Control method of visual welding process system

A technology of welding process and control method, applied in the control field of visual welding process system, can solve problems such as trouble, manual sorting by workers, inability to realize automatic intelligent control device, etc., and achieve the effect of ensuring consistency and fast service response

Pending Publication Date: 2022-08-05

重庆创御智能装备有限公司

0 Cites 0 Cited by

AI-Extracted Technical Summary

Problems solved by technology

[0004] 1. The traditional method cannot realize the automatic intelligent control device, causing the device to require workers to manually assist the work

[0005] 2. The traditional method cannot automatically diagnose the...

Method used

Described redundant laser positioning detection technology, based on the point set registration method ICP algorithm of profile feature, realizes the profile feature of three-dimensional element or the accurate matching and fast recognition of point set, utilizes iteration to calculate correct correspondence step by step, many Viewing angle registration realizes the contour measurement and recognition of robot welding workpieces.

Described redundant laser positioning detection technology, based on the point set registration method ICP algorithm of profile feature, realizes the profile feature of three-dimensional element or the accurate matching of point set and fast identification, utilizes iteration to calculate correct correspondence step by step, many Viewing angle registration realizes the contour measurement and recognition of robot welding workpieces.

Described redundant laser positioning detection technology, based on the point set registration method ICP algorithm of profile feature, realizes the profile feature of three-dimensional element or the accurate matching of point set and fast recognition, utilizes iteration to calculate correct correspondence step by step, many Viewing angle registration realizes the contour measurement and recognition of robot welding workpieces.

Redundant laser location detection technology, by setting up the visual recognition and location algorithm based on deep learning model (deep belief network (DBN), convolutional neural network (CNN) and recursive neural network multi-layer perceptron (RNN)), Combined with the boundary pixel detection algorithm, and adopts a dual-camera vision system to realize the detection, identification and precise positioning of welding workpieces and weld trajectory;

Redundant laser positioning detection technology, by establishing the visual recognition and positioning algorithm based on deep learning model (deep belief network (DBN), convolutional neural network (CNN) and recursive neural network multi-layer perceptron (RNN)), Combined with the boundary pixel detection algorithm, and adopts a dual-camera vision system to realize the detection, identification and precise position...

Abstract

The invention discloses a control method of a visual welding process system. The control method comprises a visual guidance welding seam locating technology, a redundant laser positioning detection technology, a welding seam tracking and laser deviation correction technology, a robot autonomous motion planning technology, an autonomous adaptive welding technology and a welding seam defect identification technology. A visual guidance welding seam position searching technology comprises the steps of visual image collection and preprocessing, welding seam recognition, key point position coordinate calculation, welding seam teaching information and robot automatic path planning, and welding seam teaching information is formed by conducting image processing on a welding seam image of a workpiece to be welded and calculating welding seam starting and ending point position coordinates. In order to solve the problems of multi-layer multi-pass welding deviation accumulation and the like, the non-standard part welding seam locating technology based on machine vision guidance is researched, intelligent workpiece welding seam recognition and locating are achieved, and a robot is guided to conduct intelligent welding operation.

Application Domain

Programme-controlled manipulatorWelding/cutting auxillary devices +3

Technology Topic

Industrial engineeringImaging processing +11

Image

Examples

Experimental program(3)

Example Embodiment

[0041] Embodiment 1: A control method for a visual welding process system, including vision-guided welding seam locating technology, redundant laser positioning detection technology, welding seam tracking and laser deviation correction technology, robot autonomous motion planning technology, autonomous adaptive welding technology and Weld defect identification technology;

[0042]Vision-guided welding seam locating technology, including visual image acquisition and preprocessing, welding seam identification, key point position coordinate calculation, welding seam teaching information and robot automatic path planning, through image processing of the weld seam image of the workpiece to be welded, Calculate the position coordinates of the start and end points of the welding seam to form the teaching information of the welding seam;

[0043] Redundant laser positioning detection technology, through the establishment of visual recognition and positioning algorithms based on deep learning models (deep belief network (DBN), convolutional neural network (CNN) and recurrent neural network multilayer perceptron (RNN)), combined with boundary pixels Detection algorithm, and adopts dual-camera vision system to realize the detection, identification and precise positioning of welding workpiece and welding seam trajectory;

[0044] Weld seam tracking and laser deviation correction technology, integrating vision sensor and laser sensor in the robot welding system, realizes welding seam tracking and laser deviation correction in the process of multi-pass welding of medium and thick plates, and realizes real-time adjustment through welding seam tracking through visual sensing The motion state of the welding end, the welding seam rectification is realized by the laser sensor, the adaptability of the robot to the change of working conditions during the welding process is improved, and the motion accuracy and welding quality of the welding process are improved;

[0045] Robot autonomous motion planning technology, on the basis of visual aided teaching information, human-computer interaction information and welding process information, performs fast path planning of robot host computer, forms complete robot welding planning information, and establishes welding work instruction library;

[0046] Self-adaptive welding technology can adapt to defects such as oil stains and large gaps on the plate to ensure welding quality;

[0047] Weld defect identification technology, collect images of welded seams, develop porosity and pit weld defect identification and positioning algorithm based on deep learning semantic segmentation function, and complete weld quality inspection through deep learning target detection method after collecting workpiece images .

[0048] The vision-guided welding seam locating technology processes the spatial position coordinates of key points as robot teaching information through the host computer, and guides the end of the robot welding torch to move to the position to be welded, so that the welding robot has the ability to adapt to the spatial position of the workpiece to be welded, Realize welding robot machine vision intelligent welding seam locating.

[0049] The redundant laser positioning detection technology, the point set registration method ICP algorithm based on contour features, realizes accurate matching and rapid identification of contour features or point sets of three-dimensional elements, and uses iteration to calculate the correct correspondence step by step, multi-view registration , to realize the contour measurement and identification of robot welding workpiece.

[0050] The welding seam tracking and laser deviation correction technology, the integration of the vision sensor and the laser sensor, is specifically, the control robot scans the welding seam at a speed of 5mm/s and obtains the image of the structured light welding seam, combined with the coordinates of the robot tool point and the The processing result of the structured light weld image, the actual position of the weld is calculated. By analyzing the data of 10 groups of random welding seam tracking conditions, the system has high tracking accuracy in the X-axis and Y-axis directions, and the overall average error is within 0.5mm, while the tracking error in the Z-axis direction is larger, and the overall average error It is 0.95mm, which can meet the accuracy requirements of V-shaped fillet weld tracking of medium and heavy plates.

[0051] The robot autonomous motion planning technology, the robot autonomous motion planning establishes a welding work instruction library, and based on the autonomous motion planning results, by calling the corresponding instructions in the instruction library, the complete robot motion control instructions and job execution instructions are automatically generated to realize the robot fast path. Planning, teaching and job execution.

[0052] The self-adaptive welding technology reduces the welding spatter rate by 10%, improves the quality of solder joints by 15%, and reduces energy consumption by 20% compared to constant current.

[0053] The defect identification technology of the welding seam locates and identifies the welding bead and the welding seam with serious unqualified defects.

[0054] The welding robot remote operation and maintenance technology of the cloud platform can collect and monitor industrial robot working conditions, faults and task information through the cloud service platform, which can improve the safe operation rate of equipment and provide data support for predictive fault diagnosis, thereby avoiding unplanned failures. The waste of manpower and financial resources caused by downtime, collect and monitor the running status of the robot, and grasp the statistical information of equipment startup rate, operation rate, utilization rate, failure rate and OEE in real time, so as to adjust the utilization rate of robots in the workshop more reasonably, thereby improving the Productivity and return on assets, realize the remote automatic update and upgrade of the robot's firmware, realize the remote networked collaborative service of service personnel, realize localized personnel scheduling, and reduce the operating cost of technical services.

Example Embodiment

[0055] Embodiment 2: A control method for a visual welding process system, including vision-guided welding seam locating technology, redundant laser positioning detection technology, welding seam tracking and laser deviation correction technology, robot autonomous motion planning technology, autonomous adaptive welding technology and Weld defect identification technology;

[0056] Vision-guided welding seam locating technology, including visual image acquisition and preprocessing, welding seam identification, key point position coordinate calculation, welding seam teaching information and robot automatic path planning, through image processing of the weld seam image of the workpiece to be welded, Calculate the position coordinates of the start and end points of the welding seam to form the teaching information of the welding seam;

[0057] Redundant laser positioning detection technology, through the establishment of visual recognition and positioning algorithms based on deep learning models (deep belief network (DBN), convolutional neural network (CNN) and recurrent neural network multilayer perceptron (RNN)), combined with boundary pixels Detection algorithm, and adopts dual-camera vision system to realize the detection, identification and precise positioning of welding workpiece and welding seam trajectory;

[0058] Weld seam tracking and laser deviation correction technology, integrating vision sensor and laser sensor in the robot welding system, realizes welding seam tracking and laser deviation correction in the process of multi-pass welding of medium and thick plates, and realizes real-time adjustment through welding seam tracking through visual sensing The motion state of the welding end, the welding seam rectification is realized by the laser sensor, the adaptability of the robot to the change of working conditions during the welding process is improved, and the motion accuracy and welding quality of the welding process are improved;

[0059] Robot autonomous motion planning technology, on the basis of visual aided teaching information, human-computer interaction information and welding process information, performs fast path planning of robot host computer, forms complete robot welding planning information, and establishes welding work instruction library;

[0060] Self-adaptive welding technology can adapt to defects such as oil stains and large gaps on the plate to ensure welding quality.

[0061] The vision-guided welding seam locating technology processes the spatial position coordinates of key points as robot teaching information through the host computer, and guides the end of the robot welding torch to move to the position to be welded, so that the welding robot has the ability to adapt to the spatial position of the workpiece to be welded, Realize welding robot machine vision intelligent welding seam locating.

[0062] The redundant laser positioning detection technology, the point set registration method ICP algorithm based on contour features, realizes accurate matching and rapid identification of contour features or point sets of three-dimensional elements, and uses iteration to calculate the correct correspondence step by step, multi-view registration , to realize the contour measurement and identification of robot welding workpiece.

[0063] The welding seam tracking and laser deviation correction technology, the integration of the vision sensor and the laser sensor, is specifically, the control robot scans the welding seam at a speed of 10mm/s and obtains the image of the structured light welding seam, and combines the coordinates of the robot tool point with the coordinates of the robot tool point each time an image is taken. The processing result of the structured light weld image, the actual position of the weld is calculated. By analyzing the data of 10 groups of random welding seam tracking conditions, the system has high tracking accuracy in the X-axis and Y-axis directions, and the overall average error is within 0.4mm, while the tracking error in the Z-axis direction is larger, and the overall average error It is 0.95mm, which can meet the accuracy requirements of V-shaped fillet weld tracking of medium and heavy plates.

[0064] The robot autonomous motion planning technology, the robot autonomous motion planning establishes a welding work instruction library, and based on the autonomous motion planning results, by calling the corresponding instructions in the instruction library, the complete robot motion control instructions and job execution instructions are automatically generated to realize the robot fast path. Planning, teaching and job execution.

[0065] The self-adaptive welding technology reduces the welding spatter rate by 10%, improves the quality of solder joints by 15%, and reduces energy consumption by 20% compared to constant current.

Example Embodiment

[0066] Embodiment 3: A control method for a visual welding process system, including vision-guided welding seam locating technology, redundant laser positioning detection technology, welding seam tracking and laser deviation correction technology, robot autonomous motion planning technology, autonomous adaptive welding technology and Weld defect identification technology;

[0067] Vision-guided welding seam locating technology, including visual image acquisition and preprocessing, welding seam identification, key point position coordinate calculation, welding seam teaching information and robot automatic path planning, through image processing of the weld seam image of the workpiece to be welded, Calculate the position coordinates of the start and end points of the welding seam to form the teaching information of the welding seam;

[0068] Redundant laser positioning detection technology, through the establishment of visual recognition and positioning algorithms based on deep learning models (deep belief network (DBN), convolutional neural network (CNN) and recurrent neural network multilayer perceptron (RNN)), combined with boundary pixels Detection algorithm, and adopts dual-camera vision system to realize the detection, identification and precise positioning of welding workpiece and welding seam trajectory;

[0069] Weld seam tracking and laser deviation correction technology, integrating vision sensor and laser sensor in the robot welding system, realizes welding seam tracking and laser deviation correction in the process of multi-pass welding of medium and thick plates, and realizes real-time adjustment through welding seam tracking through visual sensing The motion state of the welding end, the welding seam rectification is realized by the laser sensor, the adaptability of the robot to the change of working conditions during the welding process is improved, and the motion accuracy and welding quality of the welding process are improved;

[0070] Robot autonomous motion planning technology, on the basis of visual aided teaching information, human-computer interaction information and welding process information, performs fast path planning of robot host computer, forms complete robot welding planning information, and establishes welding work instruction library;

[0071] Self-adaptive welding technology can adapt to defects such as oil stains and large gaps on the plate to ensure welding quality;

[0072] Weld defect identification technology, collect images of welded seams, develop porosity and pit weld defect identification and positioning algorithm based on deep learning semantic segmentation function, and complete weld quality inspection through deep learning target detection method after collecting workpiece images .

[0073] The vision-guided welding seam locating technology processes the spatial position coordinates of key points as robot teaching information through the host computer, and guides the end of the robot welding torch to move to the position to be welded, so that the welding robot has the ability to adapt to the spatial position of the workpiece to be welded, Realize welding robot machine vision intelligent welding seam locating.

[0074] The redundant laser positioning detection technology, the point set registration method ICP algorithm based on contour features, realizes accurate matching and rapid identification of contour features or point sets of three-dimensional elements, and uses iteration to calculate the correct correspondence step by step, multi-view registration , to realize the contour measurement and identification of robot welding workpiece.

[0075] The welding seam tracking and laser deviation correction technology, the integrated vision sensor and the laser sensor, are specifically controlled to scan the welding seam at a speed of 8mm/s and obtain the image of the structured light welding seam, combined with the coordinates of the robot tool point and the The processing result of the structured light weld image, the actual position of the weld is calculated. By analyzing the data of 15 groups of random welding seam tracking, the system has high tracking accuracy in the X-axis and Y-axis directions, and the overall average error is within 0.5mm, while the tracking error in the Z-axis direction is larger, and the overall average error It is 0.95mm, which can meet the accuracy requirements of V-shaped fillet weld tracking of medium and heavy plates.

[0076]The robot autonomous motion planning technology, the robot autonomous motion planning establishes a welding work instruction library, and based on the autonomous motion planning results, by calling the corresponding instructions in the instruction library, the complete robot motion control instructions and job execution instructions are automatically generated to realize the robot fast path. Planning, teaching and job execution.

[0077] The self-adaptive welding technology reduces the welding spatter rate by 10%, improves the quality of solder joints by 15%, reduces energy consumption by 20% compared to constant current, and saves time by 30%.

[0078] The defect identification technology of the welding seam locates and identifies the welding bead and the welding seam with serious unqualified defects.

PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

Classification and recommendation of technical efficacy words

Ensure consistency

Control method of visual welding process system

AI-Extracted Technical Summary

Problems solved by technology

Method used

Abstract

Image

Examples

Example Embodiment

Example Embodiment

Example Embodiment

PUM

Description & Claims & Application Information

Similar technology patents

Audience rating survey system and audience rating information searching method based on audio frequency characteristics

Individualized clothing template implementation method and system

High-bandwidth large-scale MIMO channel simulation method and device

A power module structure capable of improving welding quality

Pixel circuit, pixel driving method and display device

Classification and recommendation of technical efficacy words

Soybean milk machine with small-space crushing cover

Testing system for mobile terminal and testing method thereof

3D printing device and method

Automatic test method based on case library

Digital predistortion processing method and digital predistortion processing system