CN114800574A - Robot automatic welding system and method based on double three-dimensional cameras - Google Patents

Abstract

The invention provides a robot automatic welding system and method with dual three-dimensional cameras. The system includes a robot, a first three-dimensional camera installed outside the robot, a second three-dimensional camera installed at the end of the robot and welding equipment, a control box for controlling the movement of the robot, and a system host computer; the first three-dimensional camera is used to photograph and obtain a complete workpiece point cloud; the system host computer is used for system coordinate system alignment, shooting trajectory planning of the second three-dimensional camera, welding seam locating and welding path planning, wherein the shooting trajectory planning of the second three-dimensional camera includes, according to the complete workpiece The point cloud plans the shooting points of the second three-dimensional camera to form the shooting motion trajectory of the robot; the second three-dimensional camera is used to obtain the workpiece point cloud within the field of view of the current shooting point according to the shooting motion trajectory; the welding equipment is used for The welding task is completed according to the workpiece point cloud and welding trajectory within the field of view of the current shooting point.

Description

A robot automatic welding system and method based on dual three-dimensional cameras

technical field

The invention belongs to the technical field of welding, and in particular relates to a robot automatic welding system and method based on dual three-dimensional cameras.

Background technique

The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

Although welding robots are widely used at present, the complex programming work before welding has always been an obstacle to the development of welding robots. The programming methods of welding robots mainly include manual teaching programming and offline programming. Manual teaching programming needs to be reprogrammed at a certain time interval to ensure the welding accuracy under the workpiece clamping error; the way of offline programming is to model the robot and the workpiece, realize the path planning on the software, and for the 3D model of the workpiece Dependence is great. For small batch, multi-variety, and non-standard workpiece welding scenarios, the above two programming methods are less applicable.

In order to solve the above problems, with the development of machine vision, the use of three-dimensional cameras to automatically locate the welding trajectory can realize the automatic welding of robots without programming and teaching. However, this process requires the planning of the shooting point of the 3D camera, and the shooting point planning is still carried out manually in the current industrial scene.

In addition, the existing software is used to plan the shooting trajectory of the 3D camera with eyes on the hand, and the disadvantages are:

1. It depends on the 3D model of the workpiece. If the workpiece does not have a 3D model, it will be difficult to realize;

2. The quality of the operator is high, and the learning of professional software requires a certain cost; 3. The degree of automation is low, and the use of software for planning cannot avoid manual intervention.

SUMMARY OF THE INVENTION

In order to solve the above problems, the present invention proposes a robot automatic welding system and method based on dual three-dimensional cameras. The present invention aims at the planning of the shooting point of the camera installed at the end of the robot before welding, and uses the overhead shooting installed outside the robot workbench. The camera, combined with 3D vision technology, realizes the automatic planning of camera shooting points before welding; for welding workpieces, the camera installed at the end of the robot is used, combined with 3D vision technology, to realize the welding seam location and the shooting point as a unit. Automated welding.

According to some embodiments, the present invention adopts the following technical solutions:

In a first aspect, the present invention provides a robot automatic welding system with dual three-dimensional cameras.

A robot automatic welding system with dual three-dimensional cameras, comprising: a robot, a first three-dimensional camera installed outside the robot, a second three-dimensional camera installed at the end of the robot and welding equipment, a control box for controlling the movement of the robot, and a system host computer;

The first three-dimensional camera is used to capture and obtain a complete workpiece point cloud;

The system host computer is used for system coordinate system alignment, shooting trajectory planning of the second three-dimensional camera, welding seam locating and welding path planning, wherein the shooting trajectory planning of the second three-dimensional camera includes planning according to the complete workpiece point cloud. The shooting point of the second 3D camera forms the shooting trajectory of the robot;

The second three-dimensional camera is used to obtain the workpiece point cloud within the field of view of the current shooting point according to the shooting motion trajectory;

The welding equipment is used to complete the welding task according to the workpiece point cloud and the welding track in the field of view of the current shooting point.

In a second aspect, the present invention provides a robot automatic welding method with dual three-dimensional cameras.

A robot automatic welding method with dual three-dimensional cameras, using the robot automatic welding system with dual three-dimensional cameras described in the first aspect, comprising:

Calibrate the relationship between the welding gun tool coordinate system, the two camera coordinate systems and the robot base coordinate system;

Use the first three-dimensional camera to capture the complete workpiece point cloud;

Plan the shooting point of the second 3D camera according to the complete workpiece point cloud to form the shooting trajectory of the robot;

The second three-dimensional camera is used to obtain the workpiece point cloud within the field of view of the current shooting point according to the shooting motion trajectory;

According to the point cloud captured by the current shooting point, the welding seam is located and the welding trajectory of the robot is planned;

Using welding equipment, the welding task is completed according to the workpiece point cloud and welding trajectory in the field of view of the current shooting point.

Compared with the prior art, the beneficial effects of the present invention are:

(1) The present invention realizes the whole process automation of the robot welding system. At present, before the 3D camera guides the robot to weld, it is necessary to manually plan the shooting point of the camera. The invention uses dual three-dimensional cameras, and solves the problem of manual planning of shooting points through the combination of three-dimensional vision technology and robot technology.

(2) The present invention uses three-dimensional visual information to plan the shooting point of the hand camera, which has high efficiency and strong universality. It is suitable for welding workpieces with multiple varieties, multiple welds and poor processing consistency.

(3) The present invention uses a three-dimensional camera to acquire high precision, and uses a positioning method to pick up the welding seam with high precision, which ensures high quality of welding.

Description of drawings

The accompanying drawings forming a part of the present invention are used to provide further understanding of the present invention, and the exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention.

1 is a schematic diagram of a robotic automatic welding system with dual three-dimensional cameras shown in Embodiment 1 of the present invention;

FIG. 2 is a flowchart of a robot automatic welding method for dual three-dimensional cameras shown in Embodiment 2 of the present invention;

Among them, 1. Multi-DOF robot, 2. Camera outside the hand, 3. Camera on hand, 4. Welding equipment, 5. Control box, 6. System host computer.

Detailed ways

The present invention will be further described below with reference to the accompanying drawings and embodiments.

It should be noted that the following detailed description is exemplary and intended to provide further explanation of the invention. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

It should be noted that the terminology used herein is for the purpose of describing specific embodiments only, and is not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural as well, furthermore, it is to be understood that when the term "comprising" or "comprising" is used in this specification, it indicates the presence of a characteristic , steps, operations, devices, components and/or combinations thereof.

In the present invention, terms such as "connected" and "connected" should be understood in a broad sense, indicating that it can be a fixed connection, an integral connection or a detachable connection; it can be directly connected or indirectly connected through an intermediate medium. For the relevant scientific research or technical personnel in the field, the specific meanings of the above terms in the present invention can be determined according to the specific situation, and should not be construed as a limitation of the present invention.

Example 1

This embodiment provides a robot automatic welding system with dual three-dimensional cameras.

As shown in Figure 1, a robot automatic welding system based on dual 3D cameras includes a multi-degree-of-freedom robot 1, a 3D camera installed outside the robot, a 3D camera installed at the end of the robot and welding equipment 4, a control box 5 that controls the movement of the robot, and the system Host computer 6; wherein the system host computer 6 is respectively connected with the multi-degree-of-freedom robot 1, the 3D camera installed outside the robot, the 3D camera installed at the end of the robot, the welding equipment, and the control box 5 that controls the movement of the robot, and the control box 5 is connected to the robot.

The three-dimensional camera installed outside the robot is combined with the robot in an Eye-to-Hand, which is simply referred to as the out-of-hand camera 2 in the description of this embodiment.

The three-dimensional camera installed at the end of the robot is combined with the robot in an Eye-in-Hand, which is referred to as the camera 3 on the hand for short in the description of this embodiment.

The welding equipment described is not limited to the use of various brands of welding torches, welding machines and wire feeders according to different welding processes.

The out-of-hand camera is used for the shooting point planning of the on-hand camera. According to the requirements of the welding scene, it is usually installed above the working scene of the robot, and the installation method includes but is not limited to the gantry type. The device has a large field of view and can obtain a complete workpiece point cloud.

The hand-held camera is used for welding seam locating, and the device selects a camera with a smaller field of view, and can acquire the workpiece point cloud within the field of view of the current shooting point.

The 3D camera has the function of acquiring the point cloud on the surface of the scanned object, its measurement accuracy is less than 1mm, and the frame rate of acquiring the depth map is greater than 1 frame per second. Usually, a high-precision 3D area array structured light camera is used, but no It is limited to three-dimensional visual scanning devices that use multiple brands, multiple structures, similar principles and have the same function.

The multi-degree-of-freedom robot is used for carrying a welding gun and a camera on hand, and performing the shooting motion of the camera on the hand and the welding motion of the welding gun, and is not limited to the use of industrial robots of various brands.

The system host computer includes system coordinate system alignment function, point cloud processing function, communication function, interactive function, file reading and storage function, model visualization function and other functions that meet the needs of the welding system. Limited to Windows and Linux system platforms.

The main purpose of the system coordinate system alignment function is to calibrate the coordinate system of the entire system from one to the base coordinate system of the multi-degree-of-freedom robot, including the spatial transformation relationship between the base coordinate system of the multi-degree-of-freedom robot and the coordinate system of the welding gun tool. The calibration of the spatial transformation relationship between the base coordinate system of the multi-degree-of-freedom robot and the coordinate system of the camera outside the hand, and the calibration of the spatial transformation relationship between the base coordinate system of the multi-degree-of-freedom robot and the coordinate system of the camera on the hand, the three calibration relationships have no sequence relationship, specifically:

First, the 3D camera itself is calibrated with precise internal parameters, and the internal parameters of the camera are known;

Secondly, calibrate the base coordinate system of the multi-degree-of-freedom robot and the welding gun tool coordinate system: guide the robot to drive the end of the welding gun to point to a fixed space point for many times, change the posture of the robot end for each pointing, without changing the position of the end, after calculating The spatial transformation relationship ^Base T _Tool of the welding gun tool coordinate system relative to the multi-degree-of-freedom robot base is obtained.

Then, calibrate the base coordinate system of the multi-degree-of-freedom robot and the camera coordinate system outside the hand: the calibration plate is fixedly installed at the end of the robot, and the camera shoots the calibration plate in different robot poses multiple times, and records the multi-degree-of-freedom robot. position and attitude. Define the homogeneous transformation matrix from the robot end to the base as ^End T _Base , and the homogeneous transformation matrix from the 3D camera to the calibration plate coordinate system as ^CameraA T _Object . Since the relative position of the robot end and the calibration plate does not change, the time between two different shots can be expressed as:

^End T _Base2 · ^Base2 T _CameraA2 · ^CameraA2 T _Object = ^End T _Base1 · ^Base1 T _CameraA1 · ^CameraA1 T _Object

And because the position of the camera to the robot base is relatively unchanged, namely:

^Base2 T _CameraA2 = ^Base1 T _CameraA1 = ^Base T _CameraA ,

Available:

Through multiple shots, the above equations are solved, and the spatial transformation relationship ^Base T _CameraA between the base coordinate system of the multi-degree-of-freedom robot and the coordinate system of the out-of-hand camera is obtained;

Finally, the base coordinate system of the degree-of-freedom robot and the coordinate system of the camera on the hand are calibrated: the fixed position of the calibration board is unchanged, the posture of the robot is changed, and the camera on the hand can shoot multiple times of different poses on the calibration board. Define the homogeneous transformation matrix from the 3D camera to the calibration board coordinate system as ^CameraB T _Object . Since the relative position of the robot base and the calibration plate does not change, the time between two different shots can be expressed as:

^Base T _End2 · ^End2 T _CameraB2 · ^CameraB2 T _Object = ^Base T _End1 · ^End1 T _CameraB1 · ^CameraB1 T _Object

And because the relative position of the 3D camera to the end of the robot does not change, that is:

^End2 T _CameraB2 = ^End1 T _CameraB1 = ^End T _CameraB ,

Have:

Through multiple shots, the above equations are solved, and the spatial transformation relationship ^End T _CameraB between the base coordinate system of the multi-degree-of-freedom robot and the coordinate system of the out-of-hand camera is obtained;

Finally, the hand-eye relationship matrix between the camera on the hand and the robot base is obtained as:

^Base T _CameraB = ^Base T _End ^End T _CameraB

In order to allow the camera to accurately photograph the welding seam of the workpiece, it is necessary to determine the shooting position of the camera. There may be more than one position. These points plus the intermediate position during the welding process constitute the shooting trajectory of the robot. The current acquisition method of shooting track is mainly manual teaching and some special software settings, and the degree of automation of these two methods is not high.

The process of guiding robot welding with the camera in hand is as follows:

1. Weld seam locating

The point cloud is obtained by shooting at the shooting point, the point cloud is processed by algorithm, the characteristics of the welding seam are obtained, and then the welding trajectory is planned according to the welding process requirements. Finally, the welding trajectory is transformed from the camera coordinate system to the robot base coordinate system.

2. Robot welding

Analyze the welding trajectory, generate control scripts or programs for different robots, and control the robot to complete the welding.

Therefore, in order to solve the above problems, the present embodiment proposes a solution of adding a three-dimensional camera to realize automatic planning of shooting points, so as to improve the automation degree of the existing solution.

In this embodiment, the system host computer plans the shooting point of the second 3D camera according to the complete workpiece point cloud to form the shooting motion trajectory of the robot; the second 3D camera is used to obtain the current shooting point field of view according to the shooting motion trajectory. The point cloud of the workpiece; the welding equipment is used to complete the welding task according to the point cloud of the workpiece and the welding track within the field of view of the current shooting point.

Embodiment 2

This embodiment provides a robot automatic welding method with dual three-dimensional cameras.

A robot automatic welding method with dual 3D cameras, the robot automatic welding system using the dual 3D cameras described in Embodiment 1 includes:

(1) To calibrate the relationship between the welding gun tool coordinate system, the two camera coordinate systems and the robot base coordinate system;

(2) Use an off-hand camera to capture the complete workpiece point cloud;

(3) Plan the shooting point of the camera on the hand according to the complete workpiece point cloud to form the shooting trajectory of the robot;

(4) Use the hand-held camera to obtain the workpiece point cloud within the field of view of the current shooting point;

(5) Obtain and generate the welding track through the welding seam locating method;

(6) The robot completes the welding task of the current shooting point;

(7) Repeat steps (4) to (6) until the welding tasks of all shooting points are completed.

The step (1) is to calibrate the space conversion relationship between the multi-degree-of-freedom robot base coordinate system and the welding torch tool coordinate system, the multi-degree-of-freedom robot base coordinate system and the space conversion relationship calibration of the out-of-hand camera coordinate system, and the multi-degree-of-freedom robot. For the calibration of the spatial transformation relationship between the base coordinate system and the hand-held camera coordinate system, the three calibration processes have no sequence relationship, specifically:

First, the 3D camera itself is calibrated with precise internal parameters, and the internal parameters of the camera are known;

Secondly, calibrate the base coordinate system of the multi-degree-of-freedom robot and the welding gun tool coordinate system: guide the robot to drive the end of the welding gun to point to a fixed space point for many times, change the posture of the robot end for each pointing, without changing the position of the end, after calculating The spatial transformation relationship ^Base T _Tool of the welding gun tool coordinate system relative to the multi-degree-of-freedom robot base is obtained.

Then, calibrate the base coordinate system of the multi-degree-of-freedom robot and the camera coordinate system outside the hand: the calibration plate is fixedly installed at the end of the robot, and the camera shoots the calibration plate in different robot poses multiple times, and records the multi-degree-of-freedom robot. position and attitude. Define the homogeneous transformation matrix from the robot end to the base as ^End T _Base , and the homogeneous transformation matrix from the 3D camera to the calibration plate coordinate system as ^CameraA T _Object . Since the relative position of the robot end and the calibration plate does not change, the time between two different shots can be expressed as:

^End T _Base2 · ^Base2 T _CameraA2 · ^CameraA2 T _Object = ^End T _Base1 · ^Base1 T _CameraA1 · ^CameraA1 T _Object

And because the position of the camera to the robot base is relatively unchanged, namely:

^Base2 T _CameraA2 = ^Base1 T _CameraA1 = ^Base T _CameraA ,

Available:

Through multiple shots, the above equations are solved, and the spatial transformation relationship ^Base T _CameraA between the base coordinate system of the multi-degree-of-freedom robot and the coordinate system of the out-of-hand camera is obtained;

Finally, the multi-degree-of-freedom robot base coordinate system and the hand camera coordinate system are calibrated: the fixed position of the calibration board is unchanged, the robot's posture is changed, and the hand camera can shoot the calibration board with different poses for many times. Define the homogeneous transformation matrix from the 3D camera to the calibration board coordinate system as ^CameraB T _Object . Since the relative position of the robot base and the calibration plate does not change, the time between two different shots can be expressed as:

^Base T _End2 · ^End2 T _CameraB2 · ^CameraB2 T _Object = ^Base T _End1 · ^End1 T _CameraB1 · ^CameraB1 T _Object

And because the relative position of the 3D camera to the end of the robot does not change, that is:

^End2 T _CameraB2 = ^End1 T _CameraB1 = ^End T _CameraB ,

Have:

Through multiple shots, the above equations are solved, and the spatial transformation relationship ^End T _CameraB between the base coordinate system of the multi-degree-of-freedom robot and the coordinate system of the out-of-hand camera is obtained;

Finally, the hand-eye relationship matrix between the camera on the hand and the robot base is obtained as:

^Base T _CameraB = ^Base T _End ^End T _CameraB

In the step (3), the purpose of planning the shooting points of the camera on the hand is to make the camera on the hand complete the point cloud of all the weld areas of the workpiece obtained by the camera outside the hand by shooting at multiple points.

In the step (3), the methods for planning the shooting point of the camera on the hand mainly include the following two:

1) By geometrically segmenting the workpiece point cloud, the shooting angle of the hand camera is determined according to the line and surface features of the point cloud. Combined with the working distance range of the hand camera, set the shooting position of the hand camera. The methods of geometrically segmenting the workpiece point cloud include but are not limited to segmentation using point cloud processing algorithms and manual segmentation using interactive equipment. The manual segmentation includes but is not limited to mouse drawing or selection, touch screen drawing or selection, and the like.

2) When there is a three-dimensional model of the workpiece, use a third-party modeling software to set the shooting point on the three-dimensional model. The point cloud model obtained by shooting is registered with the three-dimensional model, and the shooting point is converted into the robot base coordinate system.

The sources of the three-dimensional model include but are not limited to: scanning, splicing, and reverse point cloud fusion using the 3D camera at the end of the robot; modeling using three-dimensional CAD software; using existing models for transformation; In the case that the original CAD design 3D model is available and the consistency of the workpiece is good, the CAD 3D digital model of the design is selected, otherwise the digital 3D model obtained in reverse is preferred.

In the step (5), the positioning method mainly includes the following two methods, but as long as the welding seam feature can be picked up, the effect of the present invention can be achieved.

1) By geometrically segmenting the three-dimensional point cloud model, extracting the feature points of the workpiece weld through line and surface features, and fitting and smoothing the feature points to obtain the weld track. The methods of geometrically segmenting the workpiece point cloud include but are not limited to segmentation using point cloud processing algorithms and manual segmentation using interactive equipment. The manual segmentation includes but is not limited to mouse drawing or selection, touch screen drawing or selection, and the like. At the same time, using the neighborhood information of the feature points, the welding torch pose is planned.

2) When there is a usable 3D digital model, directly pick up features in the digital model as welding tracks. The sources of the digital 3D model include, but are not limited to: scanning, splicing, and reverse point cloud fusion using the 3D camera at the end of the robot; modeling using 3D CAD software; using existing models for transformation; , in the case that the original CAD design 3D model is available and the consistency of the workpiece is good, the CAD 3D digital model of the design is selected, otherwise the digital 3D model obtained in reverse is preferred. The planning of the welding torch posture is the same as step 1).

In the step (6), the welding trajectory obtained in the step (5) is converted into a robot control language, sent to the robot controller, and the robot is used for welding.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. A robotic automated welding system of dual three-dimensional cameras, comprising: the robot comprises a robot, a first three-dimensional camera arranged outside the robot, a second three-dimensional camera and welding equipment arranged at the tail end of the robot, a control box for controlling the robot to move and a system upper computer;

the first three-dimensional camera is used for shooting and acquiring a complete workpiece point cloud;

the system upper computer is used for system coordinate system alignment, shooting track planning, welding seam locating and welding path planning of the second three-dimensional camera, wherein the shooting track planning of the second three-dimensional camera comprises the steps of planning the shooting point position of the second three-dimensional camera according to complete workpiece point cloud to form a shooting motion track of the robot;

the second three-dimensional camera is used for acquiring workpiece point clouds in the current shooting point location view according to the shooting motion track;

the welding equipment is used for completing a welding task according to the point cloud and the welding track of the workpiece in the current shooting point location view.

2. The dual three-dimensional camera robotic automated welding system according to claim 1, wherein the system coordinate system alignment comprises: calibrating the space conversion relation between the robot base coordinate system and the welding equipment coordinate system, calibrating the space conversion relation between the robot base coordinate system and the first three-dimensional camera coordinate system, and calibrating the space conversion relation between the robot base coordinate system and the second three-dimensional camera coordinate system.

3. The automated welding system of robots with two three-dimensional cameras according to claim 2, wherein the calibration of the spatial transformation relationship between the robot base coordinate system and the welding equipment coordinate system comprises guiding the robot to drive the welding equipment to point to a fixed spatial point for a plurality of times, changing the posture of the robot end pointed each time, and obtaining the spatial transformation relationship between the welding equipment coordinate system and the robot base without changing the position of the end.

4. The robotic automated welding system of dual three-dimensional cameras of claim 2, wherein the calibration of the spatial transformation relationship between the robot base coordinate system and the first three-dimensional camera coordinate system comprises fixedly mounting a calibration plate at a distal end of the robot, photographing the calibration plate of the robot at different poses by the first three-dimensional camera for a plurality of times, and recording the position and the posture of the robot at each photographing to obtain the spatial transformation relationship between the robot base coordinate system and the first three-dimensional camera coordinate system.

5. The automated robotic welding system of dual three-dimensional cameras of claim 2, wherein the calibration of the spatial transformation relationship between the robot-based coordinate system and the second three-dimensional camera coordinate system comprises a plurality of shots of different poses taken by the second three-dimensional camera on the calibration plate with the fixed position of the calibration plate unchanged to obtain the spatial transformation relationship between the multi-degree-of-freedom robot-based coordinate system and the hand-outside camera coordinate system.

6. The dual three-dimensional camera robotic automated welding system of claim 1, wherein said planning a shot point location of a second three-dimensional camera specifically comprises:

geometrically dividing the complete workpiece point cloud, determining the shooting angle of a second three-dimensional camera according to the line-surface characteristics of the point cloud, and setting the shooting position of the second three-dimensional camera by combining the working distance range of the second three-dimensional camera;

or,

when a three-dimensional model of a workpiece exists, setting a shooting point location on the three-dimensional model by using third-party modeling software, registering a point cloud model obtained by shooting with the three-dimensional model to obtain a shooting point, and converting the shooting point location into a robot base coordinate system;

alternatively, the welding track is generated by a weld locating method.

7. The automated welding system with double three-dimensional cameras according to claim 1, wherein the welding seam locating method comprises the steps of performing geometric segmentation on the three-dimensional point cloud model, extracting feature points of a workpiece welding seam through line-surface features, and fitting and smoothing the feature points to obtain a welding seam track.

8. The dual three-dimensional camera robotic automated welding system of claim 1, wherein the weld locating method comprises picking features directly in a digital model as a weld trajectory when a three-dimensional digital model is present.

9. The dual three-dimensional camera robotic automated welding system according to claim 1, wherein the three-dimensional digital model is modeled using three-dimensional CAD software.

10. A robot automated welding method using a dual three-dimensional camera according to any one of claims 1 to 9, comprising:

calibrating the relation among a welding gun tool coordinate system, two camera coordinate systems and a robot base coordinate system;

shooting by adopting a first three-dimensional camera to obtain complete workpiece point cloud;

planning a shooting point position of a second three-dimensional camera according to the complete workpiece point cloud to form a shooting motion track of the robot;

acquiring a workpiece point cloud in the current shooting point location view by adopting a second three-dimensional camera according to the shooting motion track;

carrying out welding seam locating according to the point cloud obtained by shooting at the current shooting point position, and planning the welding track of the robot;

and (4) completing a welding task according to the point cloud and the welding track of the workpiece in the current shooting point location view by adopting welding equipment.

Images