JP2736715B2 - Positioning control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a positioning control system for positioning a controlled object such as a magnetic head at a high speed. In a magnetic disk device or an optical disk device, the head is moved by controlling an actuator from a current track (cylinder) position on a disk to a target track (target cylinder) position, and the head is moved to a commanded target track position. By the positioning, data writing or reading is performed. Also in a printer or an X / Y plotter, printing is performed at the target position by moving and positioning the head for printing from the current position to the target position.
In such positioning control, there is a demand for enabling high-speed positioning without generating vibration in an actuator or the like.

[0002]

2. Description of the Related Art In a magnetic disk drive, when a target track position is given, a head is moved to a target track position on a disk by an actuator such as a voice coil motor so that data can be written or read from the target track. Done. In this case, for example, a target speed curve as shown in FIG. 11 is given from a table or the like corresponding to the number of tracks from the current track position to the target track position. The target speed curve shows a deceleration characteristic for stopping the head from a certain speed to the target track position, and the actuator is controlled according to a difference between the actual speed of the head and the target speed curve. . Accordingly, since the difference is large at the start of the seek operation, the actuator is driven with the output having the full driving capability, and when the actual speed of the head matches the target speed curve, thereafter, the deceleration control according to the target speed curve is performed. . Such a control is generally performed by an analog operation. At the start of a seek operation, the actuator is driven with an output that saturates the power amplifier.
Since the change in the acceleration of the head is large, there is a possibility that vibration may occur.

A digital signal processor (D)
A method of performing positioning control by digital calculation using SP) is also known. For example, a control method that obtains a theoretically optimal movement trajectory that accelerates and decelerates a mechanism section equipped with a head with low impact by digital calculation and performs feedback and feedforward control so as to follow the movement trajectory. Proposed earlier (Japanese Patent Application No. 2-28197)
No.). In this control method, as shown in FIG. 12, a movement trajectory of an acceleration a, a velocity v, and a position x expressed by using a low-order algebraic function with respect to time is set as a target trajectory, and an actuator is controlled so as to follow the trajectory. Things. In addition,
Horizontal axis represents time t _n normalized, and the ordinate shows the normalized value. The equation of motion of such a system that moves the mass by a certain distance L is as shown in the following equation.

[0004]

(Equation 1)

Where x = moving position, v = moving speed, a =
Moving acceleration, u = driving amount, M = mass, Kf = force constant, T
= Target travel time. Note that Kf is a force constant called BL [N / A] of the voice coil motor in the magnetic disk device. Also, t = 0 and t = T in the equation (0)
Under the boundary conditions of (1), (0a) and (0b) are obtained.

In order to design a target trajectory, a target position x _obj , a target speed v _obj , and a target acceleration a _obj are set so as to satisfy the equation (0). The optimum trajectory in the previously proposed invention is the evaluation function J = ∫ (0
T) (da / dt) ² dt (where ∫ (0 to T) is t
= 0 to t = T), and the function representing the trajectory is represented by the following equation. ^{_{x obj / L = t n 3}} (10-15t n + 6t n 2) ... (1) v obj / (L / T) = 30t n 2 (1-t n 2) ... (2) a obj / (L / T ² ) = 120t _n (1−t _n ) (0.5−t _n ) (3) where t is the elapsed time, T is the movement target time, L is the movement distance, and t _n = t / T = normal. Indicates the activation time. FIG. 12 described above.
Is (1) to (3) target position by an equation, the target speed, the target acceleration, which shows the curve x0, v0, a0 each normalized, the normalized abscissa with the t _n = t / T Indicates time.

[0007]

The moving trajectory represented by the equations (1) to (3) for minimizing the above-mentioned evaluation function J minimizes the acceleration / deceleration change during the moving time. Therefore, stable movement can be performed. However, settling after the end of the movement is not considered. That is, the target acceleration during the movement is given by equation (3), and after the movement, the target speed and the target acceleration become zero.
(da _obj / dt) at t = T becomes infinite and t
= T, the change in the target acceleration is not continuous. Therefore, if the integration range of the evaluation function J is expanded not to be 0 to T but to include the settling time after the movement, (1) to
The trajectory according to equation (3) is not optimal. In the case where the settling time after the movement greatly affects the access performance as in the case of a disk device, at t = T (da _obj / d
t) needs to be continuous.

When the actual position, actual speed and actual acceleration are controlled so as to follow the target trajectory according to the above-mentioned equations (1) to (3), the actual measured value always follows the target trajectory with a delay. Is common. Also, the actual trajectory may fluctuate near the target trajectory due to disturbances, fluctuations in control parameters, and the like. For example, if the force constant (the proportional constant between the force generated by the current supplied to the voice coil motor and the current) in the disk device is larger than the design value due to product variation, etc., the actual position trajectory is larger than the target position trajectory. Previously, a case may occur where the target position is reached before the speed becomes zero. For example, with respect to the target position x _obj , the target speed v _obj , and the target acceleration a _obj of the solid line in FIG. 13, the actual position x _r , the actual speed v _r , and the actual acceleration a _r are indicated by dotted lines, respectively. When the actual position _xr becomes zero (T−ε), the target speed v _obj and the target acceleration a _obj are not zero, so feedback control is performed to make them zero. _r and the actual acceleration _ar increase, the actual position x _r , the actual speed v _r , and the actual acceleration a _r change in a damped manner as indicated by the dotted lines, and are positioned at the target position x _obj . There is a drawback that the settling time for performing is long.

In the positioning control system using the trajectory according to the above-mentioned equations (1) to (3), a large amount of feedforward is used, and parameters of a controlled body such as a head and an actuator vary. Sometimes, the difference between the target trajectory and the actual trajectory becomes large, and the phenomenon shown in FIG. 13 frequently occurs. That is, the robustness is poor, which is a problem in mass production. SUMMARY OF THE INVENTION It is an object of the present invention to perform stable and high-speed positioning of a controlled object such as a head.

[0010]

A positioning control system according to the present invention will be described with reference to FIG. 1. Referring to FIG. 1, a drive unit 2 for moving a controlled body 1 and positioning it at a command position, and a position of the controlled body 1 And a driving unit 2 based on the detected position of the controlled object 1 and the commanded position by the position detecting unit 3.
And an arithmetic control unit 4 for controlling the
A trajectory generator 5 that calculates a target trajectory based on the total travel distance between the current position of the controlled object 1 and the commanded position, and a target trajectory from the trajectory generator 5 and a position detector 3 A tracking control unit 6 for controlling the driving unit 2 during acceleration and deceleration so as to follow the movement target trajectory based on the detected position, and the trajectory generation unit 5 issues a command to complete the movement of the controlled body 1 A moving target trajectory is obtained by an algebraic polynomial indicating an acceleration / deceleration pattern at which the differential value of the target acceleration at the position becomes zero,
Following the movement target trajectory, the following control unit 6
Is tracked and positioned.

The target movement trajectory is defined as T, the movement target time is T, the moving time with the initial value being 0 is t, and the constant is V ₀ ,
It is represented by a polynomial of V ₀ (t / T) ^k [(1- (t / T) ^j ], where the power j is set to 3 or more and the power k is set to 2 or more.

Further, the position of the controlled object 1 is detected by the position detecting unit 3, and the position of the controlled object 1 is estimated from the detected position information for each sample and the driving amount of the driving unit 2 for moving the controlled object 1. The speed is obtained, and the follow-up control unit 6 controls the drive unit 2 so that the difference between the estimated speed and the target speed based on the moving target trajectory becomes zero.

The moving time between the current position of the controlled body 1 and the command position is set in proportion to the root of the moving distance.

Further, the position of the controlled object 1 is detected by the position detecting unit 3, and the position of the controlled object 1 is estimated from the detected position information for each sample and the driving amount of the driving unit 2 for moving the controlled object 1. The speed is obtained, and a speed feedback gain corresponding to the estimated speed is set.

Further, a normalized time is calculated from position information detected by the position detecting section 3, and a response delay is added to the normalized time to compensate for the response delay.

When the moving distance of the controlled body 1 is equal to or more than a predetermined value, the moving distance between the acceleration period and the deceleration period based on the moving target trajectory is determined.
There is a constant speed period.

[0017]

The driving section 2 is constituted by a voice coil motor or the like, and moves the controlled body 1 such as a head to position it at a command position. The arithmetic control unit 4 includes a trajectory generation unit 5 and a follow-up control unit 6. The trajectory generation unit 5 determines the total travel distance between the commanded position and the current position of the controlled object 1 by the position detection unit 3. To calculate the moving target trajectory. This movement target trajectory is represented by an algebraic polynomial in which the differential value of the target acceleration becomes zero at the command position where the movement of the controlled body 1 is completed. The tracking control unit 6 controls the control body 1 to follow the movement target trajectory, and positions the controlled body 1 at the command position.

The movement target trajectory is obtained by the trajectory generation unit 5 of the arithmetic control unit 4. The movement target time for moving the controlled object 1 from the current position to the command position is T, the time during movement is t, and the constant is V. _{Assuming 0} , V ₀ (t / T) ^k [1- (t /
T)] A moving target trajectory is represented by a polynomial of ^j , and its power j
Is set to 3 or more, the differential value of the target acceleration when the controlled body 1 is moved to the command position becomes zero. And k = j
Then, the pattern of the acceleration period and the deceleration period of the target acceleration becomes similar.

The tracking control unit 6 obtains the estimated speed of the controlled object 1 from the value of the position detection information by the position detection unit 3 for each sample and the drive amount of the drive unit 2, and calculates the estimated speed and the movement. The drive amount of the drive unit 2 is controlled so that the difference between the target trajectory and the target speed becomes zero, and the controlled body 1 follows the target speed.

The trajectory generating section 5 increases the target movement time T between the current position of the controlled object 1 and the commanded position in accordance with the movement distance. Is not made to correspond to 1: 1 and the movement target time T
Is set in proportion to the second to third powers of the moving distance.

The position detecting unit 3 detects the position of the controlled object 1 and detects the detected position information for each sample for digital operation and the driving amount of the driving unit 2 for moving the controlled object 1. , The moving speed of the controlled object 1 is estimated. The speed feedback gain corresponding to the estimated speed is set.

The controlled body 1 is moved by the drive unit 2, but a delay occurs with respect to the movement target trajectory according to the delay of each unit. Therefore, the response delay can be compensated for by adding this response delay time to the normalized time.

When the moving distance of the controlled object 1 is large,
Since a case where the upper limit of the limit speed is exceeded occurs, a constant speed period can be provided between the acceleration period and the deceleration period, and when the target speed reaches the upper limit, the operation can be shifted to the constant speed period.

[0024]

FIG. 2 is a block diagram showing an embodiment of the present invention.
This shows a case where the present invention is applied to a magnetic disk device. In the figure, 11 is a magnetic disk device, 12 is a magnetic disk, 1
3 is a spindle motor, 14 is a head, 15 is a head arm, 16 is an actuator composed of a voice coil motor, 17 is a servo signal decoder, 18 is a cylinder counter, 19 is an AD converter (ADC), and 20 is an amplifier ( AM
P), 21 is a DA converter (DAC), 22 is an operation control unit, and 23 is a digital signal processor (DS).
P) and 24 are memories (MEM), 25 is a transmission / reception register (SR) for transmitting / receiving data to / from a higher-level device,
Reference numeral 26 denotes a bus, and the illustration of a control structure for reading and writing data by the head 14 is omitted.

The magnetic disk drive 11 comprises a magnetic disk 12 rotated at a constant speed by a spindle motor 13.
And an actuator 16 that controls the head 14 to move a head arm 15 supported at the tip to position the head 14 at a command track position of the magnetic disk 12. Can be used. The head 14 corresponds to the controlled body 1 in FIG. 1, and the actuator 16 corresponds to the drive unit 2 in FIG.

The servo signal decoder 17 decodes a servo signal read from the magnetic disk 12 by the head 14, and the cylinder counter 18 counts the number of cylinders (eg, zero-cross points) of the decoded servo signal. Number of tracks)
The number of cylinders (the number of tracks) indicates the current position of the head 14 from the reference point. Therefore, the moving distance (seek distance) of the head 14 corresponds to the number of tracks corresponding to the difference between the current track position and the command track position. The AD converter 19 converts the decoded two-phase servo signal into a digital signal. The current position of the head 14 from the reference point is detected by a combination of the cylinder counter 18 and the AD converter 19, and these correspond to the position detector 3 in FIG. The DA converter 21 converts the drive control signal from the digital signal processor 23 via the bus 26 into an analog signal, and amplifies the signal by the amplifier 20 to drive the actuator 16.

The arithmetic control unit 22 includes a digital signal processor 23, a memory 24, and a transmission / reception register 25. The digital signal processor 23 controls the head 14 based on an address signal from a host device or the like. The moving distance from the current position of the head 1 to the command position is obtained, and the moving target trajectory corresponding to the moving distance is obtained.
4 performs an operation for controlling the actuator 16 so as to follow the movement target trajectory. That is, the digital signal processor 23 of the arithmetic control unit 22 corresponds to the arithmetic control unit 4 of FIG. 1, and the trajectory generating unit 6 and the tracking control unit 7 of FIG. It is realized by the control function.

The present invention relates to (a). At t = T, that is, the differential value da _obj / dt of the target acceleration at the time when the movement of the head 14 as the controlled body 1 is completed is continuous;
(B). At t = T−ε (ε = control error time), the differential value da _obj / dt of the target acceleration is as small as possible; (c). It is assumed that the target trajectory is represented by as simple a function as possible so that the trajectory can be calculated digitally at high speed. The velocity trajectory satisfying the conditions (a) to (c) is as follows: It is expressed by an equation. v _obj / (L /
^{_{T) = V 0 t n k}} (1-t n) j ... (4) where, V ₀ is a constant, k is an integer of 2 or more, j is an integer of 3 or more.

The above equation (4) shows that t _n = 1, that is, t =
Since it becomes a triple root or more at T, da _obj / dt =
It can be seen that (d / dt) ² · v _obj becomes zero at t = T. Note that the constant V _{0 in the} equation (4) is L = x _obj (T) = ∫ (0−T) v _obj (t) dt of the boundary condition.
(Where ∫ (0 to T) indicates the integration from t = 0 to t = T) and x _obj (0) = 0, and V ₀ =
(K + j + 1)! /(K!.j!). However,! Is a factorial symbol.

[0030] For example, in a power multiplier k in the above equation (4), for j, When k = 2, j = 3, x obj / L = t n 3 (20-45t n + 36t n 2 -10t n _{^{3) ... (5) v obj}} / (L / T) = 60t n 2 (1-t n) 3 ... (6) a obj / (L / T 2) = 300t n (1-t n) 2 (0 .4-t _n ) (7). Further When k = 3, j = 3, x obj / L = t n 4 (35-84t n + 70t n 2 -20t n 3) ... (8) v obj / (L / T) = 140t n 3 (1 −t _n ) ³ (9) a _obj / (L / T ² ) = 840t _n ² (1−t _n ) ² (0.5−t _n ) (10)

FIG. 3 is an explanatory view of a target trajectory according to the embodiment of the present invention. The horizontal axis represents normalized time t _n = t / T, and the vertical axis represents normalized position, normalized speed, and normalized acceleration. The target position x _obj and the target speed v according to the above equations (5) to (7)
_obj and target acceleration a _obj are shown by curves x1, v1, and a1. The target acceleration a _obj is _calculated as the normalized time t _n = t / T
= 0.4, it becomes zero, the period before that is the period of the acceleration mode, and the period after that is the period of the deceleration mode. Also, the normalized time t _n =
Near t / T = 1, the value is close to zero, and the normalized time t
_{At n} = t / T = 1, the differential value of the target acceleration a becomes zero.

FIG. 4 is an explanatory diagram of the target trajectory according to the embodiment of the present invention. The target position x _obj , the target speed v _obj , and the target acceleration a _obj in the case of the aforementioned equations (8) to (10) are shown. 3
The curve x2, v2, and a2 are shown in the same manner as in the case of. The target acceleration a _{obj in} this case is represented by a normalized time t _n = t / T =
It becomes zero at 0.5, and the acceleration mode and the deceleration mode have symmetry. When k = 3 and j = 3, the target acceleration at the time of acceleration becomes smaller as compared with the case where k = 2 and j = 3. Therefore, the acceleration current becomes smaller under a given power supply voltage. There are advantages that can be.

FIG. 5 is a functional block diagram of the arithmetic and control unit according to the embodiment of the present invention, where L is the moving distance of the head 14, Lm is the maximum moving distance that does not require constant speed control, that is, the distance longer than Lm. When moving, the constant speed mode is inserted during the acceleration / deceleration period. 31 is a trajectory generator, 32 is a tracking controller,
33 is a drive time calculator, 34 is a track following gain calculator, 3
5 is a drive state classifier, 36 is a normalization calculator, 37 is a target trajectory calculator, 38 is a state estimator, 39 is a state controller,
0 is a drive circuit, 41 is an actuator, and 42 is a detector. The actuator 41 corresponds to the actuator 16 in FIG. 2, and the detector 42 detects the position of the head arm (see 15 in FIG. 2) by the actuator 41, that is, the current position of the head. The function of each unit can be realized by the arithmetic control function of the digital signal processor.

FIGS. 6 and 7 are flowcharts of the arithmetic control unit according to the embodiment of the present invention. The operation of each unit will be described with reference to FIG. 5 in steps (a) to (q). The trajectory generating unit 31 receives the moving distance L of the difference between the current position and the commanded position, the predetermined constant speed mode start distance Lm, and the position detected by the detector 42. Step (a): The drive time calculation unit 33 compensates for the acceleration / deceleration distance Ldrv, the movement target time T, the normalized sampling time Tu, and the control delay based on the movement distance L and the constant speed mode start distance Lm. And an advance compensation time Tlead to be calculated. Ldrv = min (L, Lm) (11) T = Tm (L / Lm) ⁱ (12) Tu = Ts / T (13) Tlead = const. … (14)

The acceleration / deceleration distance Ldrv in the above equation (11)
Indicates that the smaller one of the moving distance L and the constant speed mode starting distance Lm is selected, and the moving target time T in the equation (12) is a value proportional to the i-th power of L / Lm. And the power i can be a value of 1/2 to 1/3, for example, 1 / 2.8. Further, the normalized sampling time Tu in the equation (13) is represented by a ratio between the sampling period Ts and the movement target time T. The lead compensation time Tlead in the equation (14) is used to compensate for the control delay time of each unit to reduce the error with respect to the target trajectory, and shows a case where it is set to a constant value (const.). = 2Tu.
Also, it can be changed as needed. The sampling counter N is initialized and its value is set to zero.

Step (b): track following gain calculating section 34
Is the normalized sampling time Tu and the acceleration / deceleration distance Ldrv
Is input, and the feedforward gain Gff and the maximum orbital velocity Vmax are obtained as follows. Gff = 840 (Ldrv / T ² ) (15) Vmax = 140 (Ldrv / T) (16)

Step (c): The driving state classification unit 35
The acceleration / deceleration distance Ldrv, the movement distance L, and the detection position Xn from the detector 42 are input, and the processing in step (c) of the flowchart in FIG. 6 is performed. (D) to (q) based on the detection position Xn obtained for each sampling period
Is repeated. First, it is determined about L ≦ Lm (d). If L ≦ Lm, that is, if the moving distance L is shorter than the constant speed mode start distance Lm and does not include the constant speed mode, Xn <(Ldrv / 2) Is determined (g), and X
When n <(Ldrv / 2), that is, when the detection position Xn is less than half the acceleration / deceleration distance Ldrv, the acceleration state stat
It is determined that e = 1 (i). Target acceleration in this case, as shown in FIG. 4 shows a case where a zero at the half of the normalized time t _n, therefore, the detection position Xn is deceleration distance L
This is to determine whether or not the value exceeds 1/2 of drv.

In step (g), Xn <(Ld
If not (rv / 2), it is determined that the deceleration state state = -1 (j). In step (d), L ≦ Lm
In other words, the moving distance L is the constant speed mode starting distance L
When the constant speed mode is included between the acceleration mode and the deceleration mode, it is determined whether Xn <(Ldrv / 2) (e). If Xn <(Ldrv / 2), the acceleration is performed. It is determined that the state state = 1 (i). In step (e), if Xn <(Ldrv / 2) is not satisfied, it is determined whether (L−Xn) <(Ldrv / 2) or not (f), and (L−Xn) <(Ldrv / In the case of 2), the deceleration state state is set to −1 (j), (L−Xn) <
If not (Ldrv / 2), the constant speed state state =
It is determined to be 0 (h). The remaining moving distance is given by L-Xn.

Step (k): The normalization calculation section 36 performs a process of obtaining the normalization time Tn in step (k) of FIG. 7, and includes the normalized sampling time Tu, the control state state, and the remaining movement distance (L -Xn) and the advance compensation time Tlead from the drive time calculation unit 33 are input,
When the acceleration state state = 1, Tn = N · Tu + Tl
e (m). N in this case indicates the N-th sampling from the start of the trajectory calculation. When the constant speed state stste = 0, Tn = 0.5 (l), and when the deceleration state state = −1, Tn = NTable [L
−Xn] + Thread. In addition, NTable [L
-Xn] is y = x ⁴ (3
5-84x + 70x ² -20x ³⁾ expression of ((8) and the same) is a value of the table solving for x, it is possible to obtain the normalized time from the measured position by the table. For example, the value of x for y = 0.5 to 1.0 can be stored.

Step (o): The target trajectory calculation unit 37
The processing in step (o) of FIG. 7 is performed, and the feedforward gain G
ff, the maximum orbital velocity Vmax, and the normalized time Tn from the normalization calculation unit 36 are input, and TrjUn = Gff · Tn ² (1−Tn ² ) (0.5−Tn) (17) TrjVn = Vmax The feedforward current TrjUn and the target speed TrjVn are calculated from Tn ³ (1−Tn ³ ) (18) and input to the state controller 39 of the following control unit 32.

Step (p): The state estimator 38 of the tracking control unit 32 performs the process of step (p) in FIG. 7, and detects the detection position Xn from the detector 42 and the drive from the state controller 39. current Un and is _{input, EstXn = (1-L 1} ) EstXn-1 + EstVn-1 + GmUn-1 + L 1 · Xn ... (19) EstVn = -L 2 · EstXn-1 + EstVn-1 + 2Gm · Un-1 + L 2 · Xn (20) Gfb = GTable [EstVn] (21) An estimated speed EstVn is obtained. Here, the constant Gm
Is Gm = (Kf · Ts ² · Gdrv · Gsns) /
(2M · Ltp). Here, Gdrv is the gain of the drive circuit 40, Gsns is the gain of the detector 42, and Ltp is the track pitch. L ₁ and L ₂ are constants that determine the characteristics of the state estimator (observer), and are determined by using a state estimator design method. This design method is described in detail, for example, in "Digital Control of Dynamic Systems" (authored by GF Franklin and JD Powell). Equation (21) is equivalent to the estimated speed EstVn.
Indicates that the feedback gain Gfb is read from the table GTable [EstVn] that stores the feedback gain Gfb.

Step (q): The state controller 39 checks the state shown in FIG.
Of the feedforward current TrjUn and the target speed TrjV.
n, the estimated speed EstVn, and the feedback gain Gfb are input, and the driving current Un is calculated from Un = TrjUn + Gfb (TrjVn-EstVn) (22). By applying this drive current Un to the actuator 41 via the drive circuit 40, it is possible to follow the target movement trajectory and to perform stable positioning control. Further, the sampling counter N is incremented.

FIG. 8 is a graph in which the curves of FIGS. 4 and 12 are superimposed, and the curves of the target position x0, the target speed v0, and the target acceleration a0 in the previously proposed positioning control method. The target position according to the embodiment of the present invention is indicated by a curve x2, the target speed is indicated by v2, and the target acceleration is indicated by a2. position,
Looking at the respective normalized target trajectories of the speed and the acceleration, the target acceleration of the embodiment of the present invention at t = (0.8 to 1.0) T is previously proposed as shown by hatching. It can be understood that the target acceleration becomes smaller than the target acceleration in the determined positioning control method, and that _daobj / dt = 0 at t = T. Therefore, by performing positioning control using such a target trajectory, fluctuations and disturbances in system parameters become strong, and a transient vibration as shown in FIG.

In the embodiment of the present invention, the target speed is t =
At (0.7-1.0) T, the target speed of the previously proposed positioning control method is lower than that of the target speed indicated by hatching, so that the feedback gain can be increased earlier. Therefore, the tolerance of the tracking error can be increased, the tracking error at t = T can be reduced, and the positioning to the command position can be speeded up and stabilized. During the movement, the target speed TrjVn is controlled to follow the target speed TrjVn as shown in Expression (22). Therefore, near the end of deceleration where both the target speed TrjVn and the estimated speed EstVn have become smaller, the error (Trj) is obtained.
Vn-EstVn) also becomes a small value. Therefore,
It is effective to increase the feedback gain in order to increase the relative following degree between the estimated speed EstVn and the target speed TrjVn, and to suppress the control error at the end of deceleration. For this reason, the estimated speed E is calculated using the table shown in Expression (21).
The feedback gain is increased as stVn decreases. This table GTable is optimized experimentally, but is usually designed to have a gain of two to three times at the end of deceleration as compared with that at the time of high-speed movement.

The gain of the target trajectory is determined, for example, by the value of the target movement time T in the above-mentioned equations (8) to (10). Then, the value according to the equations (9) and (10) is
This corresponds to a case where the curve shown in FIG. 3 or FIG. On the other hand, the movement target time T is limited by the ability of the amplifier 20 for supplying the drive current to the actuator 16 and the rate of change of the current is limited by the inductance of the actuator driving coil and the like. Also has certain limitations. In general, when performing orbital calculations, it is preferable to normalize the term of the algebraic function (not to exceed ± 1.0) when using a digital signal processor that performs a fixed-point operation. The normalization function will be multiplied by the gain.

[0046] For example, (9), (10) a, v _obj = Vmax ^{_{[64t n 3 (1-t n}} 3) ] ... (23) a _obj = Amax [50 × 5 ^1/2 × t _n ² (1−t _n ) ² (0.5−t _n )] (24) where Vmax = (35/16) (L / T) (25) Amax = (84/125 ^1/2 ) (L / T ² ) (26) Therefore, the algebraic expressions in the brackets [] of the expressions (23) and (24) are normalized, and Vmax, Amax
Is the target trajectory gain.

As can be seen from the above equations (23) and (24), the acceleration increases inversely as the moving target time T decreases. However, actually, as described above, since there is a maximum current value Imax that can be supplied from the drive circuit 40 to the actuator 41, Amax is (Kf ·
The movement target time T is selected so as not to exceed (Imax) / M. Therefore, Amax = (Kf · Ima
When x) / M, T = [(84M · L) / (125 1/2 · Kf · Imax) ] by ^1/2 (27) can determine the moving target time T. That is, the movement target time T is proportional to the square root of the movement distance L.

The factor that limits the gain is the inductance Le of the actuator. If the supply voltage in this case is E, di / dt <E / L (28) is selected. In that case, i = (M
/ Kf) because it is a, (from _{10), da obj / dt = (105/2} ) (L / T 3) × _{[16t n (1-t n)} (- 1 + 5t n -5t n 2 ] ... As shown in (29), it can be decomposed into a normalization equation and a gain, and the target movement time T is obtained from (105/2) (L / T ³ ) (M / Kf) <(E / Le) (30) In this case,
T is determined in proportion to the cube root of the moving distance L.

When the moving distance L is changed in a wide range,
The trajectory gain is: (α) When the moving distance L is small, the trajectory gain is limited by the electrical inductance of the actuator, so the moving target time T is determined so as to be proportional to L ^1/3 . (Β) When the moving distance L is large, the current is limited by the maximum value of the current flowing through the actuator, and therefore, it is determined to be proportional to L ^1/2 . (Γ) Generally, the gain is (1/3 to 1) of the moving distance L.
/ 2) so that it is raised to the power.

FIG. 9 shows a logarithmic scale with the horizontal axis and the vertical axis,
It indicates the relationship between the movement distance L and the movement target time T.
The area above the straight lines (i) and (ii) of the / 2 power and the 1/3 power is the allowable selection area for the movement target time T. Then, the solid straight lines (i) and (ii) are compared with the dotted straight line (i
By using ii), the movement target time T can be easily obtained from the movement distance L. In that case, the dotted straight line (ii
i) can be, for example, T = L ^{1 / 2.8} .

There are various factors that limit the maximum speed Vmax. For example, the speed at which the position detection accuracy by the detector 42 (see FIG. 5) can be maintained, the drive capability of the actuator 41 by the drive circuit 40, and the actuator 41 are constituted. There is a back electromotive force of a voice coil motor, and the like. Therefore, it is determined whether the moving distance L has exceeded a predetermined distance Lm. That is, it is determined whether or not L ≦ Lm as in step (d) in the flowchart of FIG.
Is larger than the distance Lm, a constant speed period is provided between the acceleration period and the deceleration period as shown in FIG. 10, and in this constant speed period, Vmax = v _obj , a
_obj = 0. For a magnetic disk drive,
The distance about 1/3 of the entire movement range on the magnetic disk is Lm
In general, L> L as shown in FIG.
m, since the constant speed mode is used, T = Lm + (L
−Lm) / Vmax, which is proportional to L to the first power.

In the above-described embodiment, the case where the speed of the actuator 41 is estimated by the state estimator 39 is described. However, a speed detector capable of detecting the speed of the actuator 41 may be used. In addition, Trj is directly calculated from the remaining movement distance (L-Xn) without calculating the normalized time.
A table for reading Vn is provided, and a table NTable is provided.
[L-Xn] and the like can be omitted. Further, in order to emphasize the effect of feedback, Tn in step (m) is set as Tn in equation (17) for calculating TrjUn.
It is effective to substitute N · Tu + Tread.

[0053]

As described above, according to the present invention, the trajectory generating unit 5 of the arithmetic control unit 4 determines the target trajectory based on the moving distance of the controlled object 1, and the controlled control unit 6 controls the trajectory. The driving unit 2 is controlled so that the body 1 follows the movement target trajectory, and the controlled body 1 is positioned at the command position.
In this case, the moving target trajectory is determined such that the differential value of the target acceleration when the controlled object 1 reaches the command position becomes zero. Thereby, there is an advantage that when the controlled body 1 is moved to the commanded position, there is no sudden change in acceleration, so that the positioning can be performed stably. In addition, the parameter of each part including the controlled body 1 and the driving unit 2 can be changed. Also, even when the controlled object 1 reaches the command position before the movement target time, the change in the target speed and the target acceleration is small, so that there is an advantage that the positioning can be performed stably and at a high speed.

The moving target trajectory is represented by V ₀ (t / T) ^k [1
- represents a (t / T)] ^j, by making the power of multiplier j 3 above, as shown in target acceleration a1, a2 in FIG. 3 or FIG. 4, when the controlled device 1 is moved to the command position Of the target acceleration can be set to zero.

Further, by estimating the moving speed of the controlled object 1 using the driving amount of the driving unit 2 and the sample value of the detected position, the difference between the estimated speed and the target speed is calculated. Is controlled to be zero so that the moving speed of the controlled body 1 cannot be directly detected.
The controlled object 1 can be moved following the target speed. If the moving speed of the controlled body 1 can be directly detected, the control may be performed so that the difference between the moving speed and the target speed becomes zero.

When the target movement time T is set in accordance with the movement distance L of the controlled object 1, the movement target time T is set in proportion to the root of the second or third power of the movement distance L. In addition, the controlled body 1 can be positioned at the command position at high speed. Also, by setting the speed feedback gain in accordance with the moving speed of the controlled body 1, the moving speed decreases as the controlled body 1 approaches the command position. In this case, the gain is increased and the tracking error is reduced. It is easy to approach zero. Further, by proceeding to the normalization time and adding the compensation time to compensate for the response delay of each unit, it becomes easy to make the tracking error close to zero. When the moving distance L of the controlled body 1 exceeds a predetermined value, for example, a constant speed mode start distance Lm,
By providing a constant speed period between the acceleration period and the deceleration period and moving the controlled body 1 during the constant speed period at the maximum moving speed, the driving unit 2 is not overloaded. There is an advantage that positioning can be performed at high speed.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a block diagram of an embodiment of the present invention.

FIG. 3 is an explanatory diagram of a target trajectory according to the embodiment of the present invention.

FIG. 4 is an explanatory diagram of a target trajectory according to the embodiment of the present invention.

FIG. 5 is a functional block diagram of an arithmetic control unit according to the embodiment of the present invention.

FIG. 6 is a flowchart of an embodiment of the present invention.

FIG. 7 is a flowchart of an embodiment of the present invention.

FIG. 8 is a specific comparative explanatory diagram of the positioning control proposed earlier and the positioning control of the embodiment of the present invention.

FIG. 9 is an explanatory diagram of a moving distance and a moving target time according to the embodiment of the present invention.

FIG. 10 is an explanatory diagram of a target speed and a target acceleration including a constant speed period according to the embodiment of the present invention.

FIG. 11 is an explanatory diagram of control characteristics of a conventional example.

FIG. 12 is an explanatory diagram of a previously proposed target trajectory.

FIG. 13 is an explanatory diagram of a target trajectory and an actual trajectory.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Controlled object 2 Drive part 3 Position detection part 4 Operation control part 5 Trajectory generation part 6 Follow-up control part

Claims (7)

(57) [Claims] 1. A drive unit (2) for moving a controlled object (1) to a commanded position, a position detecting unit (3) for detecting a position of the controlled object (1), and detecting the position. Department (3)
A control unit (4) for controlling the driving unit (2) based on the detected position and the command position of the controlled body (1), wherein the calculation control unit (4) A trajectory generator (5) for calculating a target trajectory based on the total travel distance between the current position of the body (1) and the commanded position, and the target trajectory from the trajectory generator (5); Based on the position detected by the position detection unit (3), the drive unit (2) is added so that the controlled object (1) moves following the movement target trajectory.
A tracking control unit (6) for controlling at the time of speed and deceleration , wherein the trajectory generating unit (5) calculates a differential value of a target acceleration at the command position at which the movement of the controlled object (1) is completed. The moving target trajectory is obtained by an algebraic polynomial indicating an acceleration / deceleration pattern that becomes zero, and the tracking control unit (6) performs tracking control and positions the controlled object (1) according to the moving target trajectory. Positioning control method. The method according to claim 2, wherein the moving target trajectory, a movement target time T, the time of moving with an initial value 0 t, constant as _{V 0, V 0 (t /} T) k [1-(t / T)] expressed by a polynomial of ^j, the power j is 3 or more, and the power k
2. The positioning control method according to claim 1, wherein the number is set to two or more. 3. The position detector (3) detects the position of the controlled object (1), and detects the detected position information for each sample and the driving unit (2) that moves the controlled object (1). ), The estimated speed of the controlled object (1) is obtained from the driving amount, and the following control is performed by the following control unit (6) so that the difference between the estimated speed and the target speed based on the moving target trajectory becomes zero. 2. The positioning control method according to claim 1, wherein the control section controls the section. 4. The moving time between the current position of the controlled object (1) and the commanded position is set in proportion to the second to third root of the moving distance. Positioning control method. 5. The position detector (3) detects the position of the controlled object (1), and detects the detected position information for each sample and the driving unit (2) that moves the controlled object (1). 2. The positioning control method according to claim 1, wherein an estimated speed of the controlled object is obtained from the drive amount of (1), and a speed feedback gain corresponding to the estimated speed is set. 6. The apparatus according to claim 1, wherein a normalized time is calculated from position information detected by said position detecting section, and a response delay is added to the normalized time to compensate for the response delay.
The positioning control method described. 7. A constant-speed period is provided between an acceleration period and a deceleration period of the target trajectory when the moving distance of the controlled object (1) is equal to or more than a predetermined value. 1. The positioning control method according to 1.