding a condition to cancel the vibration to the control command of the robot arm, and correcting the speed control curve, it is possible to suppress the occurrence of vibration when the robot arm stops. It is another object of the present invention to provide a vibration damping control method.

[0013]

In order to achieve the above object, the present invention provides a method of calculating a speed pattern of a moving part of an industrial robot which can suppress vibration of the moving part. That is, the speed pattern of the moving part is obtained, and the torque peak generated during the acceleration period of the speed pattern is obtained.
Time difference between the torque peaks occurring click and the deceleration period, a method of modifying the speed pattern so as to be approximately an integral multiple of the natural vibration period of the moving part. Further, according to another aspect of the present invention, there is provided a moving part control method capable of suppressing vibration of a moving part of an industrial robot. That is, when the moving part accelerates the moving part
Time difference between the torque peaks occurring when the deceleration torque peaks occurring in it, is controlled to be approximately an integral multiple of the natural vibration period of the moving part. According to still another aspect of the present invention, there is provided an apparatus for calculating a speed pattern for suppressing generation of vibration of a moving part of an industrial robot. That is, the apparatus includes means for determining the speed pattern of the moving part, generates the acceleration period of the speed pattern
Has means for calculating a time difference of torque peaks occurring in the torque peak and the deceleration period, and means for the time difference to correct the speed pattern so as to be approximately an integral multiple of the natural vibration period of the moving part, the I have.

[0014]

By making the time difference between the torque peaks generated during the acceleration period and the deceleration period coincide with the natural oscillation period of the moving part,
By canceling the vibration generated during the acceleration period by the vibration generated during the deceleration period, the vibration of the moving part can be suppressed.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a flowchart of a vibration suppression control method according to one embodiment of the present invention. As shown in the drawing, steps S1 and S2 are set as advance preparations. In step S1, a natural frequency f of a movable portion such as a robot arm or a finger portion of a robot is measured, and in step S2, The free vibration cycle of the movable part, that is, the natural vibration cycle λ is calculated. Incidentally, this calculation is performed by finding the reciprocal of the natural frequency f.

The following steps S3 to S8 are set as controls when the robot arm is actually operated. First, in step S3, the speed pattern of the movement of the tip of the robot arm is calculated (FIG. 3).
(See (a)), and in step S4, a time difference T between the torque peaks of the servomotor is calculated (see FIG. 3B). In the next step S5, the torque peak time difference T is extra calculated with the natural oscillation period λ. This is performed by dividing the torque peak time difference T by the natural oscillation period λ. Here, the division result n and the remainder T ′ are obtained. In step S6, it is determined whether the remainder T ′ is close to “0 or λ”. If the remainder T 'is close to "0 or [lambda]", the process proceeds to step S8 while keeping the torque peak time difference T as it is, and the servo motor is driven. On the other hand, the remainder T '
Is not near “0 or λ”, the speed pattern is corrected so that the torque peak time difference T becomes T3 = (n + 1) λ, and in step S8, the servo motor is driven based on this.

FIG. 4 shows the velocity waveform at the tip, the current waveform (= torque) of the servomotor, and the acceleration waveform in the movement direction at the tip of the robot arm when the moving distance of the robot arm is short. Incidentally, FIG. 3A shows a velocity waveform when the moving distance is 1, FIG. 3B shows a motor current when the moving distance is 1, FIG. 3C shows a tip acceleration when the moving distance is 1, and FIG. (D) is a velocity waveform when the moving distance is 4 × l, (e) is a motor current when the moving distance is 4 × l,
FIG. 6F shows the acceleration of the tip when the moving distance is 4 × 1.

Assuming that the acceleration and deceleration are constant, the speed change is small as shown in FIG. 4A for the moving distance l, and as shown in FIG. 4D for the moving distance 4 × 1. As shown, the speed change is large. On the other hand, the motor current becomes positive during acceleration and negative during deceleration, and the torque changes accordingly. However, in the case of the moving distance l, the acceleration time and the deceleration time are short. 4B, ie, the torque peak time difference T becomes T1 as shown in FIG. 4B, and when the moving distance is 4 × 1, both the acceleration time and the deceleration time become longer, and the torque peak becomes longer as shown in FIG. The time difference T becomes T2.

In this case, the acceleration of the distal end of the robot arm changes according to the motor current, that is, the torque at the time of acceleration or deceleration, but vibrates at the time of stoppage due to the inertial force or elastic force of the robot arm. The vibration cycle in this case is shown in FIG.
As shown in (c) and (f), it is determined by the natural frequency f of the movable part including the robot arm regardless of the acceleration at the time of movement.

Incidentally, the natural frequency f of the robot arm
And the natural oscillation period λ is λ = 1 / f (1) Here, the relationship between the torque peak time difference T1 and the natural oscillation period λ when the moving distance is 1 is T1 = λ / 2 (2), and the torque peak time difference T when the moving distance is 4 × 1.
It is assumed that the relationship between 2 and the natural oscillation period λ is T2 = λ (3). In this case, as is clear from FIGS. 4C and 4F, the residual vibration at the acceleration of the tip of the robot arm is smaller at the moving distance of 4 × 1 where the torque peak time difference T2 = λ. I understand. That is,
The torque peak time difference T of the motor current is related to the magnitude of the residual vibration when the tip of the robot arm stops.

More specifically, the motor current waveform has two peaks in both positive and negative directions due to acceleration and deceleration of the robot arm. When the time difference, that is, the torque peak time difference T is T (= λ / 2), the vibration of the robot arm generated at the time of acceleration is increased at the time of deceleration, so that the residual vibration is large. In other words, the acceleration / deceleration is the same as the excitation at the frequency f, and the natural frequency f of the robot arm is excited. On the other hand, when the torque peak time difference T is T2 (= λ),
Since the vibration of the robot arm generated during acceleration is canceled and suppressed during deceleration, the residual vibration is suppressed and reduced. In other words, this acceleration / deceleration has a frequency f /
This is the same as the excitation (vibration suppression) of No. 2 and the natural frequency f of the robot arm is suppressed.

Therefore, the acceleration / deceleration timing is controlled so that the torque peak time difference T acts in the direction of suppressing the vibration in accordance with the natural frequency f of the robot arm, that is, the natural vibration period λ, in all the moving distances of the robot arm. By doing so, it is possible to suppress residual vibration when the robot arm stops. That is, the speed is set such that the time difference between the peaks of the motor current (torque) of the acceleration / deceleration of the robot arm, that is, the torque peak time difference T3 is T3 = nλ (4) where n = 1, 2, 3,. By correcting the pattern, residual vibration when the robot arm is stopped can be suppressed.

Here, the vibration suppression control method of this embodiment is shown in FIG.
3 and the block diagram of FIG. FIG. 2 shows the speed, the motor current, and the acceleration of the tip of the robot arm when the moving distance is 1; FIGS. 4A, 4B, and 4C show FIGS.
(A), (b), and (c), wherein (d) is a speed waveform based on the corrected speed pattern, and (e) is a motor current corresponding to the corrected speed pattern. (F) shows the acceleration of the robot arm tip in the case of the corrected speed pattern. FIG. 3 is a block diagram showing a specific configuration for realizing the vibration suppression control method of this embodiment. As shown in FIG. The servo motor 3 is controlled by the controller 1 via the servo amplifier 2, and the controller 1 has a CPU 11 and an internal memory 12.

First, as shown in the flowchart of FIG. 1, in step S1, the natural frequency f of the robot arm 4 is determined.
Ask for. In this case, the servo motor 3 is actually driven from the controller 1 through the servo amplifier 2, and the state of free vibration of the robot arm 4 is actually measured by an acceleration sensor, an optical measuring device, or the like (not shown). This natural frequency f is stored in the internal memory 12 through the CPU 11 of the controller 1.

Next, in step S2, the CPU 11 calls the natural frequency f of the robot arm 4 from the internal memory 12, and calculates the natural frequency λ of the robot arm 4 based on the equation (1).
The natural oscillation period λ obtained in this manner is also stored in the internal memory 12.
To be stored.

The preparation operation is completed as described above. Next, when actually performing the vibration suppression control of the robot arm 4, C
The operation is performed by the operation after step S3 based on the program by the PU11.

First, the speed pattern is obtained according to the moving distance of the robot arm 4 given in step S3. This speed pattern is determined based on conditions that maximize both the speed and the acceleration. This is to maximize the operation speed of the robot arm 4.

Next, a current pattern of the servo motor 3 for realizing the speed pattern obtained in step S4 is calculated, and a torque peak time difference T is calculated.

Then, in step S5, a surplus operation is performed on the torque peak time difference T with the natural oscillation period λ. The arithmetic expression is T / λ = n remainder T ′ (5) Here, n is an integer.

After the above operation, the remainder T 'is evaluated in step S6. As a result of this evaluation, if the remainder T ′ is extremely close to “0 or λ”, it means that the torque peak time difference T is close to an integral multiple of the natural vibration period λ. In other words, it is possible to perform control in a speed pattern that suppresses the residual vibration of the robot arm 4.
Drive. As a result, residual vibration when the robot arm 4 stops is effectively suppressed, and the work of handling the workpiece by the finger portion 5 can be immediately started.

On the other hand, as a result of the evaluation in step S6, the remainder T
′ Is not very close to “0 or λ”, it means that the torque peak time difference T is not an integral multiple of the natural vibration period λ, and the residual vibration of the robot arm will increase in this state. I can only do it. Therefore, the process shifts to step S7 and the torque peak time difference T
Is corrected to be an integral multiple of the natural oscillation period λ, that is, T3 = (n + 1) λ (6). As a result, the velocity pattern of the robot arm 4 is corrected so as to suppress the residual vibration. Therefore, the processing moves to step S8 and the servo motor 3 is driven to suppress the residual vibration when the robot arm 4 stops. become.

For example, assuming that a velocity pattern as shown in FIG. 2A is obtained in step S3 to move the robot arm 4 by the movement distance l, the torque peak time difference T of the motor current obtained in step S4 is obtained. Fig. 2 (b)
As shown in FIG. As a result, when the robot arm 4 is driven as it is, the acceleration at the tip of the robot arm has a large residual vibration as shown in FIG. 2C. However, in the calculations in steps S5 and S6 and the evaluation of the calculation results, The remainder T ′ becomes λ / 2, and does not become “0 or λ”. Therefore, in step S7, a speed pattern is determined such that the torque peak time difference T becomes an integral multiple of the natural oscillation period λ, that is, the condition of the expression (6) is satisfied. That is, the speed pattern is corrected to a pattern as shown in FIG. As a result, the torque peak time difference T becomes T3 (= λ) as shown in FIG. 2E, and the residual vibration at the time of stopping the tip of the robot arm 4 is suppressed as shown in FIG. 2F.

As a result of the above control, when the tip of the robot arm 4 is moved by a short distance, the residual vibration when the robot arm 4 stops can be effectively suppressed, so that the positioning can be performed quickly and the robot can be positioned quickly. The control can be speeded up.

Also, since it is not feedback control,
There is no need to incorporate expensive sensors or speed up the control system, and the system can be configured at low cost.

Further, unlike the configuration in which vibration is generated and then detected and suppressed, the control is performed so that vibration is not generated from the beginning, so that the vibration suppression effect is high and efficient operation can be performed from the viewpoint of energy saving. it can.

In the above-described embodiment, a configuration in which the speed pattern is corrected in real time is exemplified. However, a configuration in which the speed patterns in all the operation patterns are obtained in advance and the speed patterns are determined by referring to the table may be employed. Good.

In the above embodiment, the drive of the robot arm has been described as an example. However, the present invention is similarly applicable to other control objects, and it is needless to say that the same effects can be obtained. . Also, it goes without saying that the present invention can be applied to a case where there are a plurality of control targets, even if the control targets are not only rotational movements but also linear movements.

[0038]

As described above, according to the vibration damping control method of the present invention, the natural frequency of a driven object such as a robot arm is measured in advance, and the robot arm is controlled with a speed pattern that suppresses the natural vibration. Is moved, it is possible to increase the speed of driving the robot arm and to suppress the residual vibration when the robot arm is stopped, and there is an effect that the system can be configured simply and inexpensively.

[Brief description of the drawings]

FIG. 1 is a flowchart of a vibration suppression control method according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a state of residual vibration according to a speed pattern.

FIG. 3 is a block diagram of a configuration for implementing the present invention.

FIG. 4 is an explanatory diagram of a difference in residual vibration.

FIG. 5 is an explanatory diagram of a control system of a general robot.

FIG. 6 is an explanatory diagram of the movement of the robot arm.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Controller 2 Servo amplifier 3 Servo motor 4 Robot arm 5 Finger part 11 CPU 12 Internal memory

──────────────────────────────────────────────────続 き Continuation of front page (56) References JP-A-1-289644 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) B25J 3/00-3/10 B25J 9 / 10-9/22 B25J 13/00-13/08 B25J 19/02-19/06 G05B 19/18-19/46 B23Q 15/12 G05D 19/00-19/02

Claims (3)

(57) [Claims] 1. A method of calculating a speed pattern capable of suppressing generation of vibration of a moving part of an industrial robot, comprising: a step of obtaining a speed pattern of the moving part; and a torque generated during an acceleration period of the speed pattern. Calculating a time difference between a peak and a torque peak generated during a deceleration period; and correcting the speed pattern such that the time difference between the torque peaks is substantially an integral multiple of the natural oscillation period of the moving part. Speed pattern calculation method. 2. A method for controlling a moving part of an industrial robot, wherein the moving part is drive-controlled according to a predetermined speed pattern, wherein the speed pattern includes a torque peak generated when the moving part is accelerated. time difference between the torque peaks occurring when the deceleration is approximately an integral multiple of the natural vibration period of the moving part, the control method. 3. An apparatus for calculating a speed pattern for suppressing vibration of a moving part of an industrial robot, comprising: means for obtaining a speed pattern of the moving part; and a torque peak and a deceleration generated during an acceleration period of the speed pattern. means for calculating a time difference between the torque peaks occurring in the period, the speed pattern calculation device and means for the time difference to correct the speed pattern so as to be approximately an integral multiple of the natural vibration period of the moving part.