JPH0455061A - Method for recognizing shape of square joint - Google Patents

Abstract

PURPOSE:To stably and exactly detect even the shape of a square joint or the like by combining and computing the detection data of 1st and 2nd distance sensors which are respectively provided in the positions diametrically opposite to the front surfaces on one side of two sheets of plate materials constituting the joint. CONSTITUTION:The 1st distance sensor A is installed in the position diametrically opposite to at least the one surface of two sheets of the plate materials 1 and 2 constituting the square joint and the 2nd distance sensor B is installed to the position diametrically opposite to the other front surface. The coordinates of the points (a) and (b) at the ends of the plate materials which are diametrically opposite to each other can be detected by the two distance sensors A and B if the coordinates in the set positions of these two distance sensors A and B are previously measured by another means. The space coordinates of the resulted two points (a) and (b) are easily calculated from the coordinate of the set positions of the distance sensors A and B.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a method for recognizing the shape of a corner joint for optimally controlling welding conditions.

(Prior art) Conventionally, there are systems that recognize the shape of a weld joint during welding and perform welding by optimally controlling welding conditions such as torch position, filler supply amount, and welding current based on the obtained results, for example. JP-A-59-28608, JP-A-61-4
Proposals such as Publication No. 907 were made.

In particular, a typical system that attempts to recognize the shape of a joint in order to optimally control welding conditions is configured as shown in Figure 6. In Figure 0, 11 is the plate material to be measured. The light emitted from a light 1112 such as a semiconductor laser is input into the light emitting and scanning device 116 through the lens 14 at the overlapped joint of the plate material 11, and the light emitted from the light emitting and scanning mechanism 16 is controlled at a constant level by a vibrating mirror or the like. It is irradiated as light with a period of . Then, the reflected light passes through a light receiving lens 15 installed in a direction having a certain angle θ with respect to the light emitting axis.
The image was formed on a light receiving element 13 such as a photodiode play or a CCD linear array.

In this way, the distance to the object surface can be determined from the light projection angle and the light receiving position by simple geometric calculations (same principle as triangulation). In order to be able to perform measurements without being affected by the arc light during welding, measures have been taken such as using modulated laser light for the measurement and providing a narrow band filter in front of the light receiving element.

(Problem to be Solved by the Invention) However, in the above-mentioned method, the reflected light of the spot light emitted from the light emitting system is captured and measured by the light receiving system, so the surface shape, properties, color, etc. of the object to be measured are measured. In particular, in the case of a shape such as the square joint shown in FIG. 7, the amount of light received fluctuates extremely, making it difficult to perform stable measurements. In order to solve this problem, attempts have been made to increase the dynamic range of the light-receiving element and control the intensity of the emitting laser, but none of these methods have been able to reliably detect the shape of the corner joint.

The present invention was made in view of the above circumstances, and its purpose is to make it less susceptible to the influence of the surface shape, properties, color, etc. of the object to be measured, and to stably and accurately detect even shapes such as corner joints. In order to achieve the above object, the method for recognizing the shape of a corner joint according to the present invention is to When detecting the shape of a corner joint using a triangular distance measurement type scanning distance sensor equipped with a pair of related light emitting/receiving elements, the surface of at least one of the two plates that make up the joint is detected. The angle joint part It is characterized by recognizing the shape data of.

(Function) As described above, since the light emitting/receiving element is provided at a position directly facing one surface of one side of the corner joint, a stable amount of light can be obtained from the joint, and distance data can also be calculated accurately. Since such a light emitting/receiving element is directly opposed to one surface of the other side, the shape ms of the corner joint can be determined by calculating the distance data of both adjacent sides of the corner joint.

(Example) Examples of the present invention will be described below with reference to FIGS. 1 to 6.

Figure 1 shows the shape of the corner joint and the dimensions of the joint required for welding control. In Figure 0, 1 and 2 are plates, and these plates 1 and 2 are attached to the end face of one plate (here 1). The rear surfaces of the other plate members 2 are brought into close contact with each other and are combined at right angles to each other to form a corner portion. A small space is created at the corner of the plate material 1 and the plate material 2, resulting in a gap J. The distance between the ridgeline on the surface of plate 1 and the ridgeline on the surface of plate material 20 is referred to as the "interval", and the ridgeline on the end surface of plate material 2 that is closer to plate material 1 is referred to as "position".

Among these names, the dimensional data of "step" and "gap" are used to determine the welding current, filler supply amount, toner posture, etc., and the data of "gap" and "position" are used to determine the torch position. It is used to determine the

Figure 2 shows the arrangement position of the light emitting and receiving elements with respect to the corner joint. When placed in the α position above (direction directly facing the corner joint), it is difficult to obtain reflected light from the surfaces of plates 1 and 2 (shaded areas), so the shape data of the corner joint is difficult to obtain. Detection becomes difficult.

Therefore, the position T that directly faces the surface of one side of the plate material l that constitutes the corner joint, and the position β that directly faces the surface of one side of the plate material 2
This can be said to be the optimal position in terms of the amount of reflected light.

FIG. 3 shows the relationship between the amount of light received by the detection head and the calculated data Z in the case of a lap joint and a square joint. In the case of a shape like the lap joint shown in Figure 3 (al), if the direction perpendicular to the light irradiation direction is X and the light irradiation direction is Z, the surface of the object will be Since it is in a positional relationship directly facing the part, a stable amount of received light can be obtained over the entire scanning range.Therefore, the calculated distance data also corresponds to the shape.

In the direction facing the square joint described above in the corner joint shown in FIG.

In the corner joint shown in Fig. 3(C), the detection radar 3 is placed at a position directly facing the surface of either one side, that is, in the figure, a position directly facing the surface of the plate material 2 and including the joint part in its scanning area. By installing this, the reflected light from the surface of the board 1 disappears, and the reflected light amount data and distance data are shown in Figure 5 (C).
As shown in the figure on the right, a stable amount of light can be obtained from the joint, distance data can be calculated accurately, and shape data can be detected.

However, in reality 'l! As shown in Figure 44, 2.
There are problems such as secondary and tertiary reflections occurring, and the amount of reflected light cannot be obtained due to blind spots during projection, and reflected light amount data and distance data differ between stable and unstable regions as shown in the figure. . This stable region occurs when the surface of the plate material 2 is detected, and the unstable region occurs when the end of the surface is passed. Therefore, by detecting the point where the amount of received light becomes unstable, it is possible to detect the position of the end face of the directly facing plate, and the distance data at that position indicates the distance data of the end face of the plate.

From this point of view, as shown in FIG. A second distance sensor B is installed at the position. If the set position coordinates of these two distance sensors A and B are measured by another means, the coordinates of points a and b at the ends of the plates facing each other can be detected by the two distance sensors A and B. . The spatial coordinates of the two obtained points a and b can be easily calculated from the set position coordinates of the distance sensors A and B.

The shape data of the joint is calculated using these two Pisces [a and b. At this time, in order to obtain the gap dimension, it is only necessary to know the thickness of the target plate in advance. In the case of welding, the plate Since the thickness of the plate is often limited to the initially set thickness, the implementation of the present invention is not hindered by this.

(Effects of the Invention) As explained above, the method for recognizing the shape of a corner joint of the present invention is as follows:
When detecting the shape of a corner joint using a trigonometric scanning distance sensor equipped with a pair of light emitting and receiving elements in a fixed positional relationship, at least one of the two plates that make up the joint is detected. Detection data from a first distance sensor A provided at a position directly facing the surface of one side of the sensor and a second distance sensor B provided at a position directly facing the surface of the other side are combined and calculated. Since the shape data of corner joints can be recognized, it is possible to stably and accurately detect shapes such as corner joints without being affected by the surface shape, shape, color, etc. of the object to be measured. There are advantages that can be achieved.

[Brief explanation of drawings]

Figure 1 is a perspective view of the main parts of the corner joint (Figure 2 is a layout diagram of the distance sensor that detects the corner joint, Figure 3 is a diagram of reflected light amount data and distance data of the lap joint, Figure 3 is (b) is a graph of reflected light amount data and distance data at a detection position directly facing the corner of the corner joint, and Fig. 3 (C) is a graph of reflected light amount data and distance data at a detection position directly facing the corner of the corner joint. Figure 4 shows the details of Figure 3(C) and the reflected light quantity data and distance data. Figure 5 shows the arrangement of distance sensors A and B at opposite positions of the corner joint. , Fig. 6 is a layout diagram of shape detection in a conventional lap joint, and Figure 7 is a layout diagram of shape detection in a conventional corner joint. 1.2 Plate material A... First distance sensor B...Second distance sensor

Claims (1)

[Claims] When detecting the shape of a corner joint using a trigonometric scanning distance sensor equipped with a pair of light emitting and receiving elements in a fixed positional relationship, at least one of the two plates that make up the joint is detected. By combining and calculating the detection data of a first distance sensor A provided at a position directly facing the surface of one side and a second distance sensor B provided at a position directly facing the surface of the other side. , a method for recognizing the shape of a corner joint, characterized by recognizing shape data of the corner joint.