JPS6411396B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for teaching a weaving pattern in a welding robot.

[Conventional technology]

Conventionally, a so-called software weaving method has been put into practical use, in which weaving motion is performed as a resultant motion of the drive shaft of a welding robot.
There was a constraint that the teaching of the weaving pattern had to be done at the start of the weld line.

[Problem to be solved by the invention]

When weaving welding the shielded interior of a welded structure using this conventional method, even though the welding torch and its tip can enter, it is difficult for the operator to visually see the torch tip. The disadvantage was that the teaching work of determining the direction of the weaving pattern, the amplitude of the weaving, etc. was extremely difficult.

Even under such circumstances, the present invention provides a weaving system that allows the teaching of weaving patterns to be carried out at any location with good workability, and that allows the amplitude of the actually taught weaving pattern to be easily changed. The purpose is to provide a teaching method.

[Means for resolving assignments]

To achieve this objective, the weaving teaching method of the present invention teaches a plurality of points for a weaving pattern on an arbitrary space away from the actual welding line, and Translate each teaching point of the weaving pattern in parallel by a line segment connecting one point and any taught welding start point or any one point in its vicinity, and correspond to the specified amplitude amount. The method is characterized in that a weaving pattern is generated in which a line segment connecting each point is enlarged or reduced by a ratio of the length between two points of the weaving pattern.

〔Example〕

Hereinafter, the present invention will be specifically explained based on Examples.

FIGS. 1 and 2 are explanatory diagrams showing the relationship between the teaching pattern and the actual welding line according to the present invention.

Point Q ₁ in Figure 1 is the starting point of the welding line, and Q ₂ is the ending point, which are assumed to have been taught in advance.
(In this specification, the welding start point is defined as each teaching point of the taught welding line.) On the other hand, P ₁ , P ₂ , and P ₃ constitute a weaving pattern. This is a teaching point and is taught in a space completely different from the welding line.

Now, the orthogonal coordinates of P ₁ point, P ₂ point, P ₃ point are P ₁
(X ₁ Y ₁ Z ₁ ), P ₂ (X ₂ Y ₂ Z ₂ ), P ₃ (X ₃ Y ₃ Z ₃ ), and Q ₁
If each point is shifted in parallel by the same amount and in the same direction as the line segment _{1 1} connecting the point and the P ₁ point, the P ₁ ′, P ₂ ′, and P ₃ ′ points are realized.

Of course, the P ₁ ' point is the same as the Q ₁ point. For convenience of explanation, we assume that the weaving pattern consists of three points, but as long as there are multiple points, the number of three points moving in parallel is limited as long as it is necessary for the welding finish. There isn't. Also, there is no problem whether the point to be moved to _Q1 point is _P1 point, _P2 point, or _P3 point.

Now, set the execution direction of the weaving pattern.
Assume that the direction is P ₁ ′ point → P ₂ ′ point → P ₃ ′ point → P ₁ ′ point.
At this time, each coordinate axis component of _{1 2} ΔX ₂₁ , ΔY ₂₁ ,
ΔZ ₂₁ becomes ΔX ₂₁ =X ₂ −X ₁ ΔY ₂₁ =Y ₂ −Y ₁ ΔZ ₂₁ =Z ₂ −Z ₁ (1). Further, the length D ₂₁ of the line segment _{1 2} is D ₂₁ =√ ₂₁ ² + ₂₁ ² + ₂₁ ² …(2).

Prior to the actual weaving, some input method, e.g. a key switch (not shown) is required.
Assuming that a length A ₂₁ from P ₁ ′ is specified on ₁ ′ ₂ ′ parallel to _{1 2} as the weaving amplitude (point P ₂ ″ in Figure 1),
P ₂ ″(X ₂ ″Y ₂ ″Z ₂ ″) is on the line or extension of line segment ₁ ′ ₂ ′. Each coordinate component of this P ₂ ″ point (X ₂ ″Y ₂ ″Z ₂ ″)
is expressed by equation (3).

X ₂ ″＝X〓 ₁ +ΔX ₂₁ ×A ₂₁ /D ₂₁ Y ₂ ″＝Y〓 ₁ +ΔY ₂₁ ×A ₂₁ /D ₂₁ Z ₂ ″＝Z〓 ₁ +ΔZ ₂₁ ×A ₂₁ /D ₂₁ …(3) Here where P ₂ ″(X ₂ ″Y ₂ ″Z ₂ ″) is the coordinate of the final point in the ₁ ′ ₂ ′ direction, and (X〓 ₁ Y〓 ₁ Z〓 ₁ ) is the orthogonal coordinate data of the Q ₁ point. , P ₁ ′ point, and P ₁ ″ point coordinate data. Let v ₂₁ be the velocity coefficient from P ₁ ″ to P ₂ ″ point.

This v ₂₁ is the weaving frequency set with a key switch etc. as well as the weaving amplitude.
Alternatively, it is determined by the time to move between weaving amplitudes and the weaving amplitude.

The current value (x ₀ y ₀ z ₀ ) of the weaving trajectory moment by moment is as follows.

_{x 0 = _ _ _ _ _ _ _ _ _ _ _ _ _ _ 21} ...(4) It is assumed that v ₂₁ changes uniformly from 0 to 1 over time.

Therefore, when v ₂₁ = _{0 ,} x ₀ = X〓 ₁ y ₀ = Y〓 ₁ z ₀ = Z〓 ₁ , and when v ₂₁ = 1, _{x 0 = 2} ″.

When the current value reaches the P ₂ ″ point, the weaving direction must be changed. The direction is the line segment
The direction is ₂ ″ ₃ ″ parallel to P ₂ ′ P ₃ ′. The cosine of each direction (a ₂ , b ₂ , c ₂ ) is expressed as a ₂ =ΔX ₃₂ /D ₃₂ b ₂ =ΔY ₃₂ /D ₃₂ c ₂ =ΔZ ₃₂ /D ₃₂ (5).

ΔX ₃₂ , Y ₃₂ , Z ₃₂ is calculated by using _{X 3} , Y 3 , Z ₃ instead of X 2 , Y ₂ , Z ₂ and X ₂ instead of X ₁ , Y ₁ , Z ₁ on the right side of equation ( ₁ ). ，
Find by substituting Y ₂ and Z ₂ respectively, and D ₃₂ is (2)
Instead of ΔX ₂₁ , ΔY ₂₁ , ΔZ ₂₁ on the right side of the equation, ΔX ₃₂ ,
Find by substituting ΔY ₃₂ and ΔZ ₃₂ .

The length A ₃₂ of ₂ ″ ₃ ″ is A ₃₂ = A ₂₁ ×D ₃₂ /D ₂₁ (6). Therefore, the current values x ₀ , y ₀ , on line segment ₂ ″ ₃ ″
z ₀ is expressed by the following equation (7).

_{x 0 = _ _ _ _ _ _ _ _ _ _ _ _ _ _} ΔY ₃₂ ×v ₃₂ /D ₂₁ z ₀ =Z ₂ ″＋A ₃₂ ×c ₂ ×v ₃₂ =Z ₂ ″＋A ₂₁ ×ΔZ ₃₂ ×v ₃₂ /D ₂₁ …(7) Here, v ₃₂ is P ₂ ″ Let it be the velocity coefficient from point P ₃ ″ that starts from v ₃₂ =0 and increases uniformly over time until v ₃₂ =1.

_{When v 32 = 0 , x 0 = _ _} +A ₂₁ ×ΔX ₃₂ /D ₂₁ =X ₃ ″ y ₀ =Y ₂ ″＋A ₂₁ ×ΔY ₃₂ /D ₂₁ =Y ₃ ″z ₀ =Z ₂ ″＋A ₂₁ ×ΔZ ₃₂ /D ₂₁ =Z ₃ ″ . X ₃ ″, Y ₃ ″, and Z ₃ ″ are the coordinates of the P ₃ ″ point.

It is clear that the vector from point P ₃ '' to point P ₁ '' and the current value can be obtained in exactly the same manner as described above. That is, it is determined by the equation (8) below.

_{x 0 = _ _ _ _ _ _ _ _ _ _ _ _ _ _ 21} ...(8) Here, ΔX ₁₃ , ΔY ₁₃ , ΔZ ₁₃ are the right side of equation (1) X ₂ ,
These are the values obtained by substituting X ₁ , Y ₁ , Z ₁ in place of Y ₂ , Z ₂ , and substituting X ₃ , Y ₃ , Z ₃ in place of X ₁ , Y ₁ , Z ₁ .

v ₁₃ has the same properties as v ₂₁ and v ₃₂ .

_{When v 13} = _{1 , x 0 =} X ₃ ″ ₊ A ₂₁ × _{ΔX 13} / _{D 21 =} ΔZ ₁₃ /D ₂₁ =Z ₁ .

Now, the coordinates of the starting point of the weaving motion are
Although X ₁ , Y ₁ , and Z ₁ are used, in actual weaving motion, the origin of the weaving motion is set at point Q ₁ ,
Taking the current value of the weaving motion as a relative position from this point, the current value x ₀ when v ₂₁ = 0, v ₁₃ = 1 and the first term on the right side of equation (3) and (4) ，
y ₀ and z ₀ may be read as 0.

Now, the explanation so far has been about a method in which one of the weaving pattern teaching points coincides with the welding start point and that point is used as the origin of the weaving motion.

FIG. 2 is an explanatory diagram of a method in which the origin is set at the midpoint of one side forming a weaving pattern.

The difference from Figure 1 is that the Q ₁ point does not coincide with the P ₁ ′ and P ₁ ″ points, and the Q ₁ point is at the midpoint on the line segment ₁ ′ ₂ ′.In this case, P The coordinates of the ₂ ″ point X ₂ ″, Y ₂ ″, Z ₂ ″ are
It can be easily obtained by substituting A ₂₁ /2 in place of A ₂₁ in equation (3). The coordinates of point P ₃ ″ are X ₃ ″, Y ₃ ″, and Z ₃ ″.
In equation (7), the coordinates of point P ₁ ″, X ₁ ″, Y ₁ ″, and Z ₁ ″ are
In equation (8), it can be found by setting v ₃₂ =v ₁₃ =1, respectively. The current value on line segment ₁ ″ ₂ ″ is the first term on the right side of equation (4)
X〓 ₁ Y〓 ₁ Z〓 It can be found by substituting X″Y″Z″ in place of ₁ .

The processing when the origin of the weaving pattern is set at point _Q1 is as described above. Furthermore, when the origin of the weaving pattern is near the _Q1 point, the bias amount ΔxΔyΔz is substituted for X〓 ₁ Y〓 ₁ Z〓 ₁ in equations (3) and (4). In the case of simple harmonic motion, start by substituting A ₂₁ /2 in place of A ₂₁ in equation (4), and when P ₂ ″ is reached, A ₂₁ in equation (7)
_{Substitute −A 21 in place of _ _ _ _} Substitute ″Y″Z″
Go to P ₁ ″. Furthermore, from P ₁ ′ to P ₂ ″, equation (4) is executed. The above is an explanation of simple harmonic motion.

Next, a specific circuit example is shown in FIG.

In FIG. 3, MPX1 to MPX8 are digital multiplexers that select and output one of a plurality of groups of input data. LATCH9 and LATCH10 are digital memories, and CTR11 and CTR12 are digital counters that count up or count down every clock signal. SUBs 13 to 16 are digital subtracters that perform subtraction or sign inversion operations. FIFO1, FIFO2 and FIFO3 are
In the FIRSTIN FIRST OUT memory, all data is shifted to the adjacent cell on the right every synchronization signal.

The control circuit 33 is a part that generates signals for controlling the entire circuit, and is composed of a logic circuit group and a flip-flop group.

D/A17-20 are converters that convert digital data into analog quantities, such as voltage quantities, MPY21
27 is an analog multiplier, DVD 28 is an analog divider, and ROOT 29 is a circuit for calculating an analog square root. S/Hs 30 to 33 are analog memories, and ADDs 34 to 37 are analog adders.

Examples of analog adders, dividers, and square root circuits are shown in the fourth example.
As shown in the figure. A ₁ and A ₂ are logarithmic conversion circuits that utilize the voltage and current characteristics of semiconductors, and A ₄ is an inverse logarithmic conversion circuit.

FIG. 4 shows that three functions can be realized using the same circuit example by switching the terminals.

When the toggle switch TSI is set to M, the multiplier becomes Eout = Ex × Ey, when it is set to D, Eout = Ex/Ey, and when it is set to R, Eout = √. Here, Ex, Ey, Ez are analog inputs,
Eout means analog output.

For example, for multiplication, e _x = KlnEx, e _y =
KlnEy, and the output of adder A ₃ is A ₃ = KlnEx +
KlnEy=KlnEx×Ey. Also A ₃ = KlnEout
Therefore, the calculation results Eout=Ex and Ey are obtained. Other operations can also be easily realized by logarithmization.

Therefore, it can be considered that the corresponding circuit example is applied to FIG.

In Fig. 3, DATABUS is a processor in a welding robot control system, a program storage memory, a data storage memory, and I/Os such as an operation panel and a teachbox (all not shown).
are commonly connected data transfer lines.

For the data transferred via this data bus, one of amplitudes A, -A, and A/2 is selected by the multiplexer 7 in accordance with the ASEL signal from the control circuit 38, and is input to the counter 11. The amplitude counter 11 starts from 0 and counts up every clock ck determined by the speed rates v ₂₁ , v ₃₂ , v ₁₃ stored in the FIFO 3, and when it becomes equal to the counter input, outputs a FULL signal to the control circuit 38. At the same time, it resets itself and receives the next clock signal.
Wait for CK. This counter output is effectively the (4th)
It has a function of executing the formula A ₂₁ ×v ₂₁ , the formula (7) A ₂₁ ×v ₃₂ , and the formula (8) A ₂₁ ×v ₁₃ .

The output of the counter 11 is converted into an analog quantity by a D/A converter 20. The divider divides this data A and the output _D21 of the analog memory 30 to obtain A÷
Output D ₂₁ . This output is further output by the square root circuit 29
Each line segment from is multiplied by the corresponding data Dn,
The value of A×Dn÷D ₂₁ is MPY2
The output commands are multiplied by Δx, Δy, and Δz by 5 to 27, and output commands are x ₀ , y ₀ , and z ₀ via adders 35 to 37 .

The control circuit 38 outputs a FIFOck signal for each FULL signal of the amplitude counter 11. The FIFOck signal is
This is a signal that shifts the FIFO1, FIFO2, and FIFO3 signals to the right by one cell, and in FIFO1, X ₂ , Y ₂ ,
Z ₂ to X ₁ , Y ₁ , Z ₁ cells, X ₃ , Y ₃ , Z ₃ to X ₂ ,
X ₁ , Y 1 , and Z 1 are stored in the Y 2 and Z ₂ cells, and X ₁ , Y ₁ , and Z ₁ are stored in the X ₃ , Y ₃ , and Z ₃ cells via a multiplexer, respectively.
The same applies to FIFO2 and FIFO3.

In this way, the data in FIFO1, FIFO2, and FIFO3 is shifted cyclically, and the data does not volatilize.

The coordinate data of FIFO1 and FIFO2 is from SUB13
15, squared by MPY21-23, added by ADD34, and further ROOT29
Equation (2) is calculated by: The S/H2 signal is driven by the teaching point arrival signal FULL and is sent to the adder 35.
-37 outputs are taken into S/Hs 31-33.

In this case, S/H31 to 33 have a two-stage configuration, and after the sum of the S/H output, MPY25 to 27 output, and each axis component of the bias amount is stored in the first stage,
It shall be transferred to the output stage. That S/H31
The outputs of ~33 are held.

The S/Hs 31 to 33 also have a reset function, and the control circuit 38 outputs a reset signal S/H at the time of starting weaving, passing the origin, or passing the final point of the weaving pattern (point P1 _' ').
It is assumed that H2 is also output.

x ₀ , y ₀ , and z ₀ are turned ON/OFF by relay RLY. This ON/OFF is controlled by a signal (not shown) from a control circuit.

FIG. 5 is a schematic explanatory diagram of the servo circuit.

The command pulse is outputted from a pulse distribution circuit (not shown), and the ₁₂ pulses shown in FIGS. ₁ and 2 are given. The PG signal is a position feedback signal from a rotary encoder attached to the end of the motor shaft. The CTR uses a deviation counter to count the deviation between the command pulse and the PG pulse.

In this way, the deviation counter output becomes a position command from point _Q1 to point _Q2 and becomes an analog signal via the D/A converter.

Each axis component of position command and weaving
The sum of the pattern axis components x ₀ , y ₀ , and z ₀ becomes the input to the servo controller.

The servo controller drives the X, Y, and Z axis motors. The rotational motion of the motor is converted into linear motion via a ball screw, etc., which moves each axis of the welding robot.

Although specific examples of the present invention have been described above, the circuit example shown in FIG. 3 does not impair the purpose of the present invention even if it is an all-digital circuit example, and the same applies even if it is implemented using some kind of software. .

In addition, in the above explanation, an example using a Cartesian coordinate system robot was given, but even if the robot has other coordinate systems, such as cylindrical coordinates, polar coordinates, and joint coordinates, it is possible to change from each coordinate system to the Cartesian coordinate system. If the robot is capable of conversion, it can be assumed that the converted data is stored in FIFO1 and FIFO2.

Further, x ₀ , y ₀ , z ₀ in FIG. 3 are converted into digital quantities by an A/D converter, and then converted from the orthogonal coordinate system to the corresponding coordinate system, and the data is converted into D/D.
It is converted into an analog quantity by the A converter and becomes the figure shown in Figure 5.
It can be considered that x ₀ , y ₀ , z ₀ . It goes without saying that the X-axis motor, Y-axis motor, and Z-axis motor in FIG. 5 may be read as each arm name axis motor of the corresponding coordinate system without losing rationality.

〔Effect of the invention〕

As described above, in the present invention, the taught weaving pattern is translated in parallel onto the welding line, and the ratio between the specified amplitude amount and the corresponding length between two points of the weaving pattern is determined. However, a weaving pattern is generated by enlarging or reducing the line segments connecting each point. This makes it possible to teach weaving patterns in any space away from the welding line, and greatly improves the workability of weaving teaching. Therefore, the present invention greatly contributes to the automation of welding.

[Brief explanation of the drawing]

1 and 2 are explanatory diagrams showing the relationship between the teaching pattern and the actual welding line according to the present invention, FIG. 3 is a circuit diagram showing an embodiment for carrying out the present invention, and FIG. 4 is an analog A circuit diagram showing an example of the configuration of an arithmetic circuit,
FIG. 5 is a schematic explanatory diagram of the servo area of the welding robot.

Claims (1)

[Claims] 1. Teach multiple points for weaving pattern in arbitrary space away from the actual welding line,
Translate each teaching point of the weaving pattern by a line segment connecting any one point of the weaving pattern and any taught welding start point or any one point near it, and then specify The welding robot is characterized in that the welding robot generates a weaving pattern in which a line segment connecting each point is enlarged or reduced by a ratio between the amplitude amount and the corresponding length between two points of the weaving pattern. Weaving teaching method.