JPH0424808A - Servo controller - Google Patents

Abstract

PURPOSE:To suppress the microvibrations of a control subject by producing a manipulated variable by a feedback compensator based on a reference position produced from a function generator and applying the servo control to the control subject. CONSTITUTION:The resolution of a function generator 1 is set higher than that of a position sensor. Then it is supposed that the present position (y) of a control subject 3 is equal to the position that is inputted precedently by a sampling action and the position of the subject 3 follows a reference position (r) produced by the generator 1. Then a feedback compensator 2 produces a manipulated variable (u) based on the position (r) and applies the servo control to the subject 3. Thus it is possible to improve the resolution of the generator 1 without changing the resolution of the position sensor and to improve the apparent resolution by correcting the position (y) given from the position sensor or obtaining an estimated position yp. As a result, the noise level of the variable (u) can be reduced and the microvibrations of the subject 3 are suppressed.

Description

[Detailed Description of the Invention] [Summary] Regarding a servo control device that servo-controls a controlled object, the present invention improves the resolution of a function generator without changing the resolution of a position sensor, and corrects or estimates the current position My of a position sensor. The purpose is to increase the apparent resolution and reduce the noise of the manipulated variable U to suppress minute vibrations of the controlled object. A function generator that generates a high reference position r and a reference speed V, and a manipulated variable such that the position deviation e, which is the difference between the high-resolution reference position r generated by this function generator and the position of the controlled object, becomes zero. U
and a feedback compensator that generates a position of The feedback compensator generates a manipulated variable U based on the reference position Wr, assuming that the position follows the reference position r generated by the function generator, and servo-controls the controlled object. . In addition, the position resolution of the function generator is made higher than the resolution of the position sensor, and when the current position y of the controlled object is the same as the position input one sampling ago, the controlled object is referenced by the reference generated by the function generator. The feedback compensator generates the manipulated variable U based on the position obtained by adding the reference speed V, assuming that the speed V is being followed, and servo-controls the controlled object. In addition, the position resolution of the function generator is set higher than the resolution of the position sensor, and the average speed is calculated from the time when this position sensor detects movement by the minimum resolution, and the movement is performed by the next minimum resolution. Based on the estimated position yP that is assumed to continue until then, the feedback compensator changes the manipulated variable U.
is configured to generate and servo control the controlled object.

[Industrial application field]

The present invention relates to a servo control device that servo-controls a controlled object.

[Conventional technology]

Conventionally, a servo control device, as shown in FIG. 8 (a),
A function generator 21 that generates a reference position r and a reference speed V, which are trajectories that change over time so that the target position rr and U target speed V commanded from a higher level, and the current position y of the controlled object 23 are referenced. Position deviation e=y−r to follow position r
It is composed of a feedback compensator 22 that generates a manipulated variable U to make the value zero. When performing precise control, the current position y is detected by an encoder or the like, and digital servo control is performed.

FIG. 8(b) shows a block section of the digital servo control device. FIG. 8(a) The function generator 21 and the feedback compensator 22 are realized by software within the MPU 24. The manipulated variable U is output to the D/A converter 25 and converted into an analog variable, which rotates the DC motor 27 via the power amplifier 26 and drives the mechanical section 29 . D.C.
The rotation of the motor 27 is detected by a rotary encoder 28, the pulses from the rotary encoder 28 are counted by a counter 30, and the current position y is taken into the MPU 24.

FIG. 8(C) shows a flowchart of the conventional servo control device having the configurations of FIGS. 8(A) and 8(B).

In FIG. 8(C), the function generator 21 generates a reference position lr and a reference velocity V.

■ is the current position y of the controlled object 23 (current position 1jy counted by the low-return encoder 28 and counter 30)
is input to the MPU 24.

(2) calculates the positional deviation e.

(2) calculates the difference (differential) of the positional deviation.

(2) calculates the integral of the positional deviation.

In (2), the manipulated variable U is determined by an illustrated formula. Here, g,
is the proportional gain, gl is the integral gain, and g.

represents the differential gain.

(2) outputs the manipulated variable U to the controlled object 23;

(2) is +ll for k-, repeat the steps after (2) for the next time.

[Problem to be solved by the invention]

Since the rotary encoder 28 or the like is used as a position sensor, the position of the controlled object 23 becomes discrete in units of resolution of the rotary encoder 28. When the speed is very slow, the position My becomes a step with a height of resolution S, as shown in FIG. In order to maintain accuracy up to the resolution, the resolution of the generated reference position r is made equal to the resolution of the current position y, so the reference position r also has a step shape with a height S like the current position y. Therefore, the position deviation e occurs in a pulse form only when the reference position r and the current position y deviate, so the manipulated variable U also
Generates pulse-like noise as shown in . This noise causes minute vibrations in the controlled object 23, posing a problem in that precise speed control cannot be performed. In order to reduce this noise, it is possible to increase the resolution of the position sensor and make the step S smaller. However, increasing the resolution of the position sensor increases the cost, and it is physically difficult to further increase the resolution. The problem is that it is not possible to hire people when times are difficult.

The present invention increases the resolution of the function generator without changing the resolution of the position sensor, and also increases the current position y of the position sensor.
The purpose is to correct or obtain the estimated position y2 to increase the apparent resolution, reduce noise in the manipulated variable U, and suppress minute vibrations of the controlled object.

[Means to solve problems]

Means for solving the problem will be explained with reference to FIG.

In FIG. 1, a function generator 1 generates a reference position r and a reference speed V with high resolution based on a commanded target position and target speed.

The feedback compensator 2 generates the manipulated variable U so that the positional deviation e, which is the difference between the high-resolution reference position r generated by the function generator l and the position of the controlled object 3, becomes zero. .

[Effect]

As shown in FIG. 1, the present invention makes the position resolution of the function generator 1 higher than the resolution of the position sensor, and
When the current position y of is the same as the position input one sampling ago, it is assumed that the position of the controlled object 3 is following the reference position Wr generated by the function generator 1, and the feed is performed based on the reference position r. A bank compensator 2 generates a manipulated variable U to servo control a controlled object 3. Further, the position resolution of the function generator 1 is made higher than the resolution of the position sensor, and when the current position y of the controlled object 3 is the same as the position input one sampling ago, the controlled object 3 is
The feedback compensator 2 generates the manipulated variable U based on the position obtained by adding the reference speed ■, assuming that it is following the reference speed V generated by , and servo-controls the controlled object 3. ing. In addition, the position resolution of function generator 1 is made higher than the resolution of the position sensor, and the average speed is calculated from the time when this position sensor detects movement by the minimum resolution, and the movement is performed by the next minimum resolution. Until then, the feedback compensator 2 generates the manipulated variable U based on the estimated value WyP obtained by assuming that it will continue.
The controlled object 3 is servo controlled.

Therefore, without changing the resolution of the position sensor, by increasing the resolution of the function generator 1 and correcting the current position y from the position sensor or obtaining the estimated value Wyp, the apparent resolution can be increased. This makes it possible to reduce noise and suppress minute vibrations of the controlled object.

〔Example〕

First, the configuration and operation of one embodiment of the present invention will be sequentially explained in detail using FIGS. 1 to 3.

In FIG. 1, a function generator 1 generates a reference position r and a reference velocity V with high resolution that change over time based on a target position rf and a target velocity vf commanded from a host device.

Is the corrector 1-1 based on the current position of the controlled object 3? A corrected position y' is obtained by correcting &y, and an estimated value 1yp is obtained from the time when the position sensor moves by the minimum resolution.

The feedback compensator 2 uses the high-resolution reference position Wr generated by the function generator 1 and the corrector 1) to correct the current position y or the estimated value 1y.
The operation amount U is generated so that the positional deviation e, which is the difference from the positional deviation P, becomes zero.

The controlled object 3 is an object to be feedback-controlled by outputting the operation i1u.

Next, the operation of the configuration shown in FIG. 1 will be explained in detail in accordance with the order shown in the flowchart of FIG.

In FIG. 2, ■ is the reference speed v(k), the reference position r
(k) is generated. The function generator 1 generates a reference speed V(k) and a reference position r(k) that change over time based on the commanded target position Wrf and target speed vf using the following equations.

v(K)=v(k-1) 10a(k-1)...
...(1) r(K)I=r(k-1)+ν(k-1)
(2) Here, k represents the current sampling, k-1 represents one sub-ring before, and a represents the acceleration.

(2) Inputs y (k). This inputs the current position y (k) of the controlled object 3.

(2) determines whether y(k)·y(k-1). This determines whether the current position 1y(k) input in step (2) is equal to the position IFy(k-1) one sampling ago. Y
In the case of ES, let y' (k) = r(k) in ■.
Substituting the reference position r (k) generated in (k) into the correction value fy'(k)), and performing (2). On the other hand, in the case of No, the current position y (k) has changed from the position y (k-1) one sampling ago, so the current position y (k) after this change in ■
) to the corrected position V'(k) and perform step (2).

(2) determines whether y'(k) > y(k)+s (s represents the resolution of the position sensor in steps). This determines whether y''(k) (=reference position r (k) ) exceeds the value of current position y(k) + s. YES
In the case (when y'(k) - y(k) exceeds S), use ■ to calculate the corrected position y'(k)/current position y(k) +
s, make sure that y'(k) - y(k) does not exceed the resolution step S of the position sensor, and perform ■. On the other hand, No.
In the case of , the correction value fy "(k)
- Set the reference position r (k) and perform ■.

By the above processing, the current position 1) Fy(k) is converted to the following fi
The correction value 1ily'(
k).

fil ■When NO, set the correction value 1fy'(k)・Current position may(k), (2)■YES G■When YES, set the correction position y'(k)
)=current position y(k)+s, (3)■YES■
When No, corrected value il! y'(k)・Reference position r (
k).

Then, by ■ or 0, these (1), (2), (
The feedback compensator 2 generates the manipulated variable U based on the correction value ty''(k) corrected by either of 3),
It is output to the controlled object 3 to perform feedback control.

(2) calculates the positional deviation e (k) using the following equation (3).

e(k)su”(k)-r(k)・・・・・・・・・・・・
・(3) ■ is the difference in positional deviation (
Find the differential).

e4(k)8e(k)-e(k-1)・・・・・・・・・
...(4) [Phase] calculates the integral of the positional deviation using the following equation (5).

e, (k)2e1(k-1)+e(k)...
・(5) ■ is the manipulated variable u (k) according to the following formula (6)
seek.

u(k)=-g p e(k)-g; e r (kL
g a e a (k) · (61 Here, g represents a proportional gain, g represents an integral gain, and g4 represents a differential gain.

(2) outputs u (k). This is because the feedback compensator 2 in FIG. 1 outputs the manipulated variable u(k) to the controlled object 3 to perform feedback control.

0 is +L L for k −, for the next sampling,
■Do Ijin repeatedly.

The operation in the flowchart of FIG. 2 will be specifically explained using FIG. The solid line represents the change of the present invention, and the dotted line represents the change of the conventional method.Here, the step of the present invention has a resolution that is 1/8 of that of the conventional method.

FIG. 3(a) shows a high-resolution reference position r generated by the function generator 1. The vertical axis represents the reference position r, and the horizontal axis represents time (sampling time).

FIG. 3 (b) shows the corrected position y''. Here, the ONo shown in the figure is ``YES'' and ``N'' in FIG. 2.

, ONo represents the correction value 1y' in the case of ■NO in Figure 2, and ■YES represents the correction position y° in the case of Figure 2■
S and ■Represents the correction position y when YES.

FIG. 3(c) shows the positional deviation e. This positional deviation e is calculated by increasing the resolution of the reference position r in Fig. 3 (a) and the corrected position y' in Fig. 3 (b), and changing the step to 1/8 of the conventional one.
Accordingly, the size is reduced to 1/8 to 1/4 of the conventional size.

FIG. 3(d) shows the manipulated variable U. This operation amount U is such that the magnitude of the positional deviation e is 1/8 to 1 of the conventional value as shown in FIG.
With the reduction to /4, pulse-like noise is reduced. Therefore, by outputting the manipulated variable U with reduced pulse-like noise to the controlled object 3 and performing feedback control, it is possible to suppress minute vibrations of the controlled object 3.

Next, the configuration and operation of another embodiment of the present invention will be explained in detail using FIGS. 1, 4, and 5. this is,
The resolution of the function generator 1 is increased without changing the resolution of the position sensor of the controlled object 3, and when the current position y of the controlled object 3 detected by the position sensor is the same as the position input one sampling ago, the controlled object 3 assumes that the controlled object 3 is following the reference speed V generated by the function generator 1, and the feedback compensator 2 generates the manipulated variable U to servo control the controlled object 3. Here, by replacing ■ in Fig. 2 with the phase shown in Fig. 4, and assuming that the controlled object 3 is following the reference speed V generated by the function generator 1, Previous correction position V' (
k) to obtain a new corrected position y' (k), which will be explained below.

In FIG. 4, for [phase], the function generator 1 generates a reference velocity v (k) and a reference position r (k) according to the illustrated formula.

Is 0 the current position of controlled object 3? &y(k).

[Phase] is the current position y (k) input at 0 and the previous position? &y(k-1) are equal. YE
In the case of S, the corrected position y''(k) is increased by the reference speed v(k) in [phase].

[Phase] is the correction position y"(k) in [Phase] with reference speed ν(
It is determined whether the increase by k> exceeds the resolution step S of the position sensor. If YES (exceeds), correct position V' (k)・y(k) in [phase]
+3 and do not exceed step S [phase]
I do.

On the other hand, in the case of No (if not exceeded), the correction position y''(k)·y' (k) +v(k) substituted in [phase] is set, and [phase] is performed.

■ is [phase] and the current position y (k) is the previous position y (k
-1), so the corrected position y'(k)=
Set the current position to y (k) and perform [phase].

By the above processing, the current position y (k) can be converted to the following (1)
The correction position y'(k) is corrected to one of the correction positions y'(k) and 3).

(1) When ONo, set the corrected position y' (k) and current position y(k), and when (210Y E S and oYES, set the corrected position y'
'(k)・Current position 1fy(k)+s, (3n 0
When YES and [phase] No, correction value My'(k)=corrected position y'(k) + reference speed v(k).

And these (1), (2) by [phase] or @
, (3), the corrected position V' (
k), the feedback compensator 2 generates a manipulated variable U and outputs it to the controlled object 3 to perform feedback control. Note that 0 to @ are the same as ① to 0 in FIG. 2, so their explanation will be omitted.

The operation in the flowchart of FIG. 4 will be specifically explained using FIG. The solid line represents the variation of the present invention, and the dotted line represents the conventional variation. Here, the steps of the present invention have a high resolution of 1/8 steps compared to the conventional method.

FIG. 5(A) shows a high-resolution reference position r generated by the function generator I. The M axis represents the reference position r, and the horizontal axis represents time (subumbling time).

FIG. 5(B) shows the corrected position y''. Here, ONO shown in the figure is 0YES and [phase] N in FIG.

Corrected position y' (k) = y' (k) +v(
k), ONO is the corrected position Y when ONO is shown in Figure 4.
'(k)・y(k), oYES is 0YE in Figure 4
Correction position y”(k)・y(k) when S and oYES
Represents +s.

FIG. 5(c) shows the positional deviation e. This positional deviation e is determined by increasing the resolution of the position 1)r (see FIG. 5(a)) and the correction value Wy' (see FIG.
/8, so the size is also 1/8 of the conventional size.
It becomes smaller by 8 to 1/4.

FIG. 5(d) shows the manipulated variable U. This operation amount U is such that the magnitude of the positional deviation e is 1/8 to 1 of the conventional value as shown in Fig. 5 (c).
With the reduction to /4, pulse-like noise is reduced. Therefore, by outputting the manipulated variable U with reduced pulse-like noise to the controlled object 3 and performing feedback control, it is possible to suppress minute vibrations of the controlled object 3.

Next, the configuration and operation of other embodiments of the present invention will be explained in detail using FIGS. 1, 6, and 7. This is done by making the position resolution of the function generator 1 higher than the resolution of the position sensor, and when the position sensor moves one step, the average speed during that time is calculated from the time required to move one step, and the next The estimated position y is calculated by assuming that the object continues to move at this average speed until it moves one step.Based on this estimated position y, the feedback compensator 2 generates the manipulated variable U, and controls the The object 3 is servo controlled. This will be explained below.

FIG. 6(A) shows an explanatory diagram of the relationship between the actual position and the estimated position. Here, the broken line represents the actual position of the controlled object 3, the solid line represents the estimated position yP estimated by the present invention, and the dotted line represents the position detected by the position sensor.

FIG. 6 (b) shows an explanatory diagram when estimating the position.
This is an enlarged view of a part of FIG. 6(a). This will be explained below.

At time tJ-I, the position detected by the position sensor is (yr s) (point A), and at time 1j, the position is moved by 1 step S and the position becomes yr (point B), and the estimated position is y (point B'). At time t, from then on,
Assuming that the speed from time tj-1 to tJ continues as it is, the time at which the next step S moves is tj'
It is time 21, -1J-, after j-1 hours, and the straight line AB is extended to reach a point C' where it intersects with y=yr+S.

The estimated position at time tJ is yI, which deviates from the actual position yr, so in order to correct the estimation, time 2
So that the position when tJ-tJ- is Yr''S,
Straight line B'C' becomes the locus of the estimated position.

In fact, since the speed changes after t, the time 2t
, -tj-1, but moves by one step S at time t1°1 (point C). The estimated position at this point is yp'
” (point C°゛). Similarly, the estimated position is determined while correcting the estimation.

The servo control operation when the estimated position shown in FIG. 6 is used will be explained in detail using FIG. 7.

In FIG. 7, O determines the current reference speed ν(k) and reference position r(k).

@ is the current estimated speed ν at the estimated position y p (k-1)
2 is added to obtain the estimated position V P (k).

@ to @ calculate the manipulated variable u(k) using the estimated position y p (k).

@ inputs the current position y (k).

@ means that the current position y(k) is the previous step y(k-1)
Determine whether it is equal to or not. If YES (if they are equal), k is incremented at ■ without modifying the estimated speed ν2, and steps after ■ are repeated. On the other hand, if No (if they are not equal), increment the time and register that time t = k in [Phase], and change the current position from the time t when the position changed last time in [Phase]. Time t
Estimated speed V that moves by the sum of movement amount S and current estimation error yy during time t, -1, - until J.
, is determined by a diagrammatic formula. This estimated speed V.

From next time onwards, continue to use until the position changes (0
).

〔Effect of the invention〕

As explained above, according to the present invention, the resolution of the interval generator 1 can be increased without changing the resolution of the position sensor, the current position y from the position sensor can be corrected, and the current position y can be moved by the minimum resolution of the position sensor. Find the average speed from the time required for the estimated position y.

Since the configuration is adopted to increase the apparent resolution of the position sensor, it is possible to reduce the noise of the operation 1u and suppress minute vibrations of the controlled object 3.

[Brief explanation of drawings]

FIG. 1 is a principle block diagram of the present invention, FIG. 2 is an operation flowchart of the present invention, and FIG. 3 is a time chart of the present invention.
FIG. 4 is an operation flowchart of another embodiment of the present invention, FIG. 5 is a time chart of another embodiment of the present invention, FIG. 6 is an explanatory diagram of another embodiment of the present invention, and FIG. 7 is a diagram of another embodiment of the present invention. FIG. 8 is an operational flowchart of another embodiment configuration and is an explanatory diagram of the prior art. In the figure, 1 represents a function generator, 1-1 a corrector, 2 a feedback compensator, and 3 a controlled object. Section J: Movement to the Ise's Tufts of Ise of Honmenmoe 4-7 U-Naya 1 No. 4 Fig. 1) Ise's Time Nayat Fig. 5 Water Rate August's Great Example Ikej Structure 6th figure of the same month, wooden frame of August, Oba daj structure movement 4-7B-su↑-to-light map (d), edited by J. U.S., Kanro map No. B

Claims (3)

[Claims] (1) In a servo control device that servo-controls a controlled object, a function generator (
1), and feedback compensation that generates the manipulated variable u so that the position deviation e, which is the difference between the high-resolution reference position r generated by the function generator (1) and the position of the controlled object (3), becomes zero. The function generator (1) has a position resolution higher than that of the position sensor, and when the current position y of the controlled object (3) is the same as the position input one sampling ago, , the controlled object (
It is assumed that the position of 3) follows the reference position r generated by the function generator (1), and the feedback compensator (2) generates the manipulated variable u based on the reference position r, and performs control. A servo control device configured to servo control an object (3). (2) In a servo control device that servo-controls a controlled object, a function generator (1) that generates a reference position r and a reference speed v with higher resolution than a position sensor based on a commanded target position and target speed; , a feedback compensator (2) that generates a manipulated variable u so that the position deviation e, which is the difference between the high-resolution reference position generated by the function generator (1) and the position of the controlled object (3), becomes zero. ), the position resolution of the function generator (1) is higher than the resolution of the position sensor, and when the current position y of the controlled object (3) is the same as the position input one sampling ago, the controlled object (
3) is assumed to follow the reference speed v generated by the function generator (1), and the feedback compensator (2) calculates the manipulated variable u based on the position obtained by adding the reference speed v. What is claimed is: 1. A servo control device, characterized in that it is configured to generate and servo control a controlled object (3). (3) In a servo control device that servo-controls a controlled object, a function generator (
1), and feedback compensation that generates the manipulated variable u so that the position deviation e, which is the difference between the high-resolution reference position r generated by the function generator (1) and the position of the controlled object (3), becomes zero. (2), the position resolution of the function generator (1) is higher than the resolution of the position sensor, and the average speed is calculated from the time when the position sensor detects movement by the minimum resolution. Based on the estimated position y_P, which is assumed to continue until the next minimum resolution movement, the feedback compensator (2) generates the manipulated variable u, and the controlled object (
3) A servo control device configured to perform servo control.