JPS6394152A - Control method for scanning of robot - Google Patents

Abstract

PURPOSE:To facilitate the input of scan control data by representing the end edges of a ruled surface by coordinates and making a surface scan on a scanned area when the end effector of a flaw detecting robot makes a scan on a work with the ruled surface. CONSTITUTION:The ultrasonic flaw detecting robot is made movable in three dimensional and the end effector of the robot makes the surface scan opposite the ruled surface of the work W(c). The ruled surface is defined between the 1st end edge E1 and the 2nd end edge E2 at both axial ends of the work W(c). Then, a coordinates system (X,Y,Z) is set on the base of the robot and the 1st end edge E1 and the 2nd end edge E2 of the work W(c) on the base are given coordinates. The end effector scans the circumferential surface axially from the end point P1 of the 1st end edge E1 and is fed at the 2nd end edge E2 by specific pitch in the circumferential direction to scan the circumferential surface thereafter. At this time, a controller computes coordinates of the route of end points P1..., Q1j, Q2j,..., P3. Thus, the respective end points are given coordinates to specify the pitch, so the input of control data is facilitated to make a scan properly.

Description

Detailed Description of the Invention (Field of Industrial Application) The present invention relates to a scanning control method for a robot used when scanning a scanned area of a workpiece using an end effector. The present invention relates to a scanning control method when the object has a curved shape.

(Prior art and its problems) In the yj''?1 process of aircraft, etc., ultrasonic flaw detection is performed for the purpose of discovering flaws existing on the surface or inside of parts such as honeycomb panels.In such cases ,
Since a honeycomb panel or the like has a planar shape, it is necessary to scan the entire scanned area thoroughly with an ultrasonic probe in all directions. Because of these circumstances, flaw detection requires a certain degree of skill, and conventionally, flaw detection scanning has been performed manually by workers.

In response, there are attempts to use robots to perform such tasks efficiently and uniformly.

However, since such attempts attempt to perform scanning using almost the same conventional robot control methods, various problems arise. For example, if a conventional teaching method for cutting robots and welding robots is applied as is, it is necessary to teach all U-paths for vertical and horizontal scanning in the area to be inspected. Therefore, in such a method, the teaching operation becomes complicated.

Furthermore, there is also the problem that if the teaching points are not selected correctly, the desired surface scanning cannot be performed.

These problems occur not only when robots are used for ultrasonic flaw detection, but also when robots are used for applications such as painting or thermal spraying, where a desired surface area must be scanned with an end effector. This has become a common problem.

OBJECTS OF THE INVENTION The present invention is intended to overcome the above-mentioned problems in the prior art, in which the scanned area is defined between first and second edges formed by a predetermined type of curve. It is an object of the present invention to provide a scanning control method for a robot in which data input is easy and surface scanning is performed appropriately when the surface is a line-woven curved surface (described later).

(Means for Achieving the Object) In order to achieve the above-mentioned object, the present invention provides for a workpiece whose scanned area is a linear curved surface, with the end effector of the robot facing the scanned area. As a scan control method for alternately repeating scanning and feeding of an end effector, thereby causing the end effector to scan the surface of the scanned area, (1) Give edge specific coordinates for specifying the shape and position of the The coordinates of the scanning end points are calculated and determined, and the scanning and feeding of the end effector is alternately repeated along a path obtained by sequentially connecting the scanning end points to scan the surface of the scanned area. provide a method.

Here, the term "line weaving surface" refers to a curved surface formed by a family of straight lines with a single parameter, as described in [Iwanami Mathematics Dictionary, 2nd Edition, JP, 448 (1968)].
1! Here, this term is defined as ``a curved surface whose edges are two curved lines, and which is formed by the locus of a line segment (generating line) connecting two points that move on each of the edges.'' An example of such a linear weave curved surface is shown in FIG. 6, which will be described later. In FIG. 6, E and E2 are the edges, and the line segment l is the generatrix. The reason why this invention targets such a linear curved surface is that since the scanning line is linear, the shape of the scanned area must also be commensurate with this. .

(Example) A. - Fig. 1 is a schematic perspective view showing the mechanical configuration of a Cartesian coordinate type ultrasonic flaw detection robot as an example of a robot scan-controlled according to an embodiment of the present invention. It is. In the figure, the ultrasonic flaw detection robot RB has a movable table 2 on a base 1 that is movable in the X direction (horizontal direction) by a motor M1 (not shown). Place a workpiece (not shown). A beam 4 is installed at the top of a column 3 vertically erected on both sides of the base 1, and a motor M is attached to the beam 4, which extends in seven directions (in the vertical direction in the figure).
A movable column 5 that is movable in the Y direction is designated by 2.

A motor M4 is provided at the lower end of the moving column 5, and is moved up and down in seven directions by a motor M3.

As a result, the arm 6 provided eccentrically from the central axis of the moving column 5 rotates in the α direction in the figure. Further, a motor M5 is provided on the side of the lower end of the arm 6, and this rotates the end effector 100 for ultrasonic detection in the β direction in the figure.

Among these, the end effector 100 includes a bifurcated tube 101, as shown in FIG.

End openings 102a and 102b of this bifurcated pipe 101 are bent in opposite directions, and one end opening 102a and 102b are bent in opposite directions.
An ultrasonic transmitter (probe) 103 is provided at the center of a. Further, an ultrasonic receiver 104 is provided in the center of the other end opening 102b. These transmitter 103 and receiver 104 are connected to the ultrasonic flaw detection device 7 shown in FIG. 1 by a tassel (not shown) passing through the inside of the bifurcated tube 101 and the inside of the guide vibe 8 shown in FIG. ing.

Further, this ultrasonic flaw detection device 7 has a built-in water supply device (not shown), and water from this water supply device is supplied to the piping inside the guide vibe 8 and the water supply hose 105a shown in FIG.
, 105b, and the gap 106a of the end openings 102a, 102b.

It is designed to eject from 106b. When performing ultrasonic flaw detection, as shown in FIG.
a and 107b are sprayed onto the surface of the workpiece W. In this state, the ultrasonic wave US is transmitted from the ultrasonic transmitter 103, and the workpiece W is
The ultrasound receiver 104 receives the transmitted ultrasound waves US.

In this way, flaws on the workpiece W are detected based on the time delay and intensity changes of the received ultrasonic waves, but since the principle of this ultrasonic flaw detection itself is well known, a detailed explanation thereof will be provided. Omitted.

Furthermore, the robot RB shown in FIG. 1 is provided with a control device 9 having a built-in microcomputer, etc., as well as an operation panel 10 and an external computer 11. Among these, the control device 9 performs various controls such as scanning control, which will be described later. Further, the operation panel 10 is provided with a keyboard, a display, and the like. Furthermore, the external computer 11 is for inputting various commands, displaying information, etc.

4 is a schematic diagram of the electrical configuration of the robot RB shown in FIG. 1. In FIG. 4, the following devices are connected to a microcomputer 21 built in the control device 9 via a bus BL.

■The above motors M1 to M5 and these motors M1 to M5
Mechanical layer vJ system 23 including encoders E1 to E5 (not shown in FIG. 1) that detect rotation angles of Principle Next, the principle of scanning control in this embodiment will be explained based on a specific example. FIG. 5 is a diagram showing the placement position of the workpiece W in the absolute coordinate system S (X, Y, Z) fixed to the base 1 of the robot 1 dog B. The surface shape of this workpiece W is a shape obtained by partially cutting out a cylinder along its axial direction γ. Such a shape is called an "arc cylinder" in this specification.

This arc nC has both end edges E1. in the axial direction γ.
It is a linear weave curved surface defined between E2. In other words,
Both end points of the line segment l shown in FIG. 6(a) are arc-shaped edges E,
By moving along E2, an arc cylinder C is formed as the locus of the line segment l. These edges E and E2 correspond to "first and second edges j" in this invention.

In addition to circular cylinders, such linear weave surfaces include cylinders (
There are various types such as a part of a truncated cone (Fig. 6(c)), a part of an elliptical cylinder (Fig. 6(d)), and a wave-shaped curved surface (Fig. 6(e)). The present invention is applicable to any of these linear weave curved surfaces.

However, the type of curve to represent the edge of the scanned area is instructed during teaching. The following explanation will be given by taking as an example the case where the area to be scanned is the circular arc tube C shown in FIG. 5, and the edge is expressed as a circular arc. This is because many of the curved surfaces of actual workpieces can be represented by circular arc cylinders.

Returning to FIG. 5, in this embodiment, first, the circular arc cylinder C
Consider a case where the entire surface of the area to be scanned is set as the area to be scanned, and while the end effector 100 is opposed to this area to be scanned, the surface scanning of the area to be scanned v4 A is performed.

As mentioned above, the teaching information regarding the shape and position of the workpiece W includes four points P in the absolute coordinate system S.
- The location information of P4 is given.

Among these points P-P, points P1 and P3 are both end points (edges) of the edge E1. Moreover, point P2 is
This point is located on the edge E1 and is different from the above-mentioned both end points P1°P. Furthermore, as the point P4, the other edge E2
One of the two endpoints (in this example, the endpoint on the point P3 side) is selected.

In this way, the position and shape of the edge E1 (radius and position of the cylindrical cutout) can be determined based on the position information of points P to P3.
is specified, and by adding the position information of point P4, the height and length (H) of this circular cylinder is specified. Considering that the edge E2 is obtained by moving the edge E1 in parallel, the shape and position of the edge E2 are also specified.

Note that instead of teaching the position coordinates of point P4,
Enter the height H of the arc nc numerically to determine the shape of the edge E2.
You can also specify the location. However, since the actual workpiece W is not always an ideal circular arc tube and often has slight deviations, it is preferable to directly teach the coordinates of the end point P4.

On the other hand, when teaching these points P1 to P4, posture information of the end effector 100 is also given. Therefore, after teaching, P・= (X,,Y,,Z,,0.,φi)(i=1~
4) ...(1) is known. this(
In equation 1), θ and φi express the attitude of the end effector 100 using Euler angles. Further, in the following, the position vector (X,,Y,,Z,'') will be written as 口i.

Note that the scanning speed is input at the time of teaching the point P1.

On the other hand, in this embodiment, among the points P-P4, two points P3. The direction parallel to the line segment connecting P4 is the scanning direction,
The circumferential direction of the arc tube C is the feeding direction. Then, specify the feed pitch ΔP in the circumferential direction from the outside, and P3. The area to be scanned A is scanned by alternately repeating scanning in the P4 direction and feeding in the circumferential direction. Such surface scanning movement is shown by arrow D1 in FIG.

In such surface scanning, the edge E shown in FIG.
Position and orientation information of the end effector that can be placed at each scanning end point on E2; Qlj, Q2j (J=0, 1.2, . . . ) is required. This information can be obtained as follows using the information on the teaching points P1 to P4 and the feed pitch ΔP.

First, the plane π (Fig. 8) passing through the three points P1 to P3 is obtained by imposing the condition that the plane π (Fig. 8) passes through the three points P1 to P3 on the general formula for the plane in the absolute coordinate system S: It is thereby determined by defining constants C to C3. The result is as follows according to the well-known Cramer's formula.

(Hereinafter, blank space) Next, the radius R of the circular arc nC is determined. This is the equation of a sphere centered at O', where the coordinates of the center 0' of a circle whose edge E1 is part of the circumference are 15 - (X, Yo, Zo): % formula %) It can be determined based on the condition that passes through points P1 to P3, and the condition of equation (5) that the center coordinate r3o of this sphere is on the plane π.

CX + C2Y□ + C3Zo= 1 - (5)
This system of equations can be solved approximately or numerically, thereby opening. and the value of R is determined.

Next, a local coordinate system S' fixed to the workpiece W and having r3o as its origin is defined. This local coordinate system S' (u
, C2, u), the C3 axis is taken in the height direction of the arc portion C (see Fig. 8). That is, if the unit vector in the direction of the u3 axis is written as 03, then U = (cj -1
53) ・(6). However, the symbol () indicates normalization of the vector, and if we take any vector as Ma, we get (Ma) Mima/1v1 Mari, Ll - (U Xl - Mouth )) ... (7
).

Furthermore, the C1 axis is taken in a direction perpendicular to both of the unit vectors σ and a. That is, Ul-cJ2xt]3
...(8). If the direction of UJ3 is completely perpendicular to the plane π, this unit vector a1 is the vector O'P1. In particular, when the length of PBr3 is long, more accurate expression is possible by doing the above.

Further, the above a2 is multiplied by the sign of C2.(mouth, -1°) (9), and the sign (direction) of C2 is adjusted accordingly. This means that the feed direction in surface scanning is set to point P
This is to set the direction from P to P3 via P. Therefore, as shown in FIG. 9, point P is the origin O'
If C2 exists in the opposite direction with respect to C2, then C2
0 becomes the new axis Peltor a2. However, in the following explanation, the case where the sign of equation (6) is positive as shown in FIG. 8 will be explained as an example.

The relationship between the local coordinate system S' and the absolute coordinate system S defined in this way is that the coordinate transformation matrix G is
When defined as Ll, U)' - (10), k=G <O-r3o) ・(11
). However, the mouth is an arbitrary position vector in the absolute coordinate system S, and k is a position vector in the local coordinate system S'.

Furthermore, when the above local coordinate system S' is expressed as a local cylindrical coordinate system with an axis in the a3 direction (see FIG. 10), k 1 =RCCO3θC k =RcsinθC 3-Zo · (12). However, k= (kl, 2. and 3) ... (1
3).

Therefore, equation (12) can be transformed into a variable of the local cylindrical coordinate system: (R
, θ, Zc) and distinguish the points PH (i=1 to 4) by the subscript ゛C. Then, the coordinates of the points P1 to P4 in the above local coordinate system are R(, H-X
+ + Y; θci-tan (Y H/X H)ZCi” Z
It is expressed as i-(14).

Next, the coordinates of scanning end points Q1j and Q2j (see FIG. 7) are determined. First, for a designated feed pitch ΔP, the feed angle bit Δθ in the cylindrical coordinate system is as follows: Δe−ΔP/R (15) (FIG. 11(a)). Furthermore, the locus F! in the local cylindrical coordinate system of the scan end point Q1j shown in FIG. 11(b), which is a partial diagram of FIG. 11(a), is (Rlj, elj, Zlj
) is R1j−R 01j=j・Δθ 1 −0 (j−0,1,2,...,N)
...(16)j. Furthermore, the coordinates of the scanning end point Q2j on the edge E2 side are 2j-R e2j-J Δe z2j-1r34-131 (j-o, 1, 2.
, N)...(11). However, N is the number of times of sending, which will be described later.

To express these in the absolute coordinate system S, (11)
, the following (18), (19) obtained from equation (12)
You can use the formula.

(12j-E5o+G k2j・(18)R,-(R
cose, R5ine Z,)IJ
1j Ij 1j lj' IJ
112 ・-(Rcose R5ine, Z
,)2J2j 2j°2j 2j
2J(j-o, 1,2....,N)...(
19) However, the position vectors of points Q1j and Q2J in the absolute coordinate system S are respectively written as α d .

1j and 2j Furthermore, the posture (θ, φ,) of the end effector 100 at these scanning end points Q1j and Q2j.

j1J (θ φ,) is 2j° 2J θ θ = θ 1j (θ −01)/NIj
2J 1 3...(20) φ1j=φ2j-φ1+j(φ3-φ1)/N(j=0
．． 1, 2. ..., N) ...(21). However, normally θ = θ and φ - φ3, and the above (20) and (21) are interpolation formulas that take into account teaching errors.

Next, the number of feeds (the number of scan divisions) required for surface scanning of the area to be scanned A is determined. The number of times N of feeding is determined by the scanning area Al7) P P shown as a developed diagram in FIG. 12.
(7) When the length R(e3-91') is L and DH),
(2) The final scan is completed on the vertex P4 side in FIG. 12, and (2) The smallest integer is selected such that the cumulative p value of the feed amount is greater than or equal to the above value. Among these, the condition (2) is a condition for making it easy for the end effector 100 to move to the next working position after the completion of scanning. Further, the condition (2) is a condition for performing scanning so as to cover the entire scanning area A. Therefore, for example, point P shown in FIGS. 12(a) and 12(b). is the end point of the entire scan.

The number of feeds N required to complete the entire scan at such a point can be expressed as follows, when the Gaussian symbol for integerization is written as [1.

■ When ([L/Δl)] + 1) is an even number.

N-[L/ΔP]+1...(22)■
(When [L/ΔP] +1 > is an odd number.

N-[L/ΔP1+2 (23) The reason why these equations are derived is as follows.

First, consider the case where L/ΔP-5,6 as shown in FIG. 12(a). Then, [L/ΔP]-5
To scan an area having a length 5.6 times ΔP in the feed direction, [L/ΔP] −1−1−6 times (an even number of times) of feed is required.

Considering the relationship that when scanning is performed before and after the feed, the number of scans is one more than the number of feeds, seven scans are performed in conjunction with these six feeds. This number of 7 times is an odd number, and the end point P of all scanning is the vertex P4.
The condition e of being on the side is met. Therefore, by formula (22), the number of scans N
All you have to do is determine.

Next, consider the case where L/ΔP-4,7 as shown in FIG. 12(b). Then, [L/ΔP] −4, and 4.
To scan an interval of 7ΔP, [shi/ΔP]−←1=5
(odd number of times) is required. The number of scans associated with the five feedings is six.

However, in six scans, all the scans end on the vertex P3 side. Therefore, in order to complete the entire scan on the vertex P4 side, one additional feed is added, resulting in [L/ΔP]+2−6 scans.

In this way, equations (22) and (23) were obtained, and when these equations (22) and (23) are put together, they can be written as the following equation (24).

N-2[([L/ΔP]+2>/2]...(24) Prove that this equation (24) is equivalent to equations (22) and (23) as follows. can.

First, when ([L/ΔP]+1) is an even number, m is 0
Alternatively, as a positive integer, it can be written as [L/ΔP]+1-2m, but if this is substituted into equation (24), N-2[
(2m+1)/2] 2m - [L/ΔP]+1, which agrees with equation (22).

Also, when ([L/ΔP]+1) is an odd number, it can be written as [L/ΔP]+1-2m+1, but when this is substituted into equation (24), N-2[
(2m+2)/2] = 2m+2 - [L/ΔP]+2, which agrees with equation (23).

In this way, it can be seen that equation (24) is equivalent to equations (22) and (23).

By the way, when the length is a multiple of the feed pitch ΔP, L/ΔP - [L/ΔP], and the entire scanned area A can be covered by [L/ΔP] feeds. . Therefore, in this case, you can specifically do the following:

■ (L/ΔP) is an odd number, That is, when (L/ΔP+1) is an even number.

N-(L/ΔP)+1...(25)■
When (L/ΔP) is an even number, that is, (L/ΔP+1) is an odd number.

N-(L/ΔP)...(26) This means that L is 5ΔP and
It can be easily understood by looking at it as 4ΔP. However, (22
), (23) and (25), (26) are different only in that the number of feeds is two when (L/ΔP) is an even number. Therefore, in practice, (22), (
The number of times of feeding N may be determined using equation (23) or the equivalent equation (24). Note that the number of scans M is M=N+
1 is given by.

Based on this principle, data necessary for surface scanning of the scanned area A can be obtained from the position/orientation information of the point P-P4 and the information regarding the feed and the pitch ΔP.

Next, the control principle when the area to be scanned is a part of the curved surface of the workpiece W will be explained. FIG. 13 is a diagram illustrating such a case, in which a rectangular curved area A' is assumed to be set on the side surface of the arc cylinder C. This scanned area A' is formed as follows, and its apex is F
1 to F4. The scanning direction is parallel to the sides F1F3, and the feeding direction is the circumferential direction of the arc tube C.

Therefore, if the position/orientation information of the end effector 100 that pulls these vertices F1 to F4 is known, the data regarding the scanning path etc.
, p and p respectively as Fl.

It can be obtained by changing the readings to F and F. The feeding direction can be known from the information of P2.

Information regarding this vertex F-F4 can be given by direct teaching, but another method is used here. That is, in this embodiment, the scanned area A
Cylindrical coordinate data ea, Za specifying the scanning start position P8 (Fl) at ', and data Δe8. Δza is given. As a result, the scanned area A' is determined as follows: e ≦OA≦(ea+Δea) Z ≦ZA≦(Za+ΔZa) RA-R...(26) From this, it is possible to know the coordinates of the vertex F-F4. can. The coordinates of the scanning end points QIJ and Q2j can be determined based on the coordinates of the vertices F1 to F4, as in the case of full-plane scanning. The same applies to the number of times of sending N.

D. Control Operation Based on the above principles, the operation for controlling the robot shown in FIGS. 1 to 4 will be explained with reference to the flowchart shown in FIG. 14.

First, in step S1 in FIG. 14, teaching is performed. Of this teaching operation, the teaching order for the area scanning portion is shown in FIG. That is, in the figure, the positions indicated by white circles are the teaching points, and the numbers with the symbol T indicate the teaching order.

In addition, the scanned area A has the shape of a circular arc tube, and the edge E
, E2 are circular arcs. Furthermore, when teaching about the four points P-P4 that define the scanning area, "PLS" in the figure
As shown in , data indicating that this is a point that specifies the scanning area A is also input.

In the next step $2 in FIG. 14, the feed pitch ΔP is input from the operation panel 10 or the keyboard of the external computer 1.
Enter the value and store it in memory (not shown). Moreover, when the area to be actually scanned is only a part of the work W, θ, z lΔ in FIG.
The values a a a of e and ΔZ are also input. In the next step S3, it is determined whether the area A' (that is, a part of the flat surface of the workpiece W) has been designated as the scanned area, and if so, the process moves to step S4, and the positions related to the vertices F1 to F4 are determined. The posture and the like are determined, and the position and size of the scanned area A' are determined accordingly. However, if the input value of θa'''a'Δe8.ΔZa is not appropriate and the scanned area A' protrudes from the surface of the work W, this "protruding part" is removed and the scanned area 1aA' is determined. Then, in step S5, the number of feeds N (or the number of scans M) is calculated based on equations (22), (23), etc.

In the subsequent step S6, the operator uses a mode switch and a start button (both not shown) to start the playback operation.

To explain only the part related to surface scanning in this reproduction operation, first, in step S7, the positions of the scanning end points Q1j and Q2j for the next scanning line and the posture of the end effector 1oo in these are determined by (18) ~(21
) Calculate based on the formula etc. And these end points Q1
An interpolation calculation is performed between j and Q2j (step 88). Of these, linear interpolation is performed for scanning in the P and P4 directions, and circular interpolation is performed for sending in the circumferential direction.

When the position and posture to which the end effector 100 should move are determined in this way, the position and posture of the end effector 100 are controlled in the next step S9, and P1P3
Scanning is executed in the (F1F3) direction, and flaw detection data obtained by ultrasonic flaw detection is captured. When one scan is completed, it is determined in step 310 whether it was the last scan line.

If it is not the final scanning line, in step S11,
Feed by ΔP in the circumferential direction, and then proceed to the next step 81
After updating the scanning line in step S2, the process returns to step S7. As a result, the ultrasonic flaw detection progresses while scanning and feeding are performed alternately.

Then, when the feeding is performed the number of times N, step S1
0 becomes 'YES', and surface scanning of the area to be scanned A (or A') is completed.

The trajectory of the end effector 100 based on such a routine is shown by a thick line in FIG. Of these, Figure 16(a)
This is a case of scanning the entire plane of the workpiece W. Further, FIG. 6(b) shows the case of partial scanning.

Although not shown in the flowchart of FIG. 14, in the case of partial scanning, as shown in FIG. 16(b), P1→B
The end effector 100 is caused to pass through the path 1→F1→(plane scanning)→B2→P4. However, point B1 is on side F
Point B2 is the point where the extension line of 1F3 intersects edge E1, and point B2 is the point where the final scanning line intersects edge E2.

E. Modifications Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and for example, the following modifications are possible.

■ In the above embodiment, a case was considered in which the entire workpiece W was curved, but as shown in FIG. 17, the workpiece W itself includes a flat portion 51, 52 and a curved portion 53. , it is also possible to perform the surface scanning of the present invention targeting only the curved surface portion 53. In addition, the flat part 51.52
A surface scanning method for this invention has been developed by the inventor of this invention, and a patent application has been filed as a separate invention.

(2) The method of giving edge specific coordinates may be selected as appropriate depending on the type of curve of the edge. For example, if the edge is an ellipse, coordinate points for specifying the eccentricity may be added. However, in many cases, the linear weave curved surface can be approximately represented by a circular arc cylinder or a combination of cylinders, so it can be handled by dividing it into circular cylinders or the like. The scanned area may be approximately a linear curved surface. Further, in addition to , numerical data may be added to specify the position and shape of the edge. It is also possible to numerically input the edge specific coordinates themselves.

(2) The mechanical configuration of the robot is not particularly limited, and robots with various degrees of freedom can be used. Furthermore, the present invention is applicable not only to the control of ultrasonic flaw detection robots, but also to the control of all robots that require surface scanning, such as painting robots and thermal spraying robots.

(Effects of the Invention) As described above, according to the present invention, when the coordinates for specifying the edge of the scanned area are given and the feed pitch is specified for the linear weave curved surface, surface scanning of the scanned area is performed. Therefore, it is possible to obtain a robot scanning control method in which data input is easy and the area to be scanned can be properly scanned.

[Brief explanation of the drawing]

FIG. 1 is a schematic perspective view of a robot suitable for applying an embodiment of the present invention, FIGS. 2 and 3 are explanatory diagrams of an end effector in the embodiment, and FIG. 4 is an electrical diagram of the robot of FIG. 1. 5 is an explanatory diagram of the shape of a workpiece in an embodiment of the present invention; FIG. 6 is a diagram illustrating an example of a linear weaving surface; FIGS. 7 to 12 are explanatory diagrams of full-surface scanning. , FIG. 13 is an explanatory diagram of partial scanning, FIG. 14 is a flowchart showing the operation of the embodiment, and FIG.
16 is a diagram illustrating the trajectory of the end effector according to the embodiment. FIG. 17 is a detailed diagram illustrating the present invention. RB...Ultrasonic flaw detection robot, 7...Ultrasonic flaw detection device, 9...Control boar neck,
100... End effector, W... Work, A
, A'... scanned area

Claims (3)

[Claims] (1) For a workpiece whose scanned area is a linear curved surface defined between first and second edges formed by a predetermined type of curve, the end effector of the robot is opposed to the scanned area. A scan control method for alternately repeating scanning and feeding of the end effector in a state in which the end effector scans the area to be scanned by the end effector, the method comprising: Provide edge specific coordinates for specifying the shape and position of the edge, and each set on the first and second edges according to the edge specific coordinates and a specified feed pitch. The coordinates of scanning end points are calculated and determined, and scanning and feeding of the end effector are alternately repeated along a path obtained by sequentially connecting the scanning end points to scan the surface of the scanned area. Characterized by
Robot scanning control method. (2) The curved surface is an arc tube corresponding to a shape obtained by partially cutting a cylinder in the axial direction, and the first and second edges are both end edges of the arc tube in the axial direction. , the edge specific coordinates are (a) the coordinates of both end points of the first edge, (b) the coordinates of a point on the first edge and different from the both end points, and ( 2. The robot scanning control method according to claim 1, further comprising: c) coordinates of one end point of said second edge. (3) The area to be scanned is a part of the curved surface of the workpiece, and a first set of edge specific coordinates specifying both edges of the curved surface is given by teaching, and the area to be scanned is a part of the curved surface of the workpiece. A second set of edge specific coordinates that specify the first and second edges include: (a) the first set of edge specific coordinates; and (b) coordinates fixed to the curved surface portion. The scanning control method for a robot according to claim 1, wherein the scanning control method for a robot is given based on positional information of the scanned area specified in a system.