JPS6346515A - Pulse distributing method - Google Patents

Abstract

PURPOSE:To realize a pulse distributing method that can trace fitfully a target route by selecting a command pulse to perform a shift to the candidate point nearest to the target route among those candidate points set in each axial direction which is shifted by a single pulse. CONSTITUTION:In a linear interpolation mode an initialization part 1 writes the initial value related to each desired straight line to the memory parts 2-8 respectively after reception of an input. A code comparing part 9 discriminate the code of the value D stored in a D memory part 2 to start an X process/pulse generating part 10 with D<=0 and then a Y process/pulse generating part 11 with D>0 respectively. The part 10 adds '1' to the value X of an X memory part 5 and the value 2b to a 2b memory part 3 to produce a single command pulse of the plus (+) direction. While the part 11 adds 1 to the value Y of a Y memory part 6 and also subtracts the value 2a of a 2a memory part 4 from the value D to produce a single command pulse of the (+) direction. An end point comparing part 12 is started for each end of processing of both parts 10 and 11 to perform comparison between the value X and the value Xe of an X end point memory part 7 as well as between the value Y and the value Ye of a Y and point memory part 8. The processing is finished when coincidence is obtained in both comparisons.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a pulse distribution method, and more particularly to a method of distributing pulses so that a tool follows a target course in an NC device or the like.

[Conventional technology]

As a conventional pulse distribution method for this purpose, there is an algebraic calculation method. This method sequentially calculates algebraic equations for straight lines and circular arcs, and then performs Nords distribution so that the tool follows the line while determining whether each point that is not on the line is positive or negative. This is the way to go.

First, to explain the principle of linear pulse distribution, in Fig. 6, starting from the origin (0,0), point Q (x,
When considering a straight line ending at t)'1), the following holds true for the point P(x:, y) in the middle of the line: Here, consider a function (discriminant) called D.

D Proverbs As a result, if D<0, it means that it has reached the lower side, so next, one pulse is output in the y direction, and pulses are continued to be output one pulse at a time until D) becomes 0.

In other words, the current point is either above (D>0) or below (D<
0), and output pulses alternately with D)Old<O.

Next, the principle of pulse distribution for one circular arc will be explained. For example, as shown in FIG. 7, consider a circular arc whose center is at the coordinate origin (0,0), and set the starting point of the circular arc to (x).

yo), the end point is (Xl p 71 )- and the point on the arc is P, (x, y), then x2+ y2a* x62+y□ am x1+y, 2
holds true. Therefore, if we consider a function D, p −x
The discriminant 2+ y2 , 2 , 2 holds true.

If we think in the same way as for straight lines, we can see that we can construct a system that tracks arcs within an error of 1 nogle.

Note that 1 noguru corresponds to a minute movement amount, for example 0.0111i11, so the movement of the tool will not be step-like as shown in Figures 6 and 7, but rather a straight line or circular arc to be interpolated. The movement is smooth.

[Problem that the invention seeks to solve]

However, the drawback of the above method is that although the trajectory of the tool will never be more than 1/4' lus from the target straight line or curve, it does not necessarily mean that the point closest to the target straight line is interpolated. That's true.

For example, for a straight line with a small slope as shown in FIG. 5, it was not possible to draw a trajectory as shown in FIG. 5(&) and a desired trajectory as shown in FIG. 5(b).

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a pulse distribution method that can always distribute pulses so as to form a locus connecting points closest to a target straight line or curve.

[Means for solving problems]

According to the present invention, there is provided a pulse distribution method for distributing to each axis a command/gurus instructing movement in each axis direction of a plurality of axes so as to move along a target path, the command/gurus to be distributed next. When determining the direction of movement, a candidate point that is also close to the target path KA is determined among a plurality of candidate points in each axis direction that are moved in one pulse from the current command position, and commands in the direction of movement to proceed to this candidate point are determined. At the same time, the current command position is updated to the position of the candidate point.

[Effect]

In other words, in the past, pulses were distributed by simply determining whether the current point was on the upper or lower side of the target path, but the present invention takes this one step further and distributes pulses based on 1/4 pulse distribution. The command pulse for moving to the candidate point closest to the target path is selected among multiple candidate points for each axis direction to be moved, and the pulse distribution necessary to take the point closer to the target path is made possible. ing.

〔Example〕

The present invention will be described in detail below with reference to the accompanying drawings.

First, the principle of the present invention will be explained using an example of interpolating straight lines. Now, consider the straight line shown in the following equation.

Knee L-0...(υ breath b (a)O, b>o) 之・mi, and suppose that the first point is also interpolated as shown in Figure 3 - the coordinates at this time are pi( x, y).

Then, the next point to be interpolated is a point that is closer to a straight line among the points advanced by +1 in the xi or Y direction.

In other words, the i+1th candidate point is 2 legs, (X+1.
These are Y) and (X, Y+1).

The former will be represented by P1+x and the latter by Pi+)r.

The distance from Pi +x, Pi +y to the straight line is Ll
+x.

If Li+y, then the difference dL between Li+x and Li+y is
Considering this, this indicates which of Pl+x and pi+y is closer to the straight line. When calculated, dL = L
l +x2 Li +y2 (3) If dL is positive, proceed in the Y direction, and if dL is negative, proceed in the X direction to reach a point close to a straight line. In equation (3), a) 0. b) Since O, the discriminant formula is o-2bX-2aY+b-! It can be set as L-(4).

Therefore, by substituting the current point (x, y) into equation (4) and checking the sign of the discriminant, the command base is issued according to the rule, and each time it is issued, the current point is By repeating the procedure of updating the value, determining whether the discriminant is positive or negative based on the updated value, and issuing a command notarus again according to the above rules, the desired pulse distribution, that is, linear interpolation, becomes possible.

By the way, calculation 1]4 of the value of the discriminant in the above equation (4) can be done by only addition and subtraction without the need for multiplication, if the following recurrence formula is used. That is, if the first discriminant is Dl, the discriminant Di+x when proceeding in the X direction is expressed as Di-1x -2b (X+1)-2aY-)b-a stomach nt+2b (5) Similarly, when moving in the Y direction, the discriminant % formula% Also, in the initial state, x-o, y=o, so D
From Equation (4), D Maroba...(7) is obtained.

To summarize the above, if the initial value is D≦0, a command pulse of +X is issued, the value of X is increased by +1 as shown below, and according to equation (5), the following discriminant seek.

Similarly, if D) is 0, the +Y command/J pulse is issued, and the ( of Y is increased by +1 as shown below, and the (6th
), find the following discriminant.

In the above discussion, X and Y are arbitrary points Ps (Xs, Ym
) can be easily extended to a straight line starting with the same slope; in this case, the initial values of X and Y are set to Xs,
Just set it to Ys. Also, the end point Pe(Xs,,Ye
) is given, then P(x
, y) and Pe(Xe, Ya), and if they match, the interpolation may be terminated.

Next, the case of interpolating an ellipse will be explained.

(a) 0, b>o) As shown in Fig. 4, the first quadrant of the ellipse is a point (a.

Consider the case of interpolating from O). It is assumed that the interpolation point is advanced to a point that is closer to the ellipse among the points advanced by −1 in the X direction or +1 in the Y direction. Suppose now that the first point is interpolated.

Let the coordinates at this time be pt(x, Y). There are two i+i-th candidate points, (X-1.

These are Y) and (X, Y+1). The former is Pi-X,
The latter will be expressed as pi+y.

The intersection of the straight line connecting Pi-X (X-1, Y) and the origin is Pc
-x (Xc -x, Ye-x), and if the intersection of the straight line connecting pi+y (x, Y+ 1) and the origin is PC+y (Xe+y, Yc+y), then from point JjK to )'i-x Distance and Pc from origin
The ratio of the square of the distance at -xt is R1-x, and the ratio of the square of the distance from the origin to the circle 10y and the distance from the origin to Pc+yo is R1+y.

The smaller the absolute value of both minus 1, the closer it is to an ellipse. Therefore, DI −(R14-y
-1) (R1-x 1) ≧0...αQ
When rowing, it is sufficient to advance by -1 in the KX direction, and otherwise by +1 in the Y direction. However, R1+y)R1-x is unconditionally true for the first quadrant, and Equation 01 is D 1=(R1+y-Ri-x) (R1+y+R1-
x-2), so D2=
R1+y+R1-x-2≧0 can be set as a condition for determination. Expanding D2, we get D=2b because the denominator is always a positive constant.
X -2b X+2a2Y2+2a2Y+a2+b2-
2a2b2...αη and half j blade) j bow (・↓unishiya'+"z", ;u.

Therefore, by substituting the current point (x, y) into Equation 64,
By checking the sign of discriminant 〇, we issue commands and ruses according to the rule, update the value of the current point each time we issue them, and judge whether the discriminant is positive or negative based on the updated value.
By repeating the procedure of issuing command pulses again according to the above rules, desired pulse distribution, that is, special interpolation, becomes possible.

By the way, the value of the discriminant of Equation 64 can be calculated using only addition and subtraction without the need for multiplication, if the following recurrence formula is used as in the case of linear interpolation. That is, if the first discriminant is Dl, then the discriminant Di when proceeding in the -X direction
-: c is DI -x = DI-4b2X+4b2...-α frame +
When proceeding in the Y direction, it can be easily calculated using the recurrence formula: Di+y = DI+ 4a Y+4a2-Ql. Also, since Do is Xma and yo, DO--2ab2+ &2+ b2 .-].

To summarize the above, elliptic interpolation is possible by repeating K with the initial value as K.

If we further introduce the values DX and DY into the above equation and make D into a higher-order recurrence formula, the initial value is set to X 4-a 1 and DX+DX+4b2J? By repeating ``''7, elliptic interpolation becomes possible. It goes without saying that if amb is used, circular interpolation is also possible.

FIG. 1 is a block diagram showing an example of an apparatus for carrying out the method of the present invention, and is shown for the case where linear interpolation is performed.

In the figure, upon receiving an input, the initial setting section 1 writes initial values regarding desired straight lines into each of the storage sections 2 to 8, respectively. That is, the D storage unit 2 has (h・−&
) is written (Equation (8)), and 2b storage unit 3 and 2 & storage unit 4 each have 2b values related to the slope of the straight line.
, 2m are written in the X storage section 5 and the Y storage section 6.
The positions Xs and Ys, which are the starting points of the straight lines, are respectively written, and the positions X. and Y., which are the end points of the straight lines, are written in the X end point storage section 7 and the Y end point storage section 8, respectively.

The sign comparison section 9 discriminates the sign of the value stored in the D storage section 2, and when D≦O, starts the X processing/pulse generation section 10, and when D>0, starts the X processing/pulse generation section. 11.

When activated by the signal from the sign comparison unit 9, the X processing//9 pulse generation unit 10 adds 1 to the value X in the X storage unit 5, and adds the value in the 2b storage unit 3 to the value in the D storage unit 2. 2b
(see equation (9)), one command pulse in the +X direction is generated.

X processing...! When activated by the signal from the code comparison unit 9, the system generation unit 1 adds 1 to the value Y in the Y storage unit 6, and extracts the value 2 in the 2a storage unit 4 from the value in the D storage unit 2.
Subtract a and generate one pulse of the +Y force direction command.

The end point comparison section 12 is connected to the X processing/pulse generation section 10 or the X
It is activated every time the process of the processing/merus generation unit 11 is completed, and the value X in the X storage unit 5 and the values X@, Y in the X end point storage unit 7 are
The value Y in the storage unit 6 and the value Ye in the Y end point storage unit 8 are compared, and if they match, the process is terminated, and if not, the code comparison unit 9 is activated again.

In addition, in the above example, & + b ) 0 , Xs)
Xs, Ye) Although we have explained the case of Ys, a
+ b (When 0, the sign comparison unit 9 is D) When O, X
The processing/pulse generation section 10 is activated, and when D≦O, the X processing//4 pulse generation section 11 is activated and % and X@ (when Xs, the X processing/pulse generation section 10&i, the value
At the same time, subtract 1 from
b is subtracted to generate a command pulse in the -X direction, and Ye <
When Ys, the X processing/pulse generation section subtracts 1 from the value Y in the Y storage section 6 and adds the value 2 to the value in the D storage section 2.
All you have to do is add A and generate commands in the -Y direction) l) @ and S.

FIG. 2 is a block diagram showing another example of implementing the method of the present invention, in which elliptic interpolation is performed.

In the figure, upon receiving an input, the initial jtJ1 setting unit 21 writes initial values regarding desired ellipses into each of the storage units 22 to 30, respectively. That is, the D storage unit 22, DX storage unit 23, and DY storage unit 24 each have (
-2ab +a +b ), (-4b2m+ 4b2
) and (4a2) are written (mH formula). Further, the values 4b and 4& are written in the 4b storage unit 25 and the 41h2 storage unit 26, respectively, and the positions Xs and Y-, which are the starting points of the ellipse, are written in the X storage unit 27 and the Y storage unit 28, respectively, and the X ending point The positions Xs and Y, which are the end points of the ellipse, are written in the storage unit 29 and the X end point storage unit 30, respectively.

The sign comparison unit 31 discriminates the sign of the value stored in the D storage unit 22, and when D≧0, activates the The generator 33 is activated.

When activated by the signal from the sign comparison unit 32, the X processing/pulse generation unit 32 subtracts 1 from the value
DX is added to the value DX, and then 4 is added to the value DX in the DX storage section 23.
b Add the value 4b2 in the storage unit 25 (see formula (c)), and -X direction command x k x f 1 x occurrence f le.

When activated by a signal from the code comparison unit 31, the X processing/norse generation unit 33 adds 1 to the value Y in the Y storage unit 28, and adds 1 to the value DICDY storage unit 2 in the D storage unit 22.
Then add the value DY of 4a to the value DY of the DY storage unit 24, and then add the value 4&2 of the storage unit 26 of 4a (see equation (c)). k Stain occurrence f le.

The end point comparison section 34 is configured to perform the X processing/pulse generation section 32 or the X
It is started every time the processing of the processing/pulse generation section 33 is completed, and the value X of the X storage section 27 and the value Xe of the X end point storage section 29 are stored.
The value Y in the Y storage unit 28 and the value Ye in the X end point storage unit 30 are compared, and if they match, the process is terminated, and if not, the code comparison unit 31 is activated again.

Note that the above configuration can be configured using, for example, a logic circuit, and it is also possible to actually implement the entire configuration using an electronic computer. Further, the discriminant may be calculated directly from the current point without using the recurrence formula.

Further, in this embodiment, the case of interpolation of a straight line or an ellipse (including a circular arc) has been described, but the interpolation is not limited thereto, and any curve may be used as long as the target route is known.

Furthermore, in this embodiment, the case where the target route is two-dimensional has been explained, but it may also be three-dimensional. Just choose the closest point.

Furthermore, the present invention can be applied to drawing a desired straight line or curved line in a display device.

〔Effect of the invention〕

As explained above, according to the present invention, when determining the next command pulse to be distributed, the target path is selected from among a plurality of candidate points in each axis direction that move at 1/# rus from the current command position. Since the candidate point closest to is selected, pulse distribution that allows the target path to be followed more faithfully becomes possible.

[Brief explanation of the drawing]

1 and 2 are block diagrams showing an example of an apparatus for carrying out the method of the present invention, FIGS. 3 and 4 are diagrams used to explain the present invention in detail, and FIG. 5 5(b) is a diagram showing a preferred pulse distribution, and FIGS. 6 and 7 are used to explain the conventional pulse distribution method. This is a diagram. 1.21... Initial setting section, 2.22... D storage section,
3...2b memory section, 4...2 Hatake memory section, 5.27.
...X storage section, 6,28...Y storage section, 7,29...
・X end point storage section, 8, 30...Y end point storage section, 9, 3
1... Sign comparison section, 10.32... X processing/pulse generation section, 11.33... X processing/pulse generation section, 23
...DX storage section, 24...DY storage section, 25...
4D storage section, 26...4 & storage section. Applicant's agent Takahisa Kimura Figure 3

Claims (1)

[Claims] A pulse distribution method that distributes command pulses to each axis to instruct movement in each axis direction of a plurality of axes so as to move on a target path, the method comprising: When determining, the candidate point closest to the target path is found among multiple candidate points for each axis direction that are moved in pulses from the current command position, and the command pulse in the moving direction to advance to this candidate point is selected. A pulse distribution method characterized in that the current command position is updated to the position of the candidate point.