JPH0691082A - Feed quantity regulating sewing machine - Google Patents

Abstract

(57) [Summary] [Purpose] With a feed amount adjusting sewing machine, the feed amount can be controlled more simply and appropriately. [Structure] A detector detects a handle of each of two work cloths fed by an upper feed tooth and a lower feed tooth, and calculates a deviation amount between the two work cloths detected thereby, and also calculates a deviation amount change amount. (S235). Then, with the shift amount, the shift amount change amount, and the hanging degree set by the hanging degree setting key as input information, the upper feed dog and the upper feed dog are matched so that the patterns of the two work cloths match based on a preset fuzzy rule. The change amount of the feed amount with respect to the feed amount of the lower feed dog is calculated by fuzzy inference (S240). Then, while adjusting the feed amount by the upper feed dog according to the result of the fuzzy inference, the feed amount is fed to the sewing position, and the sewing needle is sewn without causing any sewing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a feed amount adjusting sewing machine which detects a mark on a work cloth and stitches the two work cloths so that the marks match each other.

[0002]

2. Description of the Related Art Conventionally, as this type of sewing machine, a sewing machine described in JP-A-1-192388 has been known. This sewing machine detects the intensity data of the reflected light from the work cloth on which the pattern as the mark is drawn a predetermined number of times according to the pattern interval, and the operator selects the intensity data from a plurality of different algorithms. Then, the deviation amount is calculated by processing according to the set algorithm, and the relative feed amount of the two work cloths is controlled according to the deviation amount so that the marks of the two work cloths match. It was

[0003]

However, in such a conventional device, the shift amount can be accurately detected, but the feed amount of the upper feed dog is changed even if the upper feed dog pressure and the pressing pressure are constant. At times, there are variations in the actual feed amount of the work cloths, and it may not always be possible to properly feed the two work cloths so that the marks of the work cloths match.

In such a case, since the operator has repeatedly adjusted the upper feed dog pressure and the pressing pressure so that the operator can perform appropriate control, the adjustment time and
There was a problem that it took time. Then, this invention is made in order to solve the said subject, The objective is to provide the feed amount adjusting sewing machine which can control a feed amount appropriately more simply.

[0005]

In order to achieve the above object, the feed amount adjusting sewing machine of the present invention, as illustrated in FIG. 1, includes mark detecting means M2 for detecting the marks of two work cloths M1, respectively. A stitching means M3 for overlapping and stitching the two work cloths M1 and two work cloths M1 are sent to the stitching means M3 according to a predetermined feed amount and at least one of the two work cloths M1 is In a feed amount adjusting sewing machine including a cloth feeding means M4 capable of adjusting the feed amount, a deviation amount calculating means M5 for calculating a deviation amount of the marks from the marks of the two work cloths M1 detected by the mark detecting means M2. A deviation amount change amount calculation means M6 for calculating a deviation amount change amount from the deviation amount calculated by the deviation amount calculation means M5, a engagement condition setting means M7 for setting an engagement condition, and the deviation amount,
As the input information, the shift amount change amount and the engagement degree,
A fuzzy inference means M8 is provided for calculating a feed amount change amount with respect to the feed amount of the cloth feeding means M4 by fuzzy inference so that the marks of the two work cloths M1 match based on a preset fuzzy rule.

[0006]

In the feed amount adjusting sewing machine having the above structure, the mark detecting means M2 detects the marks of the two work cloths M1 fed by the cloth feeding means M4. Then, the deviation amount calculating means M5 calculates the deviation amount from the mark of the two work cloths M1 detected by the mark detecting means M2, and the deviation amount change amount calculating means M6 calculates the deviation amount change amount from the deviation amount. calculate.

Further, the fuzzy inference means M8 processes the two sheets on the basis of the preset fuzzy rule by using the deviation amount, the deviation amount change amount, and the engagement degree set in the engagement degree setting means M7 as input information. The feed amount change amount with respect to the feed amount of the cloth feeding means M4 is calculated by fuzzy inference so that the marks of the cloth M1 are matched. Then, the cloth feed means M4 adjusts the feed quantity of at least one of the work cloths in accordance with the change in the feed quantity, and the sewing means M3 superposes the two work cloths and stitches them.

[0008]

An embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 2 is a schematic view of the essential parts of a feed amount adjusting sewing machine which is an embodiment of the present invention. The sewing needle 1 is configured to be reciprocally moved up and down by being driven by a sewing machine motor 2 (see FIG. 7), and a thread catcher driven by the sewing machine motor 2 arranged below the needle plate 3. The seams are formed in cooperation with (not shown). In this embodiment, these constitute the suturing means M3.

Further, the lower feed dog 4 performing the movement of arrow B shown in FIG. 3 in which vertical movement and horizontal movement are combined, and arrow C shown in FIG. 3 in which vertical movement and horizontal movement are also combined.
The upper feed dog 6 for performing the above movement is provided, and the lower feed tooth 4 and the upper feed tooth 6 are both configured to be driven by the sewing machine motor 2.

Further, two work cloths 1 are made by the work clamp foot 8.
0 and 12 are pressed from above, and the respective motions of the sewing needle 1, the lower feed dog 4, the upper feed dog 6 and the thread catcher (not shown) are executed in synchronization with each other. Processing cloth 1
0 and 12 are overlapped and sewn, and are transferred in the direction of arrow A in FIG.

The lower feed amount of the lower feed dog 4 in the direction of arrow A per cycle, that is, the lower feed pitch is configured to be adjustable by an adjusting mechanism (not shown). The pitch is detected by potentiometer 14 operably connected to the adjustment mechanism. Further, the upper feed amount of the upper feed dog 6 in the direction of arrow A per cycle, that is, the upper feed pitch is configured to be adjustable according to the rotation of the step motor 15.

Therefore, the lower feed pitch of the lower feed dog 4 is adjusted by the adjusting mechanism to feed the lower work cloth 12 at a constant feed amount, while the upper feed pitch of the upper feed tooth 6 is adjusted by the step motor 15. The feed amount of the upper work cloth 10 can be changed relative to the feed amount of the lower work cloth 12 by appropriately changing. As a result, the upper and lower work cloths 10,
The relative feed amount between the 12 is adjustable. In the present embodiment, these constitute the cloth feeding means M4.

The needle position detecting sensors 16 and 18 for outputting the needle down signal and needle up signal of the sewing needle 1 and the sewing machine main shaft (not shown) rotated by the driving of the sewing machine motor 2 are synchronized with the rotation. A rotation synchronization sensor 20 that outputs a signal is also provided. Further, a generating circuit 24 for generating a signal according to the depression of the foot pedal 22 is provided. The generation circuit 24 outputs an activation signal when the foot pedal 22 is depressed.

The upper and lower work cloths 10 and 12 are designed to pass between three cloth guide plates 26, 28 and 30 which are laminated at predetermined intervals. Lower cloth guide plate 2
6, pins 32 and 34 that stand through the long holes formed in the other cloth guide plates 28 and 30 are provided upright, and the pins 32 and 34 are the edge edges of the two upper and lower work cloths 10 and 12. And the lateral displacement with respect to the feeding direction is regulated.

The center cloth guide plate 28 is equipped with a detector 36 which is the mark detecting means M2 of the present invention for detecting the pattern as the mark of the upper and lower work cloths 10 and 12. As shown in FIG. 4, the detector 36 is provided with prism bodies 38 and 40. The detector 36 utilizes the reflection of the prism bodies 38 and 40 to project light onto the upper and lower work cloths 10 and 12, and The light reflected on the surfaces of the work cloths 10 and 12 is also incident by utilizing the reflection of the prism bodies 38 and 40.

An optical fiber bundle 42 is connected between the detector 36 and the light projecting section 52 and the light receiving section 54. The optical fiber bundle 42 is configured as an assembly of a large number of optical fibers, and one end of the optical fiber bundle 42 is divided into two as shown in FIG. And an optical fiber bundle 42b for the lower processing cloth, which is arranged so as to face the prism bodies 38 and 40, respectively.

As shown in FIG. 5, in the optical fiber bundle 42a for the upper work cloth, a plurality of light-transmitting optical fiber bodies Fa are located at the center and a plurality of light-receiving optical fiber bodies Fb are arranged on both sides thereof at a predetermined ratio. It is housed in. Optical fiber bundle 4 for roughing cloth
The same applies to 2b. The other ends of the upper processing cloth optical fiber bundle 42a and the lower processing cloth optical fiber bundle 42b are
An optical fiber 46 for projecting light, two sets of optical fibers 48 for receiving light,
The lens 56 of the light projecting section 52 and the photodiode sections 60 and 62 of the light receiving section 54 are arranged to face each other.

The light projecting section 52 is provided with a light source 58 for projecting white light onto the end surface of the light projecting optical fiber 46 via a lens 56. The light receiving section 54 includes a light receiving optical fiber 4
Top work cloth 10 that receives light from the end faces 8 and 50, respectively
And a photodiode portion 62 for the lower work cloth 12 are provided.

These photodiode parts 60, 62
As shown in FIG. 6, it is composed of a plurality of photodiodes provided with a color filter that transmits red (R), blue (B), and green (G) light with high sensitivity in the incident window. Since the color filters are arranged apart from one another having the same hue so as to widen the light receiving range, the light receiving optical fiber 4
Even if the light from the end faces of 8 and 50 is slightly shifted and projected on the photodiode parts 60 and 62, red (R), blue (B),
Light of each hue of green (G) can be efficiently detected.

Therefore, the white light projected from the light source 58 is reflected by the prism bodies 38 and 40 at the tip of the detector 36 via the optical fiber 46 for projecting light, and the upper and lower work cloths 10 and 1 are processed.
2 is projected. The light reflected by the upper and lower work cloths 10 and 12 is reflected by the prism bodies 38 and 40, traces the same optical path as the projection in the reverse direction, and passes through the optical fibers 48 and 50 for light reception to the photodiode parts 60 and 62. The output is input to the electronic control circuit 70 also provided in the control box 44.

The electronic control circuit 70, as shown in FIG.
In addition to the well-known CPU 73, ROM 75, RAM 78, an A / D converter 80 to which the photodiode units 60, 62 are connected, a drive circuit 94 for driving the step motor 15 for adjusting the upper feed amount, and a sewing machine motor 2 as a power source. A driving circuit 96 for driving is provided.

In addition to this, the electronic control circuit 70 is provided with a rotation synchronizing sensor which is provided in a pulley (not shown) and outputs a pulse signal of 24 [pieces / revolution] in synchronization with the rotation of a main shaft (not shown). 20, a needle position detection sensor 16, which is also provided in a pulley (not shown) and outputs a needle down signal and a needle up signal,
18, the above-mentioned potentiometer 14 for detecting the cloth feed amount of the lower processing cloth 12, and the foot pedal 22 for starting and
A generator circuit 24 for generating a stop signal is connected.

Further, an engagement degree setting key 25 as an engagement degree setting means M7 of the present invention for setting an engagement degree determined by the upper feed dog pressure, the presser foot pressure, the material quality, etc. is also connected. In this embodiment, this degree of engagement is a value indicating that the closer the value is to 100, the better the response of the actual feed amount of the upper work cloth 10 to the upper feed amount of the upper feed dog 6. When the upper feed is large in pressure, the value of the applied degree becomes large, and when the presser pressure is large, the value of the applied degree becomes small.
The value also varies depending on the material quality.

A program for a pattern matching control routine is written in the ROM 75 in advance. Then, in the pattern matching control, there are three pattern shift amount calculation modes as modes for calculating the pattern shift amount by algorithms of different methods. In the present embodiment, the brightness calculation subroutine (mode 1), the hue difference calculation subroutine (mode 2), and the hue difference emphasis calculation subroutine (mode 3) correspond to these three pattern displacement amount calculation modes, respectively. Remembered in.

Further, in the pattern matching control, the upper feed amount is properly adjusted so that the patterns of the upper and lower work cloths 10 and 12 are made to coincide with each other. A fuzzy inference program for calculating the upper feed amount change amount. Is also written in the ROM 75 in advance. Further, the RAM 78 is assigned with a memory area of a specified number Cm in which the color signal data sampled by the photodiode units 60 and 62 are sequentially stored. The RAM 78 is constantly backed up by a backup power source (not shown) so that necessary data and the like can be stored.

Next, the operation of the thus configured feed amount adjusting sewing machine will be described with reference to the flow charts of FIGS. First, interrupt routine (see Figure 8)
Let's start with. The interrupt routine is started by the fall of the synchronization signal of the rotation synchronization sensor 20. Since the rotation synchronization sensor 20 generates 24 synchronization signals each time the main shaft (not shown) makes one rotation, the main shaft (not shown) has one
The interrupt routine is repeatedly executed every time it rotates 5 degrees.

First, whether the synchronizing signal from the rotation synchronizing sensor 20 is within the feed operation period of the upper and lower work cloths 10 and 12, that is, the sewing needle 1 is raised to move the upper feed dog 6 and the lower feed dog 4 It is determined whether or not an interruption has occurred during the feeding operation period during which the upper and lower work cloths 10 and 12 are being transferred (S50).
0). This operation period can be determined by the detection data of the needle position detection sensors 16 and 18. If it is not within the feed operation period (S500: NO), nothing is executed and this routine is terminated and the process returns to the pattern matching control routine.

On the other hand, when the synchronizing signal from the rotation synchronizing sensor 20 is within the feeding operation period (S500: YES), the analog signals from the photodiode portions 60 and 62 are A / A.
Six color signals converted into digital signals by the D converter 80 (red, blue, green color signals of the upper work cloth 10, lower work cloth 1)
2 red, blue, and green color signals) are stored in the RAM 78 as one set of color signal data (S510), and the number C of color signal data is stored.
After incrementing the value of 1 by 1 (S520), the process returns to the execution of the pattern matching control routine.

As a result, only the color signal data detected while the upper and lower work cloths 10 and 12 are being transferred by the upper feed dog 6 and the lower feed tooth 4 in the direction of arrow A in FIGS. Accumulated in. Next, the pattern matching control routine will be described with reference to FIGS. 9 and 10.

First, before the sewing is started, the pattern deviation amount calculation mode, the pattern interval, and the degree of engagement determined by the upper feed dog pressure, the presser foot pressure, the material quality, etc. are set in advance (S).
100). It should be noted that the setting of the hanging degree may be manually input from the hanging degree setting key 25, or may be automatically set by trial sewing or the like.

Next, the specified length L is set based on the pattern interval, and the specified number Cm, which is the number of color signal data from the work cloths 10 and 12 corresponding to the specified length L, is calculated ( S110, 120). For example, the specified length L is 30 [m
m] and the feed amount are set to 1 [mm], the number of synchronization signals in the feed operation range per revolution of the main spindle is 10 [pulses], so 10 [pulses] × 30 [mm] / 1 The prescribed number Cm is determined to be 300 by calculating [mm].

Subsequently, the control count K and the color signal data count C assigned to the RAM 78 are cleared to the value 0 (S130, 140). And the upper and lower work cloth 1
It is determined whether 0 or 12 is present (S150). This determination is made because the output level from the detector 36 is low when the work cloths 10 and 12 are present, and the output level from the detector 36 is very high when the work cloths 10 and 12 are not present.

When it is determined that the upper and lower work cloths 10 and 12 are present (S150: YES), it is determined whether or not the operator depresses the foot pedal 22 via the generating circuit 24 (S160). ). When it is determined that the pedal is turned on (S160: YES), the sewing machine motor 2 is driven to perform sewing (S170).

During the sewing operation, the interrupt routine (see FIG. 8) described above is executed in synchronization with the rotation of the sewing machine motor 2,
New color signal data is sequentially stored in a predetermined area of the RAM 78. Next, it is judged whether or not the control count K is 0 (S180), and when the control count K is 0 (S1).
80: YES), and subsequently, it is determined whether or not the number C of color signal data has reached the specified number Cm (S190).

Then, the number C of color signal data is the specified number C
If m has not been reached (S190: NO), the process after step S150 is repeated and kept waiting, and if the specified number Cm is reached for the first time (S190: YES), or if the control count K is not 0 in step S180. When the judgment is made, that is, when the control is already started (S180: N
O), the process proceeds to the pattern matching process after step S200 shown in FIG.

As shown in FIG. 10, in the pattern matching process,
First, the number of the pattern shift amount calculation mode set in advance is determined (S200), and the brightness calculation subroutine (S210), the hue difference calculation subroutine (S220), and the hue difference emphasis calculation subroutine are determined according to the set mode. (S
230) is executed to calculate the pattern displacement amount.

The process of calculating the pattern deviation amount by executing the difference emphasis calculation subroutine when the pattern deviation amount calculation mode is the mode 3 will be briefly described below (see FIG. 12). In the difference emphasis calculation subroutine, first, the RAM 78
From the latest color signal data accumulated in the predetermined area to the specified number Cm before the color signal data, the difference emphasis signals of the upper and lower work cloths 10 and 12 are obtained.

Subsequently, the difference emphasis signal data is smoothed to remove the influence of noise and the like, and then the smoothed data is differentiated to make a large change in the smoothed data more noticeable and gradual. Smooth out changes. Then, the pattern deviation amount d and the pattern deviation direction are calculated using this differential data, and stored in association with the control count K.

When the pattern deviation amount d is obtained in this manner, the difference between the previous pattern deviation amount dn-1 and the current pattern deviation amount dn (hereinafter, the pattern deviation amount change amount dd (= dn-dn-1)) is obtained. (Referred to) is calculated (S235). Next, fuzzy inference is performed based on the pattern deviation amount d, the pattern deviation amount variation amount dd, and the degree of application h (S240). Then, the upward feed amount adjusting process (S300) is performed according to the result of the fuzzy inference.

Thereafter, the control count K is incremented by 1 (S310), and the process proceeds to step S150 described above in FIG. By adjusting the upper feed amount in this way, the relative feed amount between the upper and lower work cloths 10 and 12 is controlled, and the upper and lower work cloths 1 and 12 are controlled.
0 and 12 are sewn so that the handles of both are matched.

Next, regarding fuzzy inference, FIG. 11 and FIG.
3, with reference to FIG. The pattern shift amount d, the pattern shift amount change amount dd, and the application degree h required for the control are obtained by step S235 in FIG. 10, and from the obtained pattern shift amount d, the pattern deviation amount variation amount dd, and the application state h. , Figure 1
With the membership function shown in FIG. 3, the degree of suitability for each pattern deviation amount d, the amount of pattern deviation amount variation dd, and the degree of hanging h are obtained.

This membership function is stored in the ROM 75 in advance, and the horizontal axis of each membership function is
It corresponds to each pattern displacement amount d, the pattern displacement amount variation amount dd, and the degree of hanging h, and the vertical axis corresponds to the degree of conformity. FIG.
(A) is a characteristic view showing a membership function relating to the pattern deviation amount d. The fuzzy set of this pattern deviation amount d is 3
The fuzzy sets are classified into ranks MI, ZR, and PL.

In the rank MI, the degree of conformity gradually changes from 1 to 0 when the pattern deviation amount d is between -3.0 mm and 0 mm. In the rank ZR, the degree of conformity gradually changes from 0 to 1 when the pattern displacement amount d is between −3.0 mm and 0 mm, and the pattern displacement amount d is 0 mm to 3.0.
The goodness of fit gradually changes from 1 to 0 in mm. In the rank PL, the degree of conformity sequentially changes from 0 to 1 when the pattern deviation amount d is 0 mm to 3.0 mm.

FIG. 13B is a characteristic diagram showing a membership function relating to the pattern deviation amount change amount dd. The fuzzy set of the pattern shift amount change amount dd is a rank M of three stages.
It is divided into I, ZR, and PL fuzzy sets. Rank MI has a pattern shift amount change amount dd of -3.0 mm to 0 mm
The fitness level sequentially changes from 1 to 0. Rank ZR
Indicates that the degree of conformity changes sequentially from 0 to 1 when the pattern deviation amount change amount dd is between −3.0 mm and 0 mm, and the pattern deviation amount change amount dd is 0.
The conformity changes from 1 to 0 in the range from mm to 3.0 mm. Further, in the rank PL, the degree of conformity sequentially changes from 0 to 1 when the pattern displacement amount change amount dd is from 0 mm to 3.0 mm.

FIG. 13C is a characteristic diagram showing a membership function relating to the degree of h. The fuzzy set having the degree h is divided into three fuzzy sets of ranks S, M, and L. In the rank S, the degree of conformity changes from 0 to 50 and the fitness level changes from 1 to 0 sequentially. In the rank M, the fitness level changes from 0 to 1 in the range from 0 to 50, and the degree of h is from 50 to 1.
During 00, the goodness of fit sequentially changes from 1 to 0. In the rank L, the degree of suitability is changed from 0 to 1 in the range of 50 to 100.

Next, the control rules stored in advance in the ROM 75 will be described. The control rule is based on the above-described membership function relating to the pattern deviation amount d, the pattern deviation amount change amount dd, and the hanging degree h.
The type is set. (1) When the pattern shift amount is negative (d = MI) and the pattern shift amount change amount is negative (dd = MI), the upward feed amount change amount is large in the positive direction (ds = PL). (2) When the pattern deviation amount is negative (d = MI) and the pattern deviation amount change amount is 0 (dd = ZR), the upward feed amount change amount is small in the positive direction (ds = PS). (3) When the pattern shift amount is negative (d = MI) and the pattern shift amount change amount is positive (dd = PL), the upper feed amount change amount is 0 (ds = ZR). (4) When the pattern shift amount is 0 (d = ZR) and the pattern shift amount change amount is negative (dd = MI), the upper feed amount change amount is small in the positive direction (ds = PS). (5) When the pattern shift amount is 0 (d = ZR) and the pattern shift amount change amount is 0 (dd = ZR), the upper feed amount change amount is 0 (ds = ZR). (6) When the pattern deviation amount is 0 (d = ZR) and the pattern deviation amount change amount is positive (dd = PL), the upper feed amount change amount is small in the negative direction (ds = MS). (7) When the pattern shift amount is positive (d = PL) and the pattern shift amount change amount is negative (dd = MI), the upper feed amount change amount is 0 (ds = ZR). (8) When the pattern shift amount is positive (d = PL) and the pattern shift amount change amount is 0 (dd = ZR), the upper feed amount change amount is small in the negative direction (ds = MS). (9) When the pattern shift amount is positive (d = PL) and the pattern shift amount change amount is positive (dd = PL), the upper feed amount change amount is large in the negative direction (ds = MS). (10) When the application condition is good (h = L), the amount of change in the upper feed amount is 0 (ds = ZR). (11) When the pattern displacement amount is positive (d = PL) and the degree of engagement is normal (h = M), the upward feed amount change amount is small in the negative direction (ds = MS). (12) When the pattern shift amount is negative (d = MI) and the degree of engagement is normal (h = M), the upward feed amount change amount is small in the positive direction (ds = PS). (13) When the pattern displacement amount is positive (d = PL) and the engagement condition is bad (h = S), the upward feed amount change amount is large in the negative direction (ds = ML). (14) When the pattern displacement amount is negative (d = MI) and the engagement condition is bad (h = S), the upward feed amount change amount is large in the positive direction (ds = PL).

In the above-mentioned fuzzy inference, as shown in FIG. 11, first, the degree of conformity is calculated for each input information of the pattern displacement amount d, the pattern displacement amount variation amount dd, and the application degree h (S25).
0). For example, when the pattern deviation amount d is -1.2 mm, the pattern deviation amount variation amount is 0 mm, and the hanging degree is 30, FIG.
According to the membership function of (a), the pattern deviation amount d is 0.4 for the rank MI, 0.6 for the rank ZR, and 0 for the rank PL.

Further, according to the membership function of FIG. 13B, the pattern deviation amount variation amount dd has a suitability of 1 for the rank ZR and 0 for the ranks MI and PL. Figure 1
From the membership function of 3 (c), the degree of conformity h is 0.4 for the rank S, 0.6 for the rank M, and 0 for the rank L.

Then, for each of the control rules (1) to (14), the suitability of the entire condition part in the first half of each control rule is obtained (S).
260). When there is a plurality of input information of the condition part, the smaller one of the respective adaptability is set as the adaptability of the entire condition part. Considering the above-described example, in the control rule (1), the rank of the pattern deviation amount d is MI and the rank of the pattern deviation amount change amount dd is MI, so the suitability of the former is 0.4 and the suitability of the latter is 0.4. Is 0. Therefore, the goodness of fit of the entire condition part is 0
In this example, control rule (1) is ignored.

Similarly, in the control rule (2), the rank of the pattern deviation amount d is MI and the rank of the pattern deviation amount variation dd is ZR.
Therefore, the former fitness is 0.4 and the latter fitness is 1.
Becomes Therefore, the amount of change in the upper feed amount is small in the positive direction (d
s = PS), the fitness of the entire condition part is 0.4. Hereinafter, the control rules (3) to (14) will be obtained.

The control rules that are valid in this example are:
(2), (5), (12) and (14). In the control rule (5), since the rank of the pattern deviation amount d is ZR and the rank of the pattern deviation amount variation dd is ZR, the former suitability is 0.
At 6, the latter conformance becomes 1. Therefore, the change amount of the upper feed amount is 0 (ds = ZR), and the conformity of the entire condition part is 0.
It becomes 6.

In the control rule (12), the pattern deviation amount d
Since the rank is MI and the rank of the degree of h is M, the suitability of the former is 0.4 and the suitability of the latter is 0.6. Therefore, the change amount of the upper feed amount is small in the positive direction (ds =
PS), the fitness of the entire condition part is 0.4.

Further, according to the control rule (14), the pattern deviation amount d
The rank of MI is MI and the rank of the degree of h is S, so the suitability of the former is 0.4 and the suitability of the latter is 0.4. Therefore, the amount of change in the upper feed amount is large in the positive direction (ds =
PL), the fitness of the entire condition part is 0.4.

Then, based on the goodness of fit obtained for each control rule, the membership function of the conclusion part operation amount (here, the amount of upward feed change) is corrected as shown in FIG. 14 (S
270). Some control rules modify variables of the same rank, but in this case, the one with a higher degree of conformity is given priority.

Finally, the final change amount of the upper feed amount is obtained from the corrected membership function of the change amount of the upper feed amount (S280). Specifically, it is obtained by obtaining the center of gravity of the corrected upper feed amount variation membership function. This center of gravity can be calculated by the following formula.

Upper feed amount change amount = Σ (upper feed amount change amount × fitness) / Σ (fitness) In the above example, the upper feed amount change amount ds is +2. The step motor 15 is rotated according to the calculated change amount ds of the upper feed amount to adjust the upper feed amount. The fuzzy inference routine is executed by the fuzzy inference means M of the present invention.
Work as 8.

As described above, in the feed amount adjusting sewing machine of this embodiment, the upper feed amount change amount ds, the pattern deviation amount d, the pattern deviation amount change amount dd, the upper feed dog pressure, the presser foot pressure, the raw pressure Fuzzy inference is performed based on the degree of h that is determined by the geology and the like, and the upper feed amount is increased or decreased by the upper feed amount change amount ds.

Therefore, in the past, the upper feed dog pressure, the pressing pressure, etc. had to be adjusted in order to perform proper control, but even if the upper feed dog pressure and the pressing pressure are constant without adjustment. The upper feed amount is increased / decreased by the upper feed amount change amount ds corresponding to the material quality, the upper feed dog pressure and the pressing pressure so that the patterns of the upper and lower work cloths 10 and 12 are matched. Therefore, the feed amount can be easily and appropriately controlled.

The embodiment of the present invention has been described above.
The present invention is not limited to such embodiments, and it goes without saying that the present invention can be implemented in various modes without departing from the spirit of the present invention. Furthermore, if high-precision control is to be performed, it can be realized by increasing the number of input information and control rules, or by increasing the number of ranks of the membership function of input information. Further, in this embodiment, the upper feed amount is changed with respect to the lower feed amount, and the upper and lower work cloths 1 are changed.
Although the relative feed amount between 0 and 12 is controlled, the reverse may be possible or both may be changed.

Further, the present invention uses the work cloth 10 as a mark,
The sewing machine is not limited to the sewing machine that detects the pattern 12 and performs pattern matching sewing, and may detect the marks of the work cloths 10 and 12 to which predetermined marks are attached. As the mark, a magnetic material (in this case, a magnetic sensor is used as the mark detecting means) woven in advance on the work cloths 10 and 12 at predetermined intervals, or on the edges of the work cloths 10 and 12 are used. There are notches such as corrugated edges formed.

[0061]

As described above in detail, in the feed amount adjusting sewing machine of the present invention, the upper feed amount changes corresponding to the material quality, the upper feed dog pressure and the pressing pressure so that the marks of the two work cloths match. Since the upper feed amount is increased / decreased depending on the amount, there is an effect that the upper feed dog pressure and the pressing pressure can be simply and appropriately controlled without adjusting each time so as to appropriately control.

[Brief description of drawings]

FIG. 1 is a block diagram illustrating a basic configuration of a feed amount adjusting sewing machine of the present invention.

FIG. 2 is a main part configuration diagram schematically showing a feed amount adjusting sewing machine as one embodiment of the present invention.

FIG. 3 is a cross-sectional view of essential parts showing a configuration of a vertical feed dog and a detector of the feed amount adjusting sewing machine of the present embodiment.

FIG. 4 is an enlarged perspective view of the tip structure of the detector of this embodiment.

FIG. 5 is an explanatory diagram illustrating a configuration of a fiber bundle of the detector according to the present exemplary embodiment.

FIG. 6 is an explanatory diagram illustrating a configuration of a color filter of the detector according to the present exemplary embodiment.

FIG. 7 is a block diagram showing a configuration of an electric system of the present embodiment.

FIG. 8 is a flowchart showing an interrupt routine of this embodiment.

FIG. 9 is a flowchart showing the first half of the pattern matching control routine of this embodiment.

FIG. 10 is a flowchart showing the latter half of the pattern matching control routine of this embodiment.

FIG. 11 is a flowchart showing a fuzzy inference routine of this embodiment.

FIG. 12 is an explanatory diagram showing processed data obtained by converting color signal data from the processed cloth of this embodiment into a brightness signal, a difference signal, and a difference emphasis signal.

FIG. 13 is an explanatory diagram showing a membership function of input information in fuzzy inference according to the present embodiment.

FIG. 14 is an explanatory diagram showing a membership function modified by a goodness of fit obtained for each control rule in fuzzy inference according to the present embodiment.

[Explanation of symbols]

M1, 10, 12 ... Work cloth M2 ... Mark detecting means
M3 ... stitching means M4 ... cloth feed means M5 ... deviation amount calculation means M6 ... deviation amount change amount calculation means M7 ... engaging degree setting means M8 ... fuzzy inference means 1 ... sewing needle
2 ... Step motor 4 ... Lower feed tooth 6 ... Upper feed tooth
8 ... Cloth foot 15 ... Step motor 36 ... Detector
52 ... Emitting unit 54 ... Light receiving unit 70 ... Electronic control circuit
73 ... CPU 75 ... ROM 78 ... RAM

Claims (1)

[Claims] 1. A mark detecting means for detecting a mark on each of the two work cloths, a stitching means for superposing the two work cloths on each other, and stitching the two work cloths according to a predetermined feed amount. A feed amount adjusting sewing machine capable of adjusting the feed amount of at least one of the two work cloths while feeding the work piece to the suturing means, and the two work pieces detected by the mark detecting means. A deviation amount calculation means for calculating the deviation amount of the mark from the cloth mark, a deviation amount change amount calculation means for calculating the deviation amount change amount from the deviation amount calculated by the deviation amount calculation means, and the degree of engagement are set. The hanging condition setting means and the cloth feeding means so that the marks of the two work cloths match based on a preset fuzzy rule using the deviation amount, the deviation amount change amount, and the hanging condition as input information. Sending Feed amount adjusting sewing machine, characterized by comprising a fuzzy inference means for calculating the fuzzy inference feed rate variation to the amount.