JP3082829B2 - Automatic welding line recognition method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for recognizing the position of a welding member in a non-contact manner with high accuracy and high speed in welding by a welding robot.

[0002]

2. Description of the Related Art Conventionally, as an example of a welding method of a welding member using a welding robot, a member to be welded is accurately set and placed on a base material, and the welding robot is operated as taught to perform welding. How to do is commonly known.

As another method, there is a wire touch sensor method (currently mainstream) for correcting the welding start point and end point taught in advance during welding. In this method, a potential is applied between the welding wire and the member, and the potential is generated by touching the welding wire to the member. This is a method of detecting the position at the moment of touch and correcting (recognizing) this as a member position.

[0004]

However, in the former method in the above-mentioned prior art, the welded members are rarely machined, and most of them are gas-processed or cast products. In most cases, members are temporarily assembled by welding.

In this case, it has been impossible to accurately and accurately set the member at the same position as the teaching point.
Therefore, welding is performed with a certain degree of error, but the welding line is displaced, resulting in a defect and requiring repair.

In the latter case, the use of a welding wire causes bending due to the wire habit and the direction is not constant, so that an error always occurs.

[0007] Furthermore, if the sensing is performed at a high speed, the welding wire is bent, and the sensing speed cannot be increased so much.

Further, if there is an insulator such as scale or slag on the member surface, there is a problem that a potential is not generated there and a touch error occurs.

According to the present invention, a new automatic welding line recognition method for recognizing a position of a member in a non-contact manner as a sensing method instead of a touch sensor method in order to solve the above-mentioned disadvantages of the prior art and an apparatus used therefor It is intended to provide.

[0010]

In order to achieve the above object, a method for automatically recognizing a welding line according to claim 1 of the present invention provides a method for automatically welding a member by a welding device such as a welding robot.
A CCD camera for imaging a member and an image processing device for inputting and processing this image use a pattern matching method (normalized correlation method) using images obtained by imaging the member from front and side directions. 3 misalignment
It is characterized in that the welding is detected dimensionally and the welding start point and end point taught in advance are corrected.

As a device for use in this method, a position-adjustable welding robot having a CCD camera for picking up a member at the tip and a welding torch attached thereto is connected to a welding robot control device. It is preferable that an automatic welding line recognition device is connected to an image processing device in which the control device processes an image captured by the CCD camera and stores the image as a registered image in two directions, front and side.

Further, in the automatic welding recognition method according to the third aspect of the present invention, when a member is automatically welded by a welding device such as a welding robot, a CCD camera for imaging the member and an image for inputting and processing the image. The processing unit detects the displacement of the member mounting position on the XY plane by the pattern matching method (normalized correlation method) in the image in which the laser slit light is not seen from the front, and further, the image is irradiated with the laser slit light from the same front. The positional deviation of the slit light is subjected to pattern matching to detect the mounting positional deviation of the member in the Z direction, a three-dimensional positional deviation is obtained, and the welding start and end points taught in advance are corrected and welding is performed. It is characterized by:

As an apparatus used in this method, a CCD camera for picking up a member and a laser slit optical sensor for recognizing the position of the member in the depth direction are attached to a tip thereof, and a position shifter provided with a welding torch is also provided. A possible welding robot is connected to a welding robot control device, and the welding robot control device processes an image taken by the CCD camera and outputs an OFF image (characteristic image of a member) and an ON image (laser slit image) of laser slit light. An automatic welding line recognition device connected to an image processing device that stores an image for recognizing a light irradiation position) as a registered image is preferable.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention described in claims 1 and 2 of the present invention will be described with reference to FIGS.
This will be described below.

FIG. 1 is an overall configuration diagram of an automatic welding line recognition apparatus according to an example of an embodiment considered to be the best embodiment of the present invention.
2 (a) is a perspective view of a member temporarily attached on a base material, (b) -1 is a front registration image, (b) -2 is a front comparison image, and FIG. b) -3 shows a front view pattern matching, (c) -1 shows a side registration image, (c) -2 shows a side comparison image, and (c) -3 shows a side view pattern matching.

In these figures, a member 2 is placed on a base material 1.
Is temporarily attached, and a CCD camera 4 attached to the tip of the welding robot 3 captures an image of a reference member (hereinafter abbreviated as 2a).

The imaging is carried out by moving the welding robot 3 to a front position (hereinafter abbreviated as Pa ₁ ) which is separated from the member by a certain distance, and taking an image by the CCD camera 4.

The captured image is a memory as a front of the image (hereinafter referred to as La ₁₎ As shown in that Figure 2 (b) -1 processed by the image processing apparatus 5.

Similarly, the welding robot 3 is moved to a side surface position (hereinafter abbreviated as Pa ₂ ) of the member 2a at a predetermined distance, and an image is taken. This is also the memory as the side surface of the image (hereinafter referred to as La ₂₎ as shown in FIG. 2 (c) -1 in the same manner.

The reference starting point 2a is used for the welding start point Wa.
_1, teaches an end point Wa _2.

Next, based on these images, a position shift is detected by a pattern matching method. That is, the new member 2
Assuming that the welding robot 3b is to be welded, the welding robot 3 is first moved to the position Pa _{1 in the same} manner as when the member 2a is imaged.
Go to

Here, the member 2b is imaged by the CCD camera 4. Compare this image Lb ₁ with the registered image La ₁ ,
Perform pattern matching.

The horizontal deviation ΔX and the vertical deviation ΔY shown in FIG. 2B-3 are stored here.

At this time, if the correlation coefficient r is small, it can be treated as a poor matching.

Next, after the welding robot 3 is moved to the position Pa ₂ on the side surface of the member, the member 2b is imaged by the CCD camera 4 from the side surface.

[0026] Compared with the image Lb ₂ and the registered image La _2, to pattern matching. Then, the depth shift ΔZ shown in FIG. 2C-3 obtained here is stored. At this time, if the correlation coefficient r is small, it can be processed as a poor matching as described above.

Next, the deviation amounts ΔX, ΔY, Δ
Z is converted into a robot coordinate system based on a camera focal length, a distance between a camera and a member, a camera imaging angle, and the like, and is output to the welding robot controller 6 as a correction amount.

The welding robot 3 performs welding by moving the welding start point and the welding end point in parallel by the correction amount based on the correction amount. This enables accurate welding of the member 2b along the welding line.

At this time, if an accurate position is required for the welding end point, correction for the welding end point can be performed by performing pattern matching of the end portion of the member 2.

In the present invention, since the displacement of the member can be recognized in a non-contact manner by the CCD camera, sensing can be performed at a higher speed than in the conventional touch sensor method.

Next, the operation of the first and second aspects of the present invention will be described in detail. In the present invention having the above structure, a CCD camera for imaging a member and an image processing for inputting and processing this image are provided. The apparatus recognizes a member position shift by a pattern matching method and corrects a teaching point. Then, the welding robot performs welding using the corrected position data. As a result, the member position deviation can be corrected regardless of the mounting accuracy of the member, and accurate welding can be performed.

First, at the time of teaching, a member to be welded is imaged from two directions, substantially in front and in the lateral direction, using a CCD camera attached to the tip of the robot. The movement to the imaging position at this time is performed by a welding robot.
This image is stored in the image processing device.

Next, at the time of welding, the welding robot is similarly brought to the same imaging position, and the members are imaged by the CCD camera.

Images registered with this (front view and side view)
Then, the deviation amount of each member is detected.

In the front view, the horizontal direction and the vertical direction can be detected, and in the side view, the depth direction can be detected. By recognizing from the side direction, it is possible to detect a member position shift three-dimensionally. This displacement is performed by a pattern matching method, that is, a normalized correlation method.

Thus, the calculated position of the member is compared with the position of the member at the time of registration, and the positional deviation in the left, right, up, down, and depth directions is communicated to the welding robot and corrected.

By performing welding with the corrected three-dimensional position data, the welding robot can accurately weld on a welding line corresponding to the member, thereby providing high quality welding.

In the world of image processing, pattern matching is generally found by a normalized correlation method. Incidentally, the normalized correlation method will be described as follows.

The correlation coefficient r of the target portion in the image with a positional deviation of the offset of the comparison member L and the registered member M (u, v) is the comparison image _{L i (x i, y i} ) and (0 ≦
i <N), when the registered image and M _i, is expressed by the following equation (1).

[0040]

(Equation 1)

Here, I _i = I (u + X _i , v + Y _i ).

The value of r is always in the range of -1 ≦ r ≦ 1.

The value of r becomes 1 when the comparison image and the registered image completely match. At this time, there are values of a and b satisfying the following relationship for all i. I _i = a * M _i + b a> 0

When the value of r becomes -1, the same relationship holds except that a <0 in the case of complete mismatch.

The shift amount can be detected by obtaining (u, v) at which the correlation coefficient r becomes maximum.

Next, an example of a preferred embodiment of the present invention described in claims 3 and 4 of the present invention will be described with reference to FIGS.

FIG. 3 is an overall configuration diagram of an automatic welding line recognition apparatus according to an example of the above-described preferred embodiment of the present invention.
4 (a) is a perspective view of a member temporarily attached to a base material, (b) -1 is a laser slit light OFF registered image, and (b) -2 is a laser slit. Light OFF comparative image, (b) -3 shows pattern matching when laser slit light is OFF, and (c) -1
Is a registered image of the laser slit light ON, (c) -2 is a comparative image of the laser slit light ON, (c) -3 is a diagram illustrating pattern matching when the laser slit light is ON, and FIG. 5 is a laser slit light sensor. It is a figure explaining the principle of recognizing a depth direction.

In these figures, a small member 2 is temporarily attached on a base material 1, and is attached by a CCD camera 4 for imaging a member attached to the tip of a welding robot 3.
An image of a reference member (hereinafter referred to as 2a) is taken. A welding torch 7 is also attached to the tip of the welding robot 3 in parallel with the CCD camera 4.

The imaging is performed by moving the welding robot 3 to a front position (hereinafter, abbreviated as Pa ₁ ) that is a predetermined distance away from the member, and imaging with the CCD camera 4.

The picked-up image is processed by the image processing device 5, and as shown in FIG.
FF registration image as a (hereinafter, referred to as La _1), is a memory as an image for recognizing the characteristic points of the member.

Similarly, the laser slit optical sensor 8 for recognizing the position in the depth direction of the member attached to the tip of the welding robot 3 is turned on to take an image. This also applies to FIG.
- (c) registering the image of the laser slit light ON as shown in -1 (hereinafter referred to as La _2), is a memory as an image for recognizing the laser slit beam position.

The welding start point Wa is determined by the reference member 2a.
_1, teaches an end point Wa _2.

Next, based on these images, a position shift is detected by a pattern matching method.

That is, assuming that a new member (hereinafter referred to as 2b) is to be welded, first, the welding robot 3 is moved to the position Pa in front of the member in the same manner as when the member 2a is imaged.
To move to _1.

Here, the member 2b is imaged by the CCD camera 4. This image (hereinafter abbreviated as Lb ₁ ) is compared with the registered image La ₁ to perform pattern matching.

The horizontal deviation ΔX and the vertical deviation ΔY obtained here are stored. At this time, if the correlation coefficient r is small, it can be processed as a poor matching.

Next, the laser slit light sensor 8 is turned on to irradiate laser light, and a laser slit light ON image is taken.

This image (hereinafter abbreviated as Lb ₂ ) is compared with the registered image La ₂ to perform pattern matching.
Then, the depth shift ΔZ shown in FIG. 4C-3 obtained here is stored. At this time, if the correlation coefficient r is small, it can be processed as a poor matching as described above.

Here, since it is better to perform pattern matching only on the laser slit light, a mask in which the range of the laser slit light to be processed is limited is set.
When pattern matching is performed in the mask, the deviation of the laser slit light is obtained with high accuracy.

At this time, more accurate processing can be expected by performing pattern matching between the images after differential processing.

Next, whether the depth in the Z direction can be recognized by capturing an image from the front without moving the welding robot 3 will be described with reference to FIG.

For this, the laser slit light sensor 8 is set so that the CCD camera 4 is irradiated with the laser slit light obliquely.

Therefore, when the position of the member is the same as the registered image (the position of PZ (0)), when the laser slit light is applied to the same position as that of the registered image La ₂ and pattern matching is performed, the deviation becomes 0. Become.

When the member 2a is close to the welding robot 3 (PZ (+) position), the irradiation position of the laser slit light is shown on the right side of the image. Can be recognized as a shift amount (ΔXz1) in the horizontal direction on the XY plane, that is, in the X direction.
This is corrected from the inclination angle θ of the laser slit light (Δ
Z = ΔXz1 / tan (θ)), which can be obtained as a shift amount in the Z direction.

When the member 2a is far from the welding robot 3 (PZ (-) position), the irradiation position of the laser slit light is shown on the left side of the image. The shift can be recognized as a shift amount (−ΔXz2) in the horizontal direction on the XY plane, that is, in the X direction. This can be corrected from the tilt angle θ of the laser slit light (ΔZ = −ΔXz2 / tan (θ)) to obtain the shift amount in the Z direction.

In this way, it is possible to recognize the amount of displacement in the Z direction in the depth direction by using the laser slit light sensor 8 from an image taken from the front.

At this time, the position of the laser slit light is recognized by the pattern matching method. However, this may be recognized by binarizing the laser slit light and comparing the position with the position of the registered image to calculate. It is possible.

As described above, the obtained shift amounts ΔX, ΔY,
ΔZ is converted into a robot coordinate system based on a camera focal length, a distance between the CCD camera 4 and the member 2, a camera imaging angle, and the like, and output to the welding robot controller 6 as a correction amount.

The welding robot 3 performs welding by moving the welding start point and the welding end point in the XYZ directions by the correction amount based on the correction amount. This enables accurate welding of the member 2b along the welding line.

At this time, if an accurate position is required for the welding end point, it is possible to make a correction for the welding end point by performing pattern matching on the end portion of the member 2.

In the present invention, the CCD camera 4
By being able to recognize the member position shift without contact
Sensing can be performed at a higher speed than the conventional touch sensor method.

The preferred embodiments of the present invention described in claims 3 and 4 of the present invention have been described above. The operation of the present invention is summarized as follows.

That is, according to the present invention having the above-described configuration, a CCD camera for capturing an image of a member and an image processing device for inputting and processing the image recognize a member position shift by a pattern matching method. Correct the teaching point.

The welding robot performs welding by using the corrected position data, whereby the positional deviation of the member can be corrected irrespective of the mounting accuracy of the member, and accurate welding can be performed.

Therefore, at the time of teaching, first, the member to be welded was turned on and off from the laser slit light from almost the front using a CCD camera attached to the robot tip.
N images are taken. The movement to the imaging position at this time is performed by a welding robot. These two images are stored in the image processing device.

Next, at the time of welding, the welding robot is similarly brought to the same imaging position, and the members are imaged by the CCD camera.

This is compared with the registered images (the laser slit light OFF image and the ON image), and the shift amount of each image is detected by the pattern matching method.

In the image in which the laser slit light is OFF, XY
The depth direction in the Z direction can be detected in the horizontal and vertical directions on the plane and the image of the laser slit light ON.

By recognizing the images of the laser slit light OFF and ON, it is possible to three-dimensionally detect the displacement of the member. This displacement is performed by a pattern matching method, that is, a normalized correlation method.

Thus, the obtained member position is compared with the member position at the time of registration, and the positional deviation in the left, right, up, down, and depth directions is communicated to the welding robot and corrected.

The welding robot performs welding with the corrected three-dimensional corrected position data, so that the welding robot can accurately weld on the welding line corresponding to the member, thereby providing high quality welding.

Although an example of the embodiment of the present invention has been described in detail above, the present invention is not limited to the above-described embodiment, and various changes can be made within the scope of the technical idea of the present invention. Also belong to the technical scope of the present invention.

[0083]

According to the method and the apparatus for welding by the welding robot of the present invention, the position of the welding member can be accurately and quickly recognized in a non-contact manner by the CCD camera, thereby correcting the conventional welding defect due to the displacement of the welding line. No repair is required, and high quality and highly efficient welding can be provided.

[Brief description of the drawings]

FIG. 1 is an overall explanatory view of an apparatus according to the first and second aspects of the present invention.

FIG. 2 shows an explanatory diagram of pattern matching of the present invention,
(A) is a perspective view of a member temporarily attached on a base material, (b)-
1 front registration image, (b) -2 is front comparison image, (b)-
3 shows front view pattern matching, and (c)-
1 is a side registration image, (c) -2 is a side comparison image, (c)
-3 indicates side view pattern matching.

FIG. 3 is an overall explanatory view of a welding line automatic recognition device according to the invention described in claims 3 and 4 of the present invention.

FIG. 4 is an explanatory diagram of pattern matching according to the present invention;
(A) is a perspective view of a member temporarily attached on a base material, (b)-
1 is a laser slit light OFF registration image, (b) -2 is a laser slit light OFF comparison image, (b) -3 is a pattern matching explanatory diagram when the laser slit light is OFF,
(C) -1 is a registered image of the laser slit light ON, (c)
2 is a laser slit light ON comparative image, and (c) -3 is an explanatory diagram of pattern matching when the laser slit light is ON.

FIG. 5 is an explanatory diagram of a principle of recognizing a depth direction (Z direction) by a laser slit optical sensor according to the present invention.

[Explanation of symbols]

 Reference Signs List 1 base material 2 member 3 welding robot 4 CCD camera 5 image processing device 6 welding robot control device 7 welding torch 8 laser slit optical sensor

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-62-267086 (JP, A) JP-A-6-117827 (JP, A) JP-A-59-4796 (JP, A) JP-A-62-26786 33064 (JP, A) JP-A-61-37377 (JP, A) JP-A-62-161473 (JP, A) JP-A-55-30339 (JP, A) JP-A-58-209481 (JP, A) JP-A-7-171700 (JP, A) JP-A-55-87283 (JP, A) JP-A-60-201863 (JP, A) JP-A-60-177201 (JP, A) (58) (Int.Cl. ⁷ , DB name) B23K 9/127

Claims (4)

(57) [Claims] Upon welding automatically welding apparatus such as 1. A welding robot a plurality of members, imaging the member CC
Using a D camera and an image processing device that inputs and processes this image, images obtained by capturing members from two directions, front and side, are used, and a front image of each member and an image of a reference member are used.
By pattern matching with image, horizontal and vertical
It detects the displacement (X, Y) of the mounting position and
By pattern matching between the image and the image of the reference member
A welding line automatic recognition method characterized by detecting a displacement of a mounting position in a depth direction (Z) and correcting the welding start point and end point taught in advance to perform welding. 2. A CCD camera for picking up a member is attached to the tip and a positionally movable welding robot provided with a welding torch is connected to a welding robot controller, and the welding robot controller is imaged by the CCD camera. Connected to an image processing apparatus for processing the
The apparatus stores images of the front and side surfaces of the reference member in two directions as registered images, and stores the front image of each member and
Horizontal and horizontal by pattern matching with the registered image
When the displacement (X, Y) of the vertical mounting position is detected,
Next, pattern matching between the side image and the registered image
Of the mounting position in the depth direction (Z).
An automatic welding line recognition apparatus, wherein the welding start point and the end point taught in advance are corrected . 3. When a plurality of members are automatically welded by a welding device such as a welding robot, a laser slit is provided by an image processing device which inputs and processes the images by a CCD camera for imaging the members and a laser slit optical sensor. Optical OF
An image of each member taken from the front at the time of F and a reference
By performing pattern matching with the image of the member to be mounted, the mounting position shift of the member on the XY plane of the member is detected.
The amount of displacement of the mounting position in the depth (Z) direction of the member is detected from the irradiation position deviation of the laser slit light when the laser slit light is turned on, and the mounting position of the member is calculated three-dimensionally based on the detected amount. Automatic welding line recognition method characterized by correcting welding start and end points. 4. A CCD camera for imaging a member and a laser slit optical sensor for recognizing the position of the member in the depth direction are attached to a tip thereof, and a position-movable welding robot provided with a welding torch is connected to a welding robot control device. Further, the welding robot control device is connected to an image processing device that processes an image captured by the CCD camera, and the image processing device is configured to operate when the laser slit light is turned off.
The image of the reference member in the image is used as the first registered image.
And the reference when the laser slit light is ON
The image of the member to be
Photo taken from the front of each member when the laser slit light is OFF
For pattern matching between an image image and the first registered image
Of the mounting position of the member on the XY plane
Each member when the laser slit light is ON
Of the captured image from the front of the camera and the second registered image
Mounting position in the depth (Z) direction
The displacement amount is detected, and thereby the member is three-dimensionally removed.
The welding position is calculated and the welding start point and end point taught in advance are calculated.
An automatic welding line recognition device characterized in that the end point is corrected .