JPH06187044A - Positioning controller - Google Patents

Abstract

PURPOSE:To change the acceleration pattern of an object to be controlled by providing plural target acceleration tables and plural target speed tables and generating target acceleration and target speed while adjusting a weight coeffi cient. CONSTITUTION:A first target speed generating means 7 and a second target speed generating means 9 make prescribed formulas into tables. Then, a coefficient multiplying means 6 sets prescribed values to coefficient multiplying means 8 and 10 when starting a seek operation. Then, the target speed is calculated by calculating the sum of products while referring to two tables based on the approximate value of normalization time. First and second acceleration feed forward generating means 11 and 13 make prescribed formulas into tables and while referring to these tables, acceleration feed forward is calculated. Then, the optimum weight coefficient can be decided from a seek distance, seek time and natural frequency of residual vibrations to be suppressed. When starting the seek operation, respective coefficients can be calculated by inputting the seek distance to be coefficient multiplying means 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a positioning control device and a control method therefor, and more particularly to a positioning control device suitable for use in head positioning control for optical and magnetic disk devices.

[0002]

2. Description of the Related Art Conventionally, in a magnetic disk device, a speed control method which follows one target speed curve stored in a memory is often used as an actuator control method for moving and positioning a magnetic head to another track. It had been. This method will be described below with reference to the drawings.

FIG. 29 schematically shows the relationship between the distance to the target track (hereinafter referred to as remaining seek length) and the head speed during two seek operations with different target speed curves and moving distances. While the head is moving, the target speed is output by referring to the above-mentioned memory based on the remaining seek length, and at the same time, the head speed is detected and the value is fed back. The target speed curve specifies only the speed in the deceleration section after the maximum speed, and in the acceleration section until reaching the maximum speed, the maximum acceleration determined by the performance of the actuator is performed, and the pseudo shortest time as shown in FIG. It was generating control acceleration. FIG. 31 shows the time change of the head speed.

The above-mentioned conventional technique has the following problems.

First, in this method, the maximum acceleration is performed irrespective of the moving distance, so that the natural vibration mode of the mechanical system is excited by the reaction force of the thrust generated by the actuator during acceleration. In particular, a mode of several hundred Hz in which the disk vibrates greatly is excited to cause residual vibration, which causes a long access time.

Secondly, in this method, since the rate of change in acceleration at the start of acceleration and at the time of switching from acceleration to deceleration is large, a high frequency vibration of several kHz occurs in the head support mechanism, and the head is struck by the vibration during seek. There was a risk of crash.

Thirdly, in this method, the static reduction of the flying height (subtraction) caused by the head acceleration and the velocity is maximized almost at the same time, and there is a risk of head crash.

Regarding the above-mentioned second problem, Eye Triple E Transactions on Magnetics, Vol. 24, No. 3, issued in May 1988 (IEEE Transactions
onMagnetics vol.24 No.3) "Minimizing
Power Dissipation In Disk File Actuator "(Minimizing Power Dis
sipation in a Disk File Actuator) and the 68th Ordinary General Meeting of the Japan Society of Mechanical Engineers published in 1991, Volume C (Vol. C), "Part 1 of vibration control technology in magnetic head positioning system" 2 ”,
A method of designing a trajectory that minimizes the rate of change in acceleration during seek operation is disclosed. Further, in the latter known example, in order to generate a trajectory corresponding to an arbitrary seek distance, the designed trajectory is formulated as a function of a normalization time (a real time during seek is a value obtained by normalizing the seek operation time). The method of doing is disclosed. Further, Japanese Laid-Open Patent Publication No. 2-195581 discloses a method of using a smooth target speed based on a sinusoidal curve, also regarding the second problem.

Although the head positioning control system of the magnetic disk device has been described above as an example, suppressing the structural vibration during the movement of the movable part and at the time of stoppage thereof is an important issue in other positioning control devices as well. Is.

[0010]

The three known examples described in the prior art are all effective methods for smoothing the acceleration waveform of the controlled object and reducing mechanical high frequency vibration in a wide frequency range. is there. However, it is also necessary to reduce the natural vibration having a frequency equal to or lower than the servo band of the control system as mentioned in the first problem of the conventional technique.
Therefore, a first problem to be solved by the present invention is to provide a control method aiming at and suppressing a specific natural frequency.

A second object is to provide a control method capable of adjusting the acceleration pattern according to the moving distance of the controlled object. This problem is related to the first problem described above. Generally, the time required for the movement tends to increase according to the movement distance. The acceleration pattern that most effectively suppresses vibration is determined by the ratio of the frequency and the reciprocal of the travel time. Therefore, it is necessary to adjust the acceleration pattern when the moving distance changes.

A third object is to provide a control method for generating a smooth acceleration on a controlled object. The purpose of this is to reduce high frequency vibration in a wide frequency range as in the known example of the prior art.

A fourth object is to provide a control method in which the time point of maximum acceleration and the time point of maximum velocity are apart from each other during movement of the controlled object. The purpose is to reduce the sinking of the head mentioned in the third problem of the conventional technique.

A fifth problem relates to a control method for determining a target value based on the position of the controlled object. In the magnetic disk device, the value of the target speed is determined based on the head position as described in the related art. Compared to the method that determines the target speed based on the time from the start of movement regardless of the head position, this method reduces the fluctuation of the initial state at the time of switching to the track following position control system even if the loop gain changes. It is effective because it can be kept small. In this regard, it was designed in “Vibration control technology in magnetic head positioning system (2)” by Volume C (vol.C), the 68th Ordinary General Meeting of the Japan Society of Mechanical Engineers, published in 1991. When the relationship between the normalized position of the trajectory and the normalized time is constant regardless of the seek distance, this relationship is tabulated, the normalized time is calculated from the observed head position, and the target speed is calculated from this normalized time. It discloses how to decide. However, when trying to suppress a specific natural vibration, the relationship between the normalized time and the normalized position differs depending on the seek distance for the same reason as in the second problem. Therefore, even in such a case, it is an object to provide a control method for determining the target value based on the position of the controlled object.

[0015]

In order to solve the above problems, (1) a specific natural vibration of a mechanical part is suppressed and a smooth acceleration pattern of a controlled object is designed, and (2) the acceleration It is necessary to be able to adjust the target value so that the target value for generating the pattern is generated based on the position of the controlled object, and (3) the acceleration pattern can be changed according to the moving distance. The means therefor will be described in detail below.

[Method of Designing Acceleration Pattern for Suppressing Specific Natural Vibration] The driving force generated by the actuator when the movement distance is L and the movement time is Te is a polynomial function of the operation time t.

[0017]

[Equation 1]

The coefficients a ₀ , a ₁ , ..., _Am (m is an integer of 0 or more) are determined so as to satisfy the following conditions.

(1) If the position of the controlled object is x _c (t) and the velocity is v _c (t), these satisfy the following end point conditions.

[0020]

[Equation 2]

[0021]

[Equation 3]

[0022]

[Equation 4]

[0023]

[Equation 5]

(2) At time t = Te, the vibration energy of the eigenmode for suppressing the vibration is 0. This condition can be expressed by a mathematical formula according to the following procedure.

First, a dynamic model is constructed by using the positioning controller as a linear coupled vibration system having n degrees of freedom. The equation of motion is as follows.

[0026]

[Equation 6]

Here, X is a physical coordinate vector, F is a driving force vector, [M] is a mass matrix, [C] is a damping matrix, and [K] is a rigidity matrix. next,
Solving the generalized eigenvalue problem for [M] and [K], n
The next orthonormal mode is obtained and coordinate conversion is performed.
Since the damping of mechanical vibration is generally small, if the damping term is ignored, the following equation of motion for the non-coupling vibration mode is obtained.

[0028]

[Equation 7]

Where Q is a generalized coordinate vector,

[0030]

[Equation 8]

Is the diagonal matrix of the square of the natural frequency of each mode,

[0032]

[Equation 9]

Represents the transposed matrix of the eigenmode matrix. now,
Consider suppressing the i-th mode in the above equation. The external force acting on this mode is proportional to the equation 1 even if the coordinates are converted, so that the equation of motion is given by the following equation, where the proportional coefficient is k.

[0034]

[Equation 10]

Further, the vibrational energy of this eigenmode is

[0036]

[Equation 11]

Therefore, Ei (t = Te) = 0 is equivalent to the following two conditions.

[0038]

[Equation 12]

And

[0040]

[Equation 13]

This is the mathematical expression of condition (2). Among the above conditions (1) and (2), the expressions 2 and 3 are

[0042]

[Equation 14]

(Where m _a is the mass of the controlled object)

[0044]

[Equation 15]

This is a condition for determining the integration constant of, and does not affect the values of the coefficients a ₀ , a ₁ , ..., _Am (m is an integer of 0 or more). Therefore, the above coefficients are determined by Equation 4, Equation 5,
Since there are four conditions of Expressions 12 and 13, m = 3, and Expression 1 becomes a quintic function. If the equation 1 which is a driving force function is designed based on the above conditions, the acceleration is switched to the deceleration at time t = 0.5Te, and the acceleration / deceleration pattern symmetrical with respect to this time point, that is,

[0046]

[Equation 16]

Is obtained. In this formula, time 0 ≦ t ≦ Te
In the range of f _c (t) = 0, t = 0, 0.5 Te,
Since only Te should be used, the following equation holds as a condition for the weighting coefficient γ for that.

[0048]

[Equation 17]

The target acceleration a _c (t) of the controlled object is given by the following expression by dividing the equation 16 by the mass m _a of the controlled object.

[0050]

[Equation 18]

Here, t / Te is the moving time Te at time t.
The normalized time is normalized by, and changes from 0 to 1 during movement regardless of the movement distance. Target speed of controlled object v _c
(T) is expressed by the following equation by integrating the equation (18).

[0052]

[Formula 19]

Further, the target position x _c (t) of the controlled object is given by the following equation by integrating equation (19).

[0054]

[Equation 20]

If both sides of the equation (20) are normalized by the moving distance, the following equation is obtained.

[0056]

[Equation 21]

In order to generate the target trajectory on the basis of the position of the controlled object with respect to an arbitrary moving distance, the observed position x is normalized by the moving distance to obtain a normalized position δ, and from this value, the equation 21 The normalized time t / Te may be obtained by the inverse function of, and the target velocity and the target acceleration may be calculated using the normalized time. However, Equation 21 depends on the weighting factor γ,
Further, since this γ changes depending on the moving distance, it cannot be simply tabulated. Therefore, the term in equation 21 is

[0058]

[Equation 22]

[0059]

[Equation 23]

It is defined as Since the expressions 22 and 23 are constant irrespective of the moving distance, two normalized time tables are created by the respective inverse functions and stored in the memory. By referring to both tables based on the normalized position δ and substituting them into the following equation, an accurate approximate value of the normalized time is calculated.

[0061]

[Equation 24]

Here,

[0063]

[Equation 25]

Is a value obtained by referring to the table.

In order to calculate the target acceleration, the terms in Eq.

[0066]

[Equation 26]

[0067]

[Equation 27]

It is defined as Since the formulas 26 and 27 are constant regardless of the moving distance, two target acceleration tables are created,
Store in memory. The target acceleration is calculated by referring to both tables based on the above-mentioned approximated normalized time and substituting them in the following equation.

[0069]

[Equation 28]

Further, in order to calculate the target speed,
The terms in

[0071]

[Equation 29]

[0072]

[Equation 30]

It is defined as Since the numbers 29 and 30 are constant regardless of the moving distance, two target speed tables are created and stored in the memory. The target speed is calculated by referring to both tables based on the normalized time approximate value and substituting into the following equation.

[0074]

[Equation 31]

The coefficients by which the table reference value is multiplied by the equations 24, 28 and 31 are the moving distance L and the reciprocal 1 / T of the moving time.
e, determined by the weighting coefficient γ. So L and 1 / Te
And a relationship between L and γ are stored in the memory, and the coefficient is calculated by referring to the table before starting the movement.

The target speed is generated as described above. On the other hand, the speed of the controlled object is detected and fed back. The aforementioned target acceleration is added as acceleration feedforward to the result of compensating the velocity deviation. The result is input to the actuator as a manipulated variable to drive the controlled object.

[0077]

By providing a plurality of target acceleration tables and a plurality of target speed tables and adjusting the weighting factors to generate the target acceleration and the target speed, the acceleration pattern of the controlled object can be changed.

In particular, since the weighting coefficient values for suppressing a specific natural vibration are tabulated in correspondence with the moving distance, it is possible to set an acceleration that does not excite that vibration.

Further, since the waveforms of the target acceleration table and the target speed table are smooth, it is possible to suppress the high frequency vibration of the controlled object.

Further, the position of the controlled object is normalized by the moving distance, the normalized time corresponding to the normalized position is obtained, and the target acceleration / target speed table is referred to based on this value to determine the position of the controlled object. A target value can be generated based on

Further, the waveform of the target acceleration table is given by
By designing with Equation 24, the acceleration can be set to 0 at the time of 1/2 of the moving time at which the speed of the controlled object becomes maximum. As a result, it is possible to reduce the sinking of the head of the magnetic disk device.

Further, the velocity to be controlled is fed back to the target velocity, and the target acceleration is added as the acceleration feed forward to the result of compensating the velocity deviation to obtain the manipulated variable, thereby controlling the designed acceleration pattern. be able to.

[0083]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to head positioning control of a magnetic disk device will be described in detail below.

FIG. 1 shows a block diagram of a magnetic disk device head positioning control system to which the present invention is applied. First, the mechanism of residual vibration generation and the operation of the seek control system will be described.

The casing 32 is fixed on the base 31 via a viscoelastic body for insulating impact. The driving current supplied by the power amplifier 30 generates a driving force between the VCM (voice coil motor) coil 34 and the VCM magnet 33. As a result, the servo head 36 and the data head 37 mounted on the carriage 35 move to the target track at high speed and are positioned. However, since the reaction force of the driving force described above also acts on the housing side, a coupled vibration is generated between the housing 32, the VCM magnet 33, and the spindle 38. This system is in the form of
It can be modeled as a motion equation of coupled vibration with three degrees of freedom. From that equation, three natural frequencies of this system are obtained.
This value depends on the mass and viscoelastic coefficient of the member, but in many cases, the eigenmode near 500 Hz causes the disk 39 to vibrate, which impedes the reduction in seek time.

During the seek operation, the servo signal is detected from the surface of the disk 39 by the servo head 36 and transmitted to the position signal demodulation circuit 20. This circuit outputs a position signal proportional to the deviation from the center position of each track, and generates a track cross pulse each time the head crosses the boundary between adjacent tracks. The position signal is sampled by the AD converter 22 in the digital servo system. The difference counter 21
When this pulse is input, the number of remaining tracks is reduced by one.
If the head position information for each track and the position information within one track range represented by the position signal are combined,
It becomes continuous head position information from the target track to the current position. This signal is input to the head speed detecting means 23 to detect the head speed. A speed observer using a mathematical model of the controlled object is often used for the head speed detecting means 23. The deviation between the target speed and the detected head speed is processed by the compensation calculation means 24, the acceleration feedforward is added, and the DA converter 25 outputs the manipulated variable to the power amplifier 30.

Next, a method of generating a target trajectory according to the present invention will be described. The normalized position calculation means 1 normalizes the continuous head position information x by the seek distance L. The calculation is

[0088]

[Equation 32]

It is When a processor that requires a relatively long calculation time for the division is used, the division can be made unnecessary by storing a table showing the relationship between L and 1 / L in the memory. The first normalization time detecting means 2 is a table of the inverse function of Equation 22, and the second normalization time detecting means 4
Is a table of the inverse function of Equation 23. The coefficient adjusting means 6 sets the weighting coefficient 1-γ in the coefficient multiplying means 3 and the weighting coefficient γ in the coefficient multiplying means 5 at the start of the seek operation. By referring to the two tables based on the normalized position δ and performing the product-sum operation shown in Expression 24, the normalized time approximate value τ≈t / Te can be calculated accurately.

The first target speed generating means 7 is a table of the equation (29), and the second target speed generating means 9 is a table of the equation (30). By the coefficient adjusting means 6, the coefficient multiplying means 8

[0091]

[Expression 33]

Is set, and coefficient multiplication means 1
To 0

[0093]

[Equation 34]

Has been set. The target speed can be calculated by referring to the two tables based on the normalized time approximation value τ and performing the sum of products operation shown in Expression 31.

The first acceleration feedforward generating means 11 is a table of the equation 26, and the second acceleration feedforward generating means 13 is a table of the equation 27. The coefficient adjusting means 6 causes the coefficient multiplying means 12 at the start of the seek operation to

[0096]

[Equation 35]

Is set, and the coefficient multiplication means 14
In

[0098]

[Equation 36]

Has been set. The acceleration feed forward can be calculated by referring to the two tables based on the normalized time approximate value τ and performing the sum of products operation shown in Expression 28.

The coefficient adjusting means 6 has a table showing the relationship between the seek distance L and the reciprocal 1 / Te of the seek time, and a table showing the relationship between the seek distance L and the weighting coefficient γ. The relationship between L and Te determines the high-speed access performance of the disk device, and is predetermined at the design stage of the device. Further, the weighting coefficient γ can be determined as an optimum value from L and Te and the natural frequency of the residual vibration to be suppressed by the procedure described in “Means for solving the problem”. The seek distance L is input to the coefficient adjusting means 6 at the start of the seek operation, and two tables are referred to based on L in order to calculate each coefficient.

FIG. 2 shows a circuit configuration for realizing the present embodiment by the digital servo system.

The ROM 40 includes the tables 2 and 4 of the first and second normalized time detecting means, the tables 7 and 9 of the first and second target speed generating means, and the first and second accelerations. The tables 11 and 13 of the feedforward generating means, the table showing the relationship between the seek distance and the reciprocal of the seek time provided in the coefficient adjusting means 6, and the table showing the relationship between the seek distance and the weighting coefficient γ are stored. . The RAM 41 is used to store variable values during servo calculation. The data bus 42 is an electrical wiring for transmitting data among the arithmetic unit 44 of the digital servo system, the memories 40 and 41, and the interface unit connected to the external element. The data memory address bus 43 is connected from the arithmetic unit 44 to the memory 4
Electrical wiring for designating an address when 0, 41 is referred to. An ordinary microprocessor can be used for the arithmetic unit 44. However, in the present invention, as described above, many product-sum operations are performed, so that a digital signal processor (DSP) having a high-speed product-sum operation performance is Are suitable.

When the multiply-accumulate operation is repeatedly performed, the following procedure is performed. That is, first, the coefficient for multiplication is loaded from the data bus 42 to the multiplier register 44a.
Next, the multiplicand is loaded from the data bus 42, and the value in the multiplier register 44a is multiplied by the multiplier 44b. Since the switch 44c is connected to the output of the multiplier 44b, the value in the accumulator 44e and the result of the multiplier 44b are added by the adder 44d. The addition result is stored again in the accumulator 44e. The normalization time is calculated by the product-sum operation of Equation 22, and the result is stored in the RAM 41. This value is loaded into the pointer register 44f via the data bus 42, and the tables 7 and 9 of the target speed generating means and the tables 11 and 13 of the acceleration feedforward generating means in the ROM 41 are referred to. Using these table values, the product-sum calculation shown in Expressions 31 and 28 is performed to calculate the target velocity and the acceleration feedforward.

When the seek operation is completed, the arithmetic unit 44 notifies the upper controller 28 of the end of the seek operation via the digital output interface 27.

FIG. 3 shows a control algorithm of the digital servo system of this embodiment. When the host controller 28 sets the seek distance in the difference counter 21, this program is started. In step F1, initial processing for generating a target trajectory is performed. This will be described in detail later. In step F2, other initial processing for servo calculation is performed. In step F3, it is determined which of the seek operation and the following operation for following the center of the track to perform the servo calculation. When performing the seek operation, in step F4, the generation of a timer interrupt that starts the operation for the seek operation at a constant cycle is waited for. Details of the calculation for seek operation will be described later. When performing the following operation, in step F5, the generation of a timer interrupt for starting the calculation for the following operation at a constant cycle is awaited.

FIG. 4 shows details of the process of step F1 of FIG. In Step F10, the difference counter 2
The seek distance L is read from 1. In step F11, R
A table showing the relationship between the seek distance L and the reciprocal 1 / Te of the seek time stored in the OM 40 is referred to. In step F12, a table showing the relationship between the seek distance L and the weighting coefficient γ stored in the ROM 40 is referred to. Step F
In step 13, the coefficient 1-γ of the equation 24 is calculated, and the value is stored in the assigned address of the RAM 41. Step F14
Then, the coefficient of the equation 25 for calculating the acceleration feedforward

[0107]

[Equation 35]

And

[0109]

[Equation 36]

[0110] is calculated and the value is stored in the assigned address of the RAM 41. In step F15, the equation 33 and the equation 34, which are the coefficients of the equation 31 for calculating the target speed, are calculated, and the values are stored in the assigned addresses of the RAM 41.

FIG. 5 shows an algorithm of servo calculation during seek operation. This module executes once every time a constant period timer interrupt occurs during seek operation. In step F40, head position information is fetched. Specifically, the output of the difference counter 21 is sampled via the digital input interface 26, and the position signal is sampled via the AD converter.
After that, the position information for each track from the difference counter and the position information within one track of the position signal are combined to calculate continuous position information from the current position to the center of the target track. In step F41, the head speed is calculated based on the head position information. Step F42
Then, the processing for generating the target trajectory is performed. This will be described later in detail. In step F43, compensation calculation is performed based on the deviation between the target speed and the head detected speed. In step F44, processing for outputting the manipulated variable from the DA converter is performed. The processing up to this point is performed within one sampling period, and the process returns to step F3 in FIG.

6 and 7 show details of the target trajectory generation processing in step F3 of FIG. In step F420, the head position is divided by the seek distance L to calculate the normalized position δ. In step F421, the table 2 of the first normalized time detecting means is referred to based on δ. Step F422
Then, based on δ, Table 4 of the second normalized time detecting means
Refer to. In step F423, the normalized time approximate value τ is calculated by the equation 24. In step F424, τ
The table 7 of the first target speed generating means is referred to based on the above. In step F425, the table 9 of the second target speed generating means is referred to based on τ. In Step F426,
The target speed is calculated by the equation 31. Step F427
Then, the table 11 of the first acceleration feedforward generating means is referred to based on τ. In step F428, τ
The table 13 of the second acceleration feedforward generating means is referred to based on the above. In step F429, the acceleration feedforward is calculated by the equation 28.

FIG. 8 shows the inverse function curves of the equations 22 and 23 stored in the tables 2 and 4 of the first and second normalized time detecting means. Generally, the normalized position δ and the approximated normalized time τ ≒
The t / Te relationship is determined by the weighted sum of both curves.

FIG. 9 shows the number 2 stored in the tables 11 and 13 of the first and second acceleration feedforward generating means.
6, the curve of equation 27 is shown. The value of acceleration feedforward that generally occurs is the curve of the weighted sum of both curves. FIGS. 10, 11 and 12 show the VCM when the weighting factor γ is changed from the minimum value to the maximum value within the allowable range of Expression 17.
The drive current waveform is shown. The VCM drive current waveform becomes a waveform equal to the acceleration feed forward. When γ has a minimum value of 0, acceleration is started most gently and stopped gently.
As γ increases, changes in acceleration at the start and stop of the operation also increase.

FIG. 13 to FIG. 16 show the method according to the present invention.
The response waveform when performing a track seek is shown. In this example, the seek time is 3.7 ms, the target trajectory is designed with the frequency of the eigenmode to be suppressed being 460 Hz, and γ = 1.4.
Is. For comparison, FIGS. 17 to 20 show response waveforms of conventional 32-track seek by pseudo shortest time control. Comparing the position signals of FIG. 13 and FIG. 17,
The present invention significantly reduces the residual signal despite the shorter seek time. Furthermore, this effect is compared by vibration energy. FIG. 16 and FIG. 20 show the sum of the vibration energies of all the eigenmodes of the mechanism section and the vibration energies of only the modes having a natural frequency of 460 Hz. In either method, the 460 Hz mode occupies most of the vibration energy near the stop time. However, this method can significantly reduce the vibration not only after stopping but also during seek operation. 14 and 18 show VCM drive current waveforms, which are proportional to the head acceleration. Since the acceleration pattern according to the present invention is smoother than the conventional one, FIGS.
Comparing fluctuations in the head flying height of No. 9, high frequency vibration of about 3.5 kHz can be reduced.

FIGS. 21 to 24 show the method of the present invention for 64
The response waveform when performing a track seek is shown. In this example, the seek time is 4.9 ms, the target trajectory is designed with the frequency of the eigenmode to be suppressed being 460 Hz, and γ = 0. For comparison, FIGS. 24 to 28 show response waveforms of conventional 64-track seek by pseudo shortest time control. Even in the case of this seek distance, this method is excellent as before.

In the embodiment, the seek operation is performed by generating the target acceleration and the target velocity trajectory represented by the equations 18 and 19 based on the head position during movement. However, a similar vibration suppressing effect can be expected even if the trajectory is generated as a function of the operation time t regardless of the head position. Examples of this method will be described below.

In this method, since the normalization position and the normalization time are unnecessary, blocks 1, 2, 3, 4, and 5 related to this calculation can be excluded from the functional block diagram of FIG.

The circuit configuration when implemented by a digital servo system may be the same as in FIG.

Instead of referring to the table storing the trajectory pattern, the algorithm for generating the trajectory uses the digital servo operation unit 44 to calculate the time polynomial function number 18 and the number 19 of the trajectory.
Is calculated directly. At this time, the operation time t is measured by the clock signal of the microprocessor. Further, as in the embodiment, the ROM 40 stores a table showing the relationship between the seek distance L and the seek time Te (or its reciprocal) and a table showing the relationship between the seek distance and the weighting coefficient γ.

The head velocity is fed back to the calculated target velocity trajectory, and the target acceleration is added as an acceleration feed forward via the compensation calculation means 24, and the manipulated variable is output.

Also in this embodiment, the response waveforms are the same as those in FIGS. 13 to 16 and FIGS. 21 to 24, and residual vibration and head flying height variation can be suppressed.

Either the target acceleration or the target velocity trajectory is generated based on the head position as in the first embodiment,
A configuration in which the other is generated based on the operation time as in the second embodiment is also possible.

In this case also, the same vibration suppressing effect as that of the embodiment is obtained.

[0125]

As described above, the present invention provides a positioning control device including a controlled object that is moved to a target position and positioned, an actuator that drives the controlled object, and a structure in which the controlled object and the actuator are mounted. It is equipped with a table and multiple target speed tables. The values read from the table are weighted and summed to calculate the target value and acceleration feedforward to be input to the control system, and the weighting factor of the calculation is adjusted to control Target acceleration can be generated in various patterns. In particular, the vibration can be reduced by preliminarily calculating the weighting factor value corresponding to the moving distance and providing the table as a table so that the acceleration pattern suppresses a specific natural vibration of the mechanism unit. When applied to the head positioning control system of a magnetic disk device, residual vibration of the disk can be reduced and seek time can be shortened.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a circuit diagram of an embodiment of the present invention using a digital servo system.

FIG. 3 is an algorithm diagram of digital servo processing according to the embodiment of the present invention.

FIG. 4 is an algorithm diagram of digital servo processing according to the embodiment of the present invention.

FIG. 5 is an algorithm diagram of digital servo processing according to the embodiment of the present invention.

FIG. 6 is an algorithm diagram of digital servo processing according to the embodiment of the present invention.

FIG. 7 is an algorithm diagram of digital servo processing according to the embodiment of the present invention.

FIG. 8 is an explanatory diagram of a relationship between a normalized position of a trajectory designed by the present invention and a normalized time.

FIG. 9 is an explanatory diagram of two patterns of a target acceleration trajectory used in the present invention.

FIG. 10 is a waveform diagram of the VCM drive current at each weighting coefficient value.

FIG. 11 is a waveform diagram of the VCM drive current at each weighting coefficient value.

FIG. 12 is a waveform diagram of a VCM drive current at each weighting coefficient value.

FIG. 13 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 14 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 15 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 16 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 17 is a waveform diagram of a response in the conventional seek control.

FIG. 18 is a waveform diagram of a response in conventional seek control.

FIG. 19 is a waveform diagram of a response in the conventional seek control.

FIG. 20 is a waveform diagram of a response in conventional seek control.

FIG. 21 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 22 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 23 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 24 is a response waveform diagram when the present invention is applied to a seek control system of a magnetic disk.

FIG. 25 is a response waveform diagram in the conventional seek control.

FIG. 26 is a response waveform diagram in the conventional seek control.

FIG. 27 is a response waveform diagram in the conventional seek control.

FIG. 28 is a response waveform diagram in the conventional seek control.

FIG. 29 is an explanatory diagram of a relationship between head position and speed in conventional magnetic disk drive seek control.

FIG. 30 is a waveform diagram of acceleration in the conventional seek control of the magnetic disk device.

FIG. 31 is a velocity waveform chart in seek control of a conventional magnetic disk device.

[Explanation of symbols]

1 ... Normalized position calculating means, 2, 4 ... Normalized time detecting means, 3, 5, 8, 10, 12, 14, ... Coefficient multiplying means, 6
... Coefficient adjusting means, 7, 9 ... Target speed generating means, 11, 1
3 ... Acceleration feedforward generating means, 20 ... Position signal demodulation circuit, 21 ... Difference counter, 2
2 ... AD converter, 23 ... Head speed detection means, 24 ... Compensation calculation means, 25 ... DA converter, 30 ... Power amplifier,
31 ... Base, 32 ... Case, 33 ... VCM magnet, 34 ... VCM coil, 35 ... Carriage, 36 ... Servo head, 37 ... Data head, 38 ... Spindle,
39 ... Disk.

Claims (1)

[Claims] 1. A positioning control device comprising: a control target that moves to a target position and positions it; an actuator that moves the control target to the target position; and a structure that mounts the actuator and the control target. , One or more acceleration feedforward signal generating means,
One or more coefficient multiplying means for multiplying the acceleration feedforward signal output by each of the acceleration feedforward signal generating means by a weighting coefficient, an adding means for adding the outputs of the coefficient multiplying means, and one or more velocity targets Value generating means, and one or more coefficient multiplying means for multiplying the speed target value output by each of the speed target value generating means by a weighting coefficient,
An adding means for adding the outputs of the coefficient multiplying means, a weighting coefficient adjusting means for adjusting the weighting coefficient, a speed detecting means for detecting the speed of the controlled object, and a speed target value and a speed detection value obtained by summing the loads. A positioning control device comprising: a control means for inputting and performing an arithmetic operation; and an adding means for adding an output of the control means and an acceleration feedforward signal summed with a load.