JPH0737258A - Tracking control device - Google Patents

Abstract

PURPOSE:To absorb variations of a gain and to maintain stability and then to enhance servo rigidity by obtaining driving means for a tracking actuator and a linear motor, etc., with diminished characteristic variations in resonance frequency and damping number, etc., and hardly susceptible to disturbance and constituting a parallel servo system in using the tracking actuator and the linear motor, etc. CONSTITUTION:Outputs of the tracking actuator 1, the linear motor 2 and models 51 and 61 made to mimic characteristcs of the driving means for the above parts are compared by comparing means 52 and 62. A tracking actuator control part 50 and a linear motor control part 60, having constitution of feeding back the compared outputs through feedback elements 53 and 63 respectively are applied to the parallel servo system.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tracking control device such as a tracking actuator and a linear motor in a servo system of an optical disk device.

[0002]

2. Description of the Related Art Generally, an optical disk device has a tracking servo system for driving a lens in the radial direction of the disk in order to match a light beam spot with a target track, and in particular, an optical disk used for data recording of a computer or the like. The tracking servo system of the device is shown in FIG.
As shown in FIG. 7, a parallel servo system for positioning the light beam spot by using the tracking actuator 1 which is a fine actuator and the linear motor 2 which is a coarse actuator may be used.

In the parallel servo system of the optical disk device of FIG. 17, an error between the target displacement of the track 4 of the disk 3 and the displacement of the light beam spot 5 is tracked by an optical system and an electric circuit (not shown) in the optical pickup 6. The error signal TE is detected. This tracking error signal TE
In order to stabilize the system, the phase compensation circuits 7 and 8 perform phase compensation and then are added to the drive circuits 9 and 10, respectively.
The tracking actuator 1 and the linear motor 2 are driven. The tracking actuator 1 is a lens 11
Drive the optical beam spot 5 to follow the high frequency / small amplitude displacement of the track 4, while the linear motor 2 drives the tracking actuator 1 together with the optical pickup 6 etc. to move the low frequency / large amplitude displacement of the track 4. To follow.

FIG. 17 is a block diagram of FIG. 17, and it can be seen that the tracking actuator 1 and the linear motor 2 operate in parallel based on the tracking error signal TE.

The tracking actuator 1 is a lens 1
In many cases, a movable part including 1 is supported by a spring and is driven by an electromagnetic force, which is represented by a spring-mass physical model as shown in FIG. In this physical model, the transfer characteristic from the driving force F _A to the displacement X _A of the movable part (lens) is of the second-order low-pass filter (LPF) type as is well known, and its gain curve is shown in FIG. Be done.

However, f _A is the resonance frequency of the tracking actuator, ζ _A is the damping number that determines the swell (resonance peak) of the gain curve near f _A , and the mass of the movable part (lens or the like) of the tracking actuator is m. , F _A = (1 / 2π) ・ (k /, where k is the spring constant of the spring, and d is the viscosity of the rubber intentionally attached to the spring member or the viscosity of the spring member itself or the air resistance. m) ^1/2 ... (1) ζ _A = d / {2 · (m · k) ^1/2 } ... (2)

[0007] Next, a physical model of the linear motor is an inertial mass system in FIG. 21, the transfer characteristic from driving force F _L to the displacement X _L by the way parallel servo system as described above is shown in FIG. 22, the low A linear motor follows the frequency band and a tracking actuator follows the high frequency band. Therefore, in order to secure stability near the cutoff frequency of the entire servo system, the tracking actuator system is subjected to phase lead / lag compensation having differential characteristics at frequencies from f _P to f _{Q as} shown in FIG. On the other hand, in order to increase the gain at low frequencies where the displacement of the disk due to eccentricity is large and to improve the servo tracking performance, the linear motor system has a break frequency at f _LPF as shown in FIG.
The following low-pass filter is applied to increase the servo gain at low frequencies. FIG. 18 is a block diagram showing a combination of these phase compensations added to the spring-mass tracking actuator having the transfer characteristics shown in FIG. 20 and the inertial mass linear motor having the transfer characteristics shown in FIG. The servo gain curve of the shown parallel servo system is shown in FIG.

In FIG. 25, f _X represents the frequency at which the gains of the linear motor system and the tracking actuator system become equal. With this as a boundary, the gains of the linear motor system and the tracking actuator system, in other words, the magnitude relationship between the displacement amounts of the linear motor and the tracking actuator at the time of servo tracking are exchanged, and the movement of the linear motor exceeds the tracking actuator at a frequency lower than f _X.
On the contrary, at a frequency higher than f _X , the movement of the tracking actuator becomes dominant and the linear motor hardly moves. In FIG. 25, f _A is the resonance frequency of the tracking actuator, f _P is the start frequency of phase lead compensation, and f _C
Indicates the cutoff frequency.

In the parallel servo system as described above, in order to stabilize the system, it is necessary to keep the damping number of the tracking actuator large and reduce the gain peak (resonance peak) in the vicinity of the resonance frequency. This will be explained subsequently.

FIG. 26 shows curves of gain and phase of the servo system when the damping number ζ _A of the tracking actuator is small and a large resonance peak occurs near the resonance frequency f _A in the parallel servo system. As shown in this figure, when the gain curve of the linear motor system and the tracking curve of the tracking actuator system intersect or approach due to the resonance peak near the resonance frequency f _{A of the} tracking actuator, the frequency near f _A indicated by the diagonal lines in FIG. Then, the magnitudes of the gains of the linear motor system and the tracking actuator system are substantially equal, and the phase difference is approximately 180.
It is easy to meet the condition called °. At such frequencies, the linear motor and the tracking actuator have almost the same amplitude, and inversion movement occurs in which they are displaced in opposite directions, and as a result, the parallel servo system becomes unstable.

In order to suppress the resonance peak of the tracking actuator to a small value and avoid this unstable phenomenon, rubber with high viscosity may be attached to the spring of the tracking actuator. However, the damping number due to temperature change or secular change of rubber may occur. In order to avoid the change of
As shown in FIG. 7, the speed of the movable portion (lens) of the tracking actuator is detected by the speed sensor 12, and the amplifier 1
A method of suppressing the resonance peak by amplifying in 3 and then performing negative feedback to improve the damping number is known.

According to this method, when the thrust constant (sensitivity) of the tracking actuator is K _FA , the sensitivity of the speed sensor 12 is S _V , and the gain of the amplifier 13 is A, the damping number ζ _A 'is given by equation (2). Ζ _A is different from ζ _A , ζ _A '= (d + K _FA · S _V · A) / {2 (mk) ^1/2 } (3), and the number of damping can be adjusted by adjusting the gain of the amplifier 13. It is possible to suppress the resonance peak to a required value by changing

[0013]

However, these conventional techniques have the following problems.

In order to stabilize the parallel servo system, it is necessary to increase the damping number of the tracking actuator and suppress the resonance peak. For this reason, for example, the method of feeding back the speed of the tracking actuator as described above is used. Even if it is adopted, there is a problem that the procedure for adjusting the gain of the amplifier for feeding back the speed is increased.

This adjustment is performed, for example, by using a frequency characteristic measuring device such as an FFT analyzer and the like, until the resonance peak characteristic of the tracking actuator has a desired shape while the operator looks at the display tube surface of the measuring device and increases or decreases the variable resistor. It takes a lot of time and labor to do so, which leads to an increase in the production cost of the device.

Further, the low range servo gain G _A of the tracking actuator system which is a spring-mass system actuator
Is the cutoff frequency f _C , the phase lead compensation start frequency f _P , and the tracking actuator resonance frequency f _A
Then, G _A = (f _P · f _C ) / (f _A ) ² (4)

The values of these f _P , f _C and f _A are determined at the design stage of the servo system. Here, since f _P is determined by the electric circuit constant, it may be considered that it fluctuates only as much as a component error and practically does not change. For f _C, the gain of the entire servo system can be adjusted to a predetermined value by, for example, increasing or decreasing the output of the phase compensation circuit with a variable resistor. On the other hand, f _A is determined by the mass of the movable part of the tracking actuator and the spring constant of the spring used, as already shown in the equation (1), and f _A fluctuates due to component errors and assembly errors at the production stage. However, the adjustment cannot be performed without mechanical adjustment of the mass of the moving part, the spring constant, etc., that is, by adding a little weight to the moving part of the tracking actuator or scraping the spring. Virtually impossible.

The fluctuation of the resonance frequency f _A of the spring-mass tracking actuator changes the low-pass servo gain G _A according to the square of the relational expression. Due to the variation of the low-range servo gain G _A , the gain sharing of the linear motor system and the tracking actuator system varies in the low range. In particular, as shown in FIG. 28, when the actual resonance frequency f _A ″ of the tracking actuator fluctuates to a value lower than the design value f _A , the low range servo gain G _A rises to the low range servo gain of the linear motor system. approach. At this time, the displacement of the tracking actuator during servo tracking increases as it approaches the displacement of the linear motor, so the displacement of the light beam spot due to lens displacement also increases, and the tracking generated in the optical system inside the optical pickup accordingly. The optical offset of the error signal TE also increases, and the servo tracking performance deteriorates.

Even if the damping number of the tracking actuator can be adjusted by the method of feeding back the speed of the tracking actuator, the resonance frequency cannot be adjusted, so that this problem in the parallel servo system cannot be solved.

Further, the tracking actuator and the linear motor are usually driven by electromagnetic force, and when the magnetic flux density of the permanent magnets used in the tracking actuator system and the linear motor system fluctuates at that time, the sensitivity and the servo gain of each fluctuate. Occurs, and the same problem as in the case where the resonance frequency of the tracking actuator fluctuates occurs at this time. FIG. 29 shows the gain curve of the parallel servo system when the sensitivity of the linear motor is reduced.
For example, the frequency f _{X at} which the gains of the linear motor system and the tracking actuator system become equal to each other decreases to f _X '. This decrease in f _X leads to a change in the gain sharing of the linear motor system and the tracking actuator system, which leads to an increase in the displacement amount of the light beam spot due to the lens displacement, as in the case where the resonance frequency of the tracking actuator is reduced. As a result, the servo tracking performance deteriorates. On the contrary, when the servo gain increases due to the increase in the sensitivity of the linear motor, the above f _X increases and approaches the cutoff frequency f _C , which is due to the phase delay of the low pass filter inserted in the linear motor system, Or it means that higher-order resonances that are likely to exist in a linear motor system easily affect the vicinity of the cutoff frequency.
It is not preferable because of the stability of the servo system.

With respect to the vibration resistance of the servo system, when disturbance is applied to the servo system, the ability to hold the lens and the light beam spot at a desired position on the disk by the disturbance vibration, so to speak, the servo rigidity or vibration resistance. In the case of the conventional parallel servo system, it can be said that it is almost determined by the vibration resistance performance of the linear motor system. The higher the servo gain, the better the vibration resistance of the linear motor system. However, if the servo gain is simply increased in order to obtain this high vibration resistance, the frequency f _{X at} which the gains of the linear motor system and the tracking actuator system become equal as described above.
Becomes higher and the phase margin near the cutoff frequency f _C decreases due to the influence of the first-order low-pass filter inserted for increasing the gain of the linear motor system. Therefore, it is impossible to unnecessarily increase the gain of the linear motor system and the vibration resistance of the parallel servo system.

The present invention has been made in view of the above points. The purpose is to absorb fluctuations in the characteristics of driving means such as tracking actuators and linear motors, and to always have a constant share of gain for both linear motors and tracking actuators in a parallel servo system. It is to provide a control device of a drive means capable of maintaining the stability of the system without making complicated adjustments.

[0023]

In order to achieve the above object, in the drive means control device according to the present invention, the light generating means and the optical for guiding the light from the light generating means to the optical disk and condensing as a light beam spot. System and a tracking error signal detection system that detects a deviation amount of the light beam spot from a predetermined track of the optical disc and generates a tracking error signal, and a coarse actuator and a fine actuator based on the tracking error signal In the tracking control device for causing the light beam spot to follow the predetermined track by driving the first fine actuator drive signal based on the error signal to drive the fine actuator. And a second driving circuit based on the error signal A first model that electrically simulates a desired transfer characteristic of a dense actuator to which a dense actuator drive signal is input, and a first model that compares the output of a displacement sensor that detects the displacement of the dense actuator with the output of this first model. 1 and the output of the first comparing means is fed back to the input side of the first drive circuit by a first predetermined multiplication and fed back to the input side of the first model by a second predetermined multiplication. A fine actuator control unit having a first feedback circuit for
Second driving circuit for driving the coarse actuator by inputting the coarse actuator driving signal and the desired transfer characteristics of the coarse actuator receiving the second coarse actuator driving signal based on the error signal are electrically simulated. The second model, the second comparing means for comparing the output of the displacement sensor for detecting the displacement of the coarse actuator with the output of the second model, and the output of the second comparing means for the second driving. A coarse actuator control section having a second feedback circuit for feeding back to the input side of the circuit by a third predetermined multiplication and for feeding back to the input side of the second model for a fourth predetermined multiplication. .

At this time, the transfer characteristic of at least one of the first and second feedback circuits may be a first-order advance / delay characteristic, and the transfer characteristic of at least one of the first and second feedback circuits. May have a second-order advance / delay characteristic.

The first feedback circuit has a first
Is a predetermined multiple of (α ₁ + β ₁ ) times, and the second predetermined multiple is α ₁
In addition, the third predetermined multiple of the second feedback circuit is (α ₂ + β ₂ ) times, and the fourth predetermined multiple is α.
It may be _two-fold.

[0026]

According to the present invention, even if there are fluctuations in the resonance frequency, damping number, sensitivity, etc. in the individual fine actuators or tracking actuators and coarse actuators or linear motors, their outputs match the model. As a result of being controlled, these characteristics become equivalent to the characteristics of the respective models. Therefore, if the characteristics such as resonance frequency and damping number desired for these are set in the model in advance, it is possible to absorb the characteristic fluctuations of the actual tracking actuator and the linear motor, and to have a constant characteristic at all times. It is possible to stabilize the system without making complex adjustments and to keep the device characteristics constant.

Further, even when a disturbance such as vibration is applied to the tracking actuator or the linear motor and the output thereof is changed, the behavior different from the output of the model is suppressed as described above. It is possible to obtain a drive means control device that is strong against disturbances, in other words, excellent in servo rigidity or vibration resistance.

[0028]

【Example】

(Embodiment 1) FIG. 1 shows a drive means control device according to a first embodiment of the present invention, which is applied to a tracking servo system of an optical disk device including a tracking actuator 1 which is a fine actuator and a linear motor 2 which is a coarse actuator. 3 is a block diagram of the embodiment of FIG. In each of the following drawings, the same parts as those shown in the previous drawings are designated by the same reference numerals, and duplicate description will be omitted.

FIG. 1 shows the conventional example shown in FIG.
A tracking actuator controller 50 and a linear motor controller 60 are provided in place of the drive circuits 9 and 10, the tracking actuator 1 and the linear motor 2 in FIG.

2A shows the inside of the tracking actuator control unit 50 in FIG. 1, and FIG. 2B shows the inside of the linear motor control unit 60 in FIG.

2 (a) and 2 (b), the transfer characteristics of the tracking actuator 1 and the linear motor 2 are represented by P _{A (S)} and P _{L (S)} , respectively. The drive circuits 9 and 1
It is assumed that 0 is usually sufficiently satisfied, that is, it has no frequency characteristic and its gain is 1. Displacement sensor 5
Reference numerals 5 and 65 detect displacements of the tracking actuator 1 and the linear motor 2, and irradiate the movable parts of the actuator 1 and the linear motor 2 with, for example, light such as a light emitting diode or a part of light from a light source toward the disk. However, an optical displacement detection sensor or the like in which the reflected light or the light partially blocked by the movable portion is received by a photodetector such as a photodiode to detect the displacement can be considered.

The ideal models 51 and 61 simulate the characteristics of the tracking actuator and the linear motor, and their transfer characteristics are P _{mA (S)} and P _{mL (S), respectively.}
It is represented by. In order to compare the outputs detected by the displacement sensors 55 and 65 of the actuator 1 and the linear motor 2 with the outputs of the ideal models 51 and 61 later, P _{mA (s)} ,
P _{mL (s)} includes the displacement sensor gain. The state feedback elements 53 and 63 are elements that determine the error suppression characteristics and stability of the system, and their transfer characteristics are represented by H _{A (S)} and H _{L (S)} . Disturbances D ₁ and D ₂ represent disturbances applied to the tracking actuator and the linear motor.

The operation of the first embodiment will be summarized below with reference to FIGS. 1 and 2 (a).

The same input X is applied from the phase compensation circuit 7 of FIG. 1 to both the "actual actuator", that is, the tracking actuator 1 and the "ideal model of the actuator" 51. On the other hand, "actual actuator"
That is, the output of the tracking actuator 1 detected by the displacement sensor 55 and the output of the “ideal model” 51 are compared by the comparison means 52, and the output of the comparison means 52 is passed through the state feedback element 53 and the drive circuit 9. By adding it to the "actual actuator", that is, the tracking actuator 1, its output is the "ideal model".
Drive so as to match the output of 1.

Therefore, even if there is a characteristic variation such as a variation in the resonance frequency or the damping number of the actual driving means (tracking actuator) during production, this is an error between the "real" and the "ideal model". The characteristics of the actual tracking actuator are controlled so as to match the characteristics of the “ideal model”. Even if a disturbance D ₁ such as an external vibration is applied to the tracking actuator, the movement of the tracking actuator due to this disturbance is regarded as an error that does not exist in the “ideal model” and is suppressed. For this reason, it is possible to obtain a tracking actuator that has virtually the same characteristics as the ideal model with less influence of disturbance and fluctuation in characteristics. Similarly, FIG.
Based on the block diagram of (b), a linear motor having almost the same characteristics as the ideal model can be obtained.

The tracking controller shown in FIG. 1 has a parallel servo system using a tracking actuator and a linear motor having substantially the same characteristics as the ideal model.

Subsequently, the tracking actuator and the linear motor having the above characteristics are shown in FIGS. 2 (a) and 2 (b).
What is obtained in the tracking actuator control unit and the linear motor control unit of 1 will be further described based on mathematical expressions and the like. Since the gains of the displacement sensors 55 and 65 are considered to be included in P _{mA (s)} and P _{mL (s)} as already described, they do not appear in the following formula.

In FIG. 2A, the input X to the output Y ₁
And the transfer function from the disturbance D ₁ to the output Y ₁ is given by the following equation.

(Y ₁ / X) = P _{mA (S)} / {1 + δ · (1-P _{mA (S)} · _{HA (S)} )} (5) (Y ₁ / D ₁ ) = P _{mA (S)}・ (1-P _{mA (S)}・_{HA (S)} ) / {1 + δ ・ (1-P _{mA (S)}・_{HA (S)} )} (6) Note that δ _A is It represents the error between the ideal model and the actual product (tracking actuator), and is expressed by the following equation.

Δ _A = {P _{mA (S)} −P _{A (S)} } / P _{A (S)} (7) In the equations (5) and (6), the coefficient term of the error δ _A , in other words, the error δ _The term that suppresses _A is (1−P _{mA (S)} · _{HA (S)} ), and the smaller this is, the smaller the influence of the error or the disturbance.
This term is sufficiently small that | δ _A · {1-P _{mA (S)} · H _{A (S)} } |
When << 1 can be regarded, the equation (5) becomes (Y ₁ / X) ≈P _{mA (S)} (5 ′), and it can be seen that almost the same transfer characteristics as the ideal model can be obtained.

At this time, the equation (6) becomes (Y ₁ / D ₁ ) ≈P _{mA (S)} · {1-P _{mA (S)} · _{HA (S)} } (6 ′). , | (1-P _{mA (S)} · _{HA (S)} ) |, the behavior of the tracking actuator due to disturbance is suppressed.

In particular, at a frequency where H _{A (S)} can be regarded as almost equal to the inverse characteristic of P _{mA (S)} , | {1-P _{mA (S)} · H
_{A (S)} } | becomes almost 0, and the influence of disturbance and error can be suppressed almost completely.

In exactly the same manner as above, Y ₁ is Y ₂ , D ₁ is D ₂ , P _{mA (S)} is P _{mL (S)} , P _{A (S)} is P _{L (S)} , and H.
_By replacing _{A (S)} with H _{L (S)} and δ _{A (S)} with δ _{L (S)} ,
Since the behavior of the linear motor system shown in FIG. 2 (b) can also be described, the description of the linear motor system is omitted.

Subsequently, P _{mA (S)} , H _{A (S)} and P _{mL (S)} , H
A graph of the transfer function of each of these expressions when a specific transfer function is set in _{L (S)} is derived.

First, when P _{mA (S)} and H _{A (S)} have characteristics as shown in FIGS. 3 (a) and 3 (b), their transfer functions are expressed by the following equations (8) and (9), respectively. It becomes like. However, s is a Laplace operator and is represented as s = 2πf.

P _{mA (S)} = K ₀ · ω ₀ ² / (s ² + 2ζ ₀ · ω ₀ · s + ω ₀ ² ) (8) (where ω ₀ = 2πf ₀ ) H _{A (S)} = ( 1 / K ₀ ) · {1+ (s / ω ₀ )} / {1+ (s / ω ₁ )} (9) (where ω ₁ = 2πf ₁ , ω ₁ >> ω ₀ ) However, generally servo Since it is not preferable that the tracking actuator (driving means) has a resonance peak in the system, the damping number ζ ₀ = 1 is assumed in FIG. 2 (b).
Also follows this.

Further, since P _{mA (S)} is an ideal model of the characteristics of a spring-mass actuator, a flat gain characteristic is clearly obtained at a frequency of f ₀ or less from the equation (a) or (8) of FIG. Have. The transfer characteristic of the feedback means H _{A (S)} shown in FIG. 3 (b) or (9) is a first-order advance / delay characteristic, which does not match the inverse characteristic of P _{mA (S)} , but f ₀ At the following frequencies, it has the same flat gain characteristics as P _{mA (S)} , and at this frequency range, P _{mA (S)}
Since H _{A (S)} ≈1, the influence of disturbances and errors can be suppressed almost completely.

When the above model is used, the error δ _A
A suppression section when _{{1-P mA (S)} · H A (S)} are expressed as R _{A (S),} the frequency of _{s << ω 1 R A (S} ) is the following formula (10) Can be approximated by

[0049] _{R A (S) = {1} -P mA (S) · H A (S)} ··· (10) ≒ s / (s + ω 0) ··· (10 ') which omega ₀ i.e. f _This is a first-order high-pass filter format with ₀ as a break point, and its gain curve is as shown in FIG. (10)
According to the equation or FIG. 4, in the frequency region of ω <ω ₀ , that is, f <f ₀ , | _{RA (S)} | = | {1-P _{mA (S)} · _{HA (S)} } | < Therefore, the error δ between the real model and the ideal model is suppressed accordingly.

Next, the improvement of resistance to disturbance such as vibration will be described.

The transfer function from the disturbance D ₁ to the output displacement Y ₁ of the tracking actuator (driving means) has already been shown by the equation (6). If the error between the tracking actuator and the transfer characteristic of the model is δ _A , Disturbance D ₁
To the displacement Y _{1 of the} movable portion of the tracking actuator, the equation (6) becomes the following equation (11).

(Y ₁ / D ₁ ) = P _{mA (S)} · { _{RA (S)} / (1 + δ _{A (S)} · _{RA (S)} )} (11) At this time, δ _{A (S )} Is extremely large | R _{A (S)}・ δ _{A (S)} | ≧ 1
Except in the case of, the equation (11) becomes (Y ₁ / D ₁ ) ≈P _{mA (S)} · _{RA (S)} ... (11 ′) and the disturbance D ₁ is the _{RA (S)} Will be suppressed accordingly.

On the other hand, in the case not according to the present invention, the above-mentioned FIG.
If the state feedback element 53 is removed in (a), the following equation (12) is obtained.

(Y ₁ / D ₁ ) = P _{A (S)} (12) Equation (12) is the transfer characteristic itself of the actual tracking actuator, and no error suppression term is applied. Therefore, by comparing these equations (11 ′) and (12), the effect of the disturbance can be obtained by applying the present invention to the equation (10 ′) as long as P _{mA (S)} is not set to be extremely larger than P _{A (S).} It can be seen that it is reduced according to R _{A (S)} shown in FIG.

Next, regarding the linear motor, P _{mL (S)} , H
_{When L (S)} has the characteristics shown in FIGS. 5 (a) and 5 (b), the transfer functions thereof are expressed by the following equations (13) and (14), respectively.

P _{mL (S)} = K _L · ω _L ² / (s ² + 2ζ _L · ω _L · s + ω _L ² ) (13) (where ω _L = 2πf _L ) H _{L (S)} = ( 1 / K _L ) · {1+ (s / ω _L )} / {1+ (s / ω ₁ )} (14) (where ω ₁ = 2πf ₁ , ω ₁ >> ω _L ) where For the same reason as in the case of the tracking actuator described above, at a frequency below f _L , P
P _{mL (S)} and _{HL (S)} are set so that _{mL (S)} · _{HL (S)} = 1.

Similar to the case of the above-mentioned tracking actuator, ζ _L = 1 is considered, and FIG. 5A is also written accordingly. In addition, when using the above linear motor system model, the ideal model and the actual product (linear motor)
The error of is expressed by the following equation δ _L = {P _{mL (S)} −PL _(S) } / PL _(S) (15) If the model of the linear motor system is set as above, It has the same characteristics as the actuator model. The term (1-P _{m (S)} · H _(S) ) that suppresses δ _L is R
_{Expressed as L (S)} , R _{L (S)} can be approximated by the following equation (16) at a frequency of s << ω ₁ .

_{RL (S)} ≈s / (s + ω _L ) ... (16) Since the gain curve is as shown in FIG.
In the frequency range of <ω _{L, that} is, f <f _L , the error δ _L between the real model and the ideal model is suppressed. Therefore, even if the original characteristic of the linear motor is that of an inertial mass system having a gain gradient of −40 dB / dec in the entire frequency band as shown in FIG. 22, this and P _{mL (in} FIG. 5A) _S)
The difference between the characteristics of the linear motor and the characteristic of the linear motor is suppressed as an error δ _L, and as a result, the characteristics of the linear motor are driven so as to match P _{mL (S)} .

As for the improvement of resistance to disturbance such as vibration, R _{L (S)}
The influence of disturbance is reduced accordingly.

Since the tracking actuator and the linear motor respectively match the outputs of the ideal model, there is no gain fluctuation due to fluctuations in these sensitivities. In particular, the resonance peak does not occur in the tracking actuator and the resonance frequency does not change. Therefore, the parallel servo system can be kept stable without making complicated adjustments such as velocity feedback to the tracking actuator system.
The low-frequency gain of the tracking actuator system does not fluctuate, and it is possible to prevent the occurrence of offset caused by the increase in the gain sharing of the tracking actuator system as seen in the conventional example. Further, the influence of disturbance such as vibration is suppressed according to _{RA (S)} or _{RL (S),} so that it is possible to construct a servo system having high rigidity against disturbance.

Subsequently, the state feedback elements 53, 6
When the characteristics H _{A (S)} and _{HL (S) of 3} are set differently, for example, the characteristics shown in FIG. 7 (b) for the tracking actuator and the characteristics shown in FIG. 9 (b) for the linear motor. Suppression of the error terms δ _A and δ _L or the disturbances D ₁ and D ₂ will be continuously described. For comparison, the transfer characteristics P _{mA (S)} and P _{mL (S)} of the models of the tracking actuator and the linear motor shown in FIGS. 7 (a) and 9 (a) are expressed by the above equation (8) or FIG. It is the same as the equations (a) and (13) or FIG. 5 (a). H _{A (S)} and H _{L (S)} are P _{mA (S)} · H _{A (S)} = 1 and P respectively in the low frequency region due to the reason described above.
Set so that _{mL (S)} · _{HL (S)} = 1.

The transfer function of H _{A (S)} shown in FIG. 7B is expressed by the following equation (17).

H _{A (S)} = (1 / K ₀ ) · {1+ (s / ω ₀ )} ² / {1+ (s / ω ₁ )} ² (where ω ₁ = 2πf ₁ , ω ₀ << ω ₁ ) (17) H _{A (S)} shown in the FIG. 7 (b) or (17) equation is f ₁
Is almost the same as the inverse characteristic of P _{mA (S) in} the frequency range up to, and in this frequency range | {1-P _{mA (S)}・_{HA (S)} } |
≈0. Therefore, the influence of disturbance or error is suppressed in a wider frequency range than in the case where H _{A (S)} has the first-order advance / delay characteristic.

At this time, R which is a coefficient term such as the error term δ _{A
A (S)} = {1-P _{mA (S)} · _{HA (S)} } is given by the equation (18).

[0065] such as _{R A (S) = (s} 2 +2 · ω 1 · s) / (s 2 +2 · ω 1 · s + ω 1 2) ··· (18) and whose gain curve is shown in FIG. 8 , (F ₁
/ 2) is approximated to a high-pass filter type whose break point is
In (f _1/2) or less in the frequency _{domain, | R A (S) |} = | {1-P mA (S) · H A (S)} | <1 , and the error between the actual and the ideal model δ The effects of _A and the disturbance D ₁ will be suppressed.

Transfer function H _{A (S) of} feedback element 53
When the equation (9) and the first-order advance-delay characteristics shown in FIG. 3 (b) are used, | _{RA (S)} | <1 is limited to frequencies below f _0. It was However, if H _{A (S)} has the second-order lead-lag characteristic shown in equation (17) and FIG. 7 (b),
by setting the f _1> 2f _0, the _{| R A (S) | <} 1 and becomes frequency is expanded to (f _1/2), the suppression effect on the error or disturbance becomes higher.

Next, the case where the state feedback element 63 of the linear motor system having the transfer characteristic _{HL (S)} of the secondary advance / delay characteristic shown in FIG. 9B is used in the same manner will be described. . For comparison, as described above, the transfer characteristic P _{mL (S)} of the linear motor model is the same as the above equation (13).

The transfer function of _{HL (S)} shown in FIG. 9B is expressed by the following equation (19).

H _{L (S)} = (1 / K ₀ ) · {1+ (s / ω _L )} ² / {1+ (s / ω ₁ )} ² (where ω ₁ = 2πf ₁ , ω _L << ω ₁ ) (19) Also at this time, like the tracking actuator,
R _{L (S)} = {1-P _{mL (S)} · H which is a coefficient term such as the error term δ _{L
L (S)} } is given by equation (20).

R _{L (S)} = (s ² + 2 · ω ₁ · s) / (s ² + 2 · ω ₁ · s + ω ₁ ² ) ··· (20) Further, the gain curve is as shown in FIG. (F
It is approximated _1/2) to the high-pass filter type of the break point, (in f _1/2) or less in the frequency _{domain, | R L (S) |} = | {1-P mL (S) · H L ( _S) } | <1 and the effects of the error δ _L between the real model and the ideal model and the disturbance D ₂ are suppressed.

From these, the error δ between the real model and the ideal model
_A, effect of [delta] _L and the disturbance D _1, D ₂ are, (f _1/2) is suppressed at frequencies below, except that the variation of the resonance frequency and damping speed of the tracking actuator (resonance peak) is absorbed, further f _X <If you set the (f _1/2), the frequency f _X the gain of the linear motor system and a tracking actuator system equals can always be made constant, in addition to immobilization of both gain sharing can be achieved , F _X can be prevented from destabilizing the parallel servo system.

Second Embodiment Next, a second embodiment of the tracking control device according to the present invention will be described.

In the second embodiment, as shown in FIGS. 11 and 12, in the tracking actuator control unit 50 and the linear motor control unit 60, the outputs of the state feedback elements 53 and 63 are output to the drive means 1 through the drive circuits 9 and 10. 2 (tracking actuator, linear motor) is newly provided with a second path for feedback, and feedback gain elements 54 and 64 having gains β _A and β _L are provided in the second path. First of
Tracking actuator control unit 5 in the embodiment
0, the performance of suppressing errors and disturbances is further improved as compared with the linear motor control unit 60.

In the tracking actuator control section 50 of FIG. 11, the transfer function from the input X from the outside and the disturbance D ₁ to the output Y ₁ of the tracking actuator 1 is expressed by equations (21) and (22), respectively.

(Y ₁ / X) = P _{mA (S)} / [1 + δ _A · {(1-P _{mA (S)} · _{HA (S)} ) / (1 + β _A · P _{mA (S)} · _{HA ( S)} )}] (21) (Y ₁ / D ₁ ) = P _{mA (S)} / [1 + δ _A・ {(1-P _{mA (S)}・_{HA (S)} ) / (1 + β _A・P _{mA (S)} · _{HA (S)} )}] × [(1-P _{mA (S)} · _{HA (S)} ) / (1 + β _A · P _{mA (S)} · _{HA (S)} )] ... (22) However, [delta] _a is shown already (7) is a term representing the error between the real tracking actuator and the ideal model.

Error δ _{A in} equations (21) and (22)
If the coefficient term of is expressed as R _{A (S)} as in the first embodiment, the following formula (23) is obtained, R _{A (S)} = {(1-P _{mA (S)} · H _{A (S)} ) / (1 + β _A · P _{mA (S)} · H _{A (S)} )} (23) | R _{A (S)} | <1 if error δ _A and disturbance D ₁ are suppressed This is similar to the first embodiment. Further, as in the case of the first embodiment, the description of the behavior of the linear motor control unit 60 is exactly the same as that of the tracking actuator control unit 50 described above, and therefore the description thereof is omitted.

Next, P _{mA (S)} , H _{A (S)} and P _{mL (S)} , H
_A specific transfer function for _{L (S)} is set to be the same as that in the first embodiment, and the characteristics at that time are obtained and compared.

First, regarding the tracking actuator system, P _{mA (S)} is of the formula (8), and H _{A (s)} is of the formula (9) and the first-order advance / delay characteristics shown in FIG. 3 (b). When a transfer function is given, _{RA (S)} is given by the equation (24), and its gain curve is as shown in FIG.

R _{A (S)} = β _A · ω ₀ / (s + β _A · ω ₀ ) ... (24) R _{A (S)} at this time has (β _A · f ₀ ) as a break point 1 The following high-pass filter type is used. Since | _{RA (S)} | <1 at frequencies below (β _A · f ₀ ), errors and disturbances are suppressed in this frequency range.

Similarly for the linear motor system, P
_{mL (S)} is of the formula (13), and _{HL (S)} is (14)
When the transfer function of the first-order lead-lag characteristic shown in FIG. 5B is given, the suppression term R _{L (S)} of the error term δ _L between the real (linear motor) and the ideal model is (25) The expression and its gain curve are as shown in FIG.

[0081] _{R L (S) = β L} · ω L / (S + β L · ω L) ··· (25) at this time R _{L (S)} is 1 to the break point of the (β _L · f _L) This is the following high-pass filter type, and | _{RL (S)} | <1 at frequencies below (β _L · f _L ), so errors and disturbances are suppressed in this frequency range.

A case where the tracking actuator and the linear motor using the above model are applied to a parallel servo system will be described.

As in the case of the first embodiment, the tracking actuator system and the linear motor system exhibit the same characteristics as the ideal model, so that it is possible to stabilize the parallel servo system without making complicated adjustments. Since the fluctuation of the resonance point of the tracking actuator and the fluctuation of the gains of both the tracking actuator system and the linear motor system can be suppressed, the tracking actuator system and the linear motor system can always be driven with a constant gain sharing. If f _X <(β _A · f ₀ ) and (β _L · f _L ) are set, the frequency f _{X at} which the gains of the linear motor system and the tracking actuator system become equal, and the cutoff frequency of the tracking actuator system It is possible to keep f _C constant, and it is possible to absorb variations in the servo tracking performance of each device.

When the transfer functions H _{A (s)} and H _{L (S)} of the feedback elements 53 and 63 having the same first-order advance / delay characteristics are used, the error and the disturbance are suppressed in the first embodiment. Is f _{0 for the} tracking actuator system and f _{L for the} linear motor system
In contrast to the following frequencies, the second embodiment is more preferable because it can suppress the frequencies up to β _A times and β _L times respectively.

Next, in H _{A (S)} , equation (17) or FIG.
In the case where the transfer function of the second-order advance-delay characteristic shown in ₍₃₎ is given, _{RA (S)} is obtained and the suppression characteristics of error and disturbance are compared.

R _{A (S)} at this time is represented by the equation (26) and has a kind of high-pass filter format. Although the shape of the gain curve is slightly different depending on β, for example, when β _A > 3 (that is, {(1 + β _A ) ^1/2 · ω ₁ }> 2ω ₁ ), it is approximately represented in FIG. It will be like that.

R _{A (S)} = (s ² + 2 · ω ₁ · s) / {s ² + 2 · ω ₁ · s + (1 + β _A ) · ω ₁ ² } (26) R in the equation (26) _{A (S)} becomes | _{RA (S)} | <1 at a frequency equal to or lower than {f ₁ (1 + β _A ) ^1/2 }, which has an effect of suppressing an error or disturbance.

Next, regarding the linear motor system (19)
The same result as that of the tracking actuator system can be obtained in the case where the equation or the transfer function of the second-order lead-lag characteristic shown in FIG. 9B is provided. That is, _{RL (S)} is (2
This is represented by the equation 7), and the gain curve has a kind of high-pass filter format as shown in FIG.

_{RL (S)} = (s ² + 2 · ω ₁ s) / {s ² + 2 · ω ₁ s + (1 + β _L ) · ω ₁ ² } (27) _{RL (S)} is _{{f 1 · (1 + β} L) 1/2} in the following the frequency _{| R L (S) | <} 1 . Therefore, in the system shown in FIG. _{12, {f 1 · (1} + β L) 1/2} There is an effect of suppressing errors and disturbance up to the following frequencies.

When H _{A (S)} and H _{L (S)} of the same second-order advance / delay characteristics are set in the first embodiment, both the tracking actuator system and the linear motor system have | _{RA (S)} |
<1, | R L (S ) | < since 1 as made had the following frequency both (f _1/2), also more suppression effect for errors and disturbances according to the second embodiment It turns out to be high.

In the parallel servo system using the tracking actuator and the linear motor as described above, the tracking actuator system is {f ₁ (1 + β _A ) ^1/2 }.
Up to {f ₁ · (1 + β _L ) ^1/2 }, the linear motor system has the effect of suppressing the error terms δ _A , δ _L , and the disturbances D ₁ and D ₂ , particularly f _X <{f ₁ · (1 + β _A ) ^1/2 }, {f ₁ · (1 + β _L )
^If set to ^1/2 }, the frequency f at which the gains of the linear motor system and tracking actuator system become equal
_X can always be constant.

[0092]

According to the drive means control device of the present invention, a model having ideal transfer characteristics desired for the drive means such as the fine actuator and the coarse actuator is provided, and the output difference between this and the actual drive means. Based on this output difference, the actual driving means is driven so that its output matches the output of the model. Therefore, it is possible to obtain a driving means having characteristics substantially equivalent to those of the model having ideal transfer characteristics. I can. For this reason, fluctuations such as the resonance frequency and damping number of the tracking actuator that is a fine actuator and the linear motor that is a coarse actuator are absorbed,
The low range gain of the servo system can be made constant. Further, the influence of disturbance such as vibration is suppressed, and the servo rigidity or vibration resistance performance can be improved.

Particularly in a tracking servo system of an optical disk device in which these tracking actuators and a linear motor are combined to form a parallel servo system, the low frequency gain increases due to the fluctuation of the resonance frequency of the tracking actuator, and the offset of the tracking error signal is increased. It is possible to prevent inconveniences such as increase in the generated amount, and it is possible to keep the frequency at which the gain of the tracking actuator system and the gain of the linear motor system are equal. Can be suppressed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a first embodiment of a tracking control device according to the present invention.

FIG. 2A is a configuration diagram showing a tracking actuator control unit 50 in FIG. (b) is a block diagram showing the linear motor control unit 60 of FIG. 1.

FIG. 3 is a diagram showing gain curves of an ideal model 51 and a state feedback element 53 of the tracking actuator 1 used in the tracking actuator control section 50 of FIG. 2 (a).

FIG. 4 Ideal model 51 and state feedback element 5
4 is a diagram showing a gain curve of a coefficient term showing an error and disturbance suppression characteristic when the characteristic of No. 3 is that of FIG. 3. FIG.

5 is a diagram showing gain curves of an ideal model 61 and a state feedback element 63 of the linear motor 2 used in the linear motor control unit 60 of FIG. 2 (b).

FIG. 6 Ideal model 61 and state feedback element 6
6 is a diagram showing a gain curve of a coefficient term showing an error or disturbance suppression characteristic when the characteristic of FIG.

7 is a diagram showing another gain curve of an ideal model 51 of the tracking actuator 1 and a state feedback element 53 used in the tracking actuator controller 50 of FIG. 2 (a).

FIG. 8: Ideal model 51 and state feedback element 5
FIG. 9 is a diagram showing a gain curve of a coefficient term showing an error or disturbance suppression characteristic when the characteristic of No. 3 is that of FIG. 7.

9 is a diagram showing another gain curve of an ideal model 61 of the linear motor 2 and a state feedback element 63 used in the linear motor control unit 60 of FIG. 2 (b).

10 is a diagram showing a gain curve of a coefficient term showing an error or disturbance suppression characteristic when the characteristics of the ideal model 61 and the state feedback element 63 are those shown in FIG. 9;

FIG. 11 is a block diagram showing the configuration of another embodiment of the tracking actuator control unit 50.

FIG. 12 is a block diagram showing the configuration of another embodiment of the linear motor control unit 60.

13 is a diagram showing a gain curve of a coefficient term showing an error or disturbance suppression characteristic when the characteristics of the ideal model 51 and the state feedback element 53 in the tracking actuator control unit 50 of FIG. 11 are the same as those of FIG. .

14 is a diagram showing a gain curve of a coefficient term indicating an error or disturbance suppression characteristic when the characteristics of the ideal model 61 and the state feedback element 63 in the linear motor control unit 60 of FIG. 12 are those of FIG. .

15 is a diagram showing a gain curve of a coefficient term indicating an error or disturbance suppression characteristic when the characteristics of the ideal model 51 and the state feedback element 53 in the tracking actuator control unit 50 of FIG. 11 are those of FIG. is there.

16 is a diagram showing a linear motor control unit 60 of FIG.
FIG. 10 is a diagram showing a gain curve of a coefficient term indicating an error or disturbance suppression characteristic when the characteristics of the ideal model 61 and the state feedback element 63 are those shown in FIG. 9.

FIG. 17 is a configuration diagram of a parallel servo system of the optical disc device.

FIG. 18 is a block diagram of a parallel servo system of the optical disc device.

FIG. 19 is an explanatory diagram of a physical model of a spring-mass system of the tracking actuator 1.

20 is a diagram showing a gain curve of a transfer function of the tracking actuator 1 represented by the physical model of FIG.

FIG. 21 is an explanatory diagram of a physical model of the linear motor 2 that is an inertial mass system.

FIG. 22 is a diagram showing a gain curve representing a transfer characteristic of the linear motor 2.

FIG. 23 is a diagram showing a gain curve of a phase lead compensation circuit.

FIG. 24 is a diagram showing a gain curve of a first-order low-pass filter.

FIG. 25 is a diagram showing gain and phase curves of a parallel servo system.

FIG. 26 is a diagram showing gain and phase curves of a parallel servo system when the damping number of the tracking actuator is small.

FIG. 27 is a configuration diagram of a method of increasing the damping number by feeding back the speed of the movable portion of the tracking actuator.

FIG. 28 is a diagram showing gain and phase curves of the parallel servo system when the resonance frequency of the tracking actuator is lowered.

FIG. 29 is a diagram showing a gain curve of the parallel servo system when the gain of the linear motor system is reduced.

[Explanation of symbols]

 1: Tracking actuator 2: Linear motor 3: Disk 4: Track 5: Light beam spot 6: Optical pickup 7, 8: Phase compensation circuit 9, 10: Drive circuit 11: Lens 12: Speed sensor 13: Amplifier 50: Tracking actuator Control unit 60: Linear motor control unit 51, 61: Ideal model 52, 62: Comparison means 53, 63: State feedback element 54, 64: Feedback gain 55, 65: Displacement sensor TE: Tracking error signal

Claims (4)

[Claims] 1. A tracking error signal is generated by detecting a deviation amount of a light generating means and an optical system for guiding light from the light generating means to an optical disk and condensing the light beam spot as a light beam spot from a predetermined track of the optical disk. A tracking error signal detection system and a tracking control device for tracking a light beam spot on the predetermined track by driving all or part of the optical system with a coarse actuator and a fine actuator based on the tracking error signal The apparatus includes a first drive circuit for inputting a first fine actuator drive signal based on the error signal to drive the fine actuator, and a fine actuator to which a second fine actuator drive signal based on the error signal is input. Of the desired transfer characteristics of A pseudo first model, first comparing means for comparing the output of the displacement sensor for detecting the displacement of the dense actuator with the output of the first model, and the first comparing means.
A first feedback circuit for feeding back the output of the comparison means to the input side of the first drive circuit by a first predetermined multiplication and feeding it back to the input side of the first model for a second predetermined multiplication. A fine actuator control unit, a second drive circuit for inputting a first coarse actuator drive signal based on the error signal to drive the coarse actuator, and a second coarse actuator drive signal based on the error signal A second model for electrically simulating a desired transfer characteristic of the coarse actuator, and second comparing means for comparing the output of the displacement sensor for detecting the displacement of the coarse actuator with the output of the second model, The output of the second comparing means is fed back to the input side of the second drive circuit by the third predetermined multiplication and fed back, and is fed to the input side of the second model by the fourth predetermined multiplication and fed. Tracking control device of an optical disk apparatus characterized by having a coarse actuator control unit and a second feedback circuit for back. 2. The tracking control device according to claim 1, wherein the transfer characteristic of at least one of the first and second feedback circuits is a first-order lead-lag characteristic. 3. The tracking control device according to claim 1, wherein the transfer characteristic of at least one of the first and second feedback circuits is a quadratic advance / delay characteristic. 4. The first predetermined multiple of the first feedback circuit is (α ₁ + β ₁ ) times, the second predetermined multiple is α ₁ times, and the third of the second feedback circuit is
Is a predetermined multiple of (α ₂ + β ₂ ) times, and the fourth predetermined multiple is α ₂
The tracking control device according to claim 1, wherein the tracking control device is doubled.