Rapid visual positioning method for industrial robot

Abstract

The invention discloses a rapid visual positioning method for an industrial robot, and relates to the technical field of rapid positioning of industrial robots, and the method comprises the steps: collecting depth image data and color image data of a target object through an image collection module, and extracting and preprocessing the collected depth image data and color image data through an image processing module; a target detection and recognition module is adopted to analyze the processed depth image data and color image data, feature information of a target object needing to be processed is recognized, and based on the target detection and recognition module, comprehensive adjustment of rapid visual positioning is carried out through a comprehensive positioning adjustment module. And the comprehensive positioning and adjusting module comprises a depth image visual coarse positioning unit, a color image feature recognition and adjusting unit and a comprehensive positioning and adjusting unit, and is used for uploading collected image data to an industrial robot control system terminal.

Description

Technical Field

[0001] The invention relates to the technical field of rapid positioning of industrial robots, and in particular to a rapid visual positioning method for industrial robots. Background Art

[0002] In the field of bearing processing, the application of industrial robots is becoming more and more extensive. Especially under the requirements of high-precision and high-efficiency processing, fast visual positioning technology is particularly important. This technology combines machine vision with industrial robots to achieve fast and accurate identification and positioning of precision parts such as bearings, thereby improving processing quality and production efficiency.

[0003] Traditional mechanical positioning methods are difficult to achieve micron-level or even nanometer-level positioning accuracy, which is particularly critical in the processing of precision parts such as bearings. Although the existing visual positioning technology has been improved, it is still affected by ambient light interference and camera resolution limitations, resulting in unstable positioning accuracy. In the bearing processing process, due to differences in workpiece size, shape, material, and fixture accuracy limitations, the position of the workpiece in each processing may be inconsistent, which increases the difficulty of visual positioning and easily leads to processing errors. In addition, in high-precision measurement, complex optical distortion calibration, projection distortion calibration, and object-image space calibration are required, and for non-planar objects, special calibration algorithms need to be designed. These calibration processes are cumbersome and prone to errors, affecting positioning accuracy. The stability and reliability of existing visual positioning technology are easily affected in complex and changeable production environments, including changes in lighting and temperature fluctuations, resulting in positioning failures or reduced accuracy. Summary of the invention

[0004] The purpose of the present invention is to provide a rapid visual positioning method for an industrial robot, which solves the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions, and the rapid visual positioning method is as follows: Data acquisition, collecting depth image data and color image data of the target object through the image acquisition module; Data processing: extract and preprocess the collected depth image data and color image data through the image processing module, and use the target detection and recognition module to analyze the processed depth image data and color image data to identify the feature information of the target object that needs to be processed; Data positioning and adjustment, based on the target detection and recognition module, a comprehensive positioning adjustment module is used to perform comprehensive adjustment of rapid visual positioning, wherein the comprehensive positioning adjustment module includes a depth image visual coarse positioning unit, a color image feature recognition and adjustment unit, and a comprehensive positioning adjustment unit; Data upload: upload the collected image data to the industrial robot control system terminal; The equipment used in the image acquisition module includes a visual sensor, which includes a depth camera and a color camera for collecting image data; The equipment used in the image processing module includes image processing hardware, and the image processing hardware includes a GPU and a DSP, which are used to accelerate the image processing process.

[0006] Optionally, the calculation formula of the depth image visual coarse positioning unit is as follows: ; in: G is the rough positioning distance value of the target workpiece; X o is the X coordinate value of the visual center, Y o is the Y coordinate value of the visual center, (X o , Y o ) reflects the coordinate position of the center of the visual sensor's field of view; X g is the X coordinate value of the target workpiece, Y g is the Y coordinate value of the target workpiece, (X g , Y g ) reflects the coordinate position of the center of the target workpiece; The positioning process of the depth image visual coarse positioning unit is as follows: The depth image data extracted by the image processing module is used to obtain the distance between the center of the output target workpiece and the center of the visual sensor field of view, that is, the rough positioning distance value G of the target workpiece. The depth image data includes the X coordinate value X of the visual center. o , Visual center Y coordinate value Y o 、Target workpiece X coordinate value X g and the target workpiece Y coordinate value Y g .

[0007] Optionally, the calculation formula of the color image feature recognition and adjustment unit is as follows: ; in: To adjust the angle value; X m is the X coordinate value of the robot end effector, Y m is the Y coordinate value of the robot end effector, (X m , Y m ) reflects the coordinate position of the end effector of the industrial robot; PT is the deviation angle adjustment value, and PT reflects the degree of adjustment of the deviation angle; MR is the target hole radius value; BR is the standard circular hole radius value; The recognition and adjustment process of the color image feature recognition and adjustment unit is as follows: Based on the target workpiece position (X) output by the depth image visual rough positioning unit g , Y g ), combined with the characteristic information of the target object identified and analyzed in the color image data, the characteristic information of the target object includes the target circular hole radius value MR and the current position of the robot end effector (X m , Y m ), and the degree of angle deviation under the standard circular hole radius value BR standard, output the angle that the industrial robot end effector needs to adjust, that is, the adjustment angle value .

[0008] Optionally, the calculation formula of the comprehensive positioning adjustment unit is as follows: ; in: X t is the X coordinate adjustment value, Y t is the Y coordinate adjustment value, (X t , Y t ) reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions; WT is the depth adjustment value; S is the processing depth value; The adjustment process of the comprehensive positioning adjustment unit is as follows: Comprehensively consider the target workpiece position (X g , Y g ), the adjustment angle value output by the color image feature recognition and adjustment unit , and the processing depth value S, the multi-dimensional output reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions, that is, (X t , Y t ).

[0009] Optionally, based on the X coordinate adjustment value X output by the comprehensive positioning adjustment unit t , Y coordinate adjustment value Y t , that is (X t , Y t ), the influence on the rapid positioning of the depth image visual coarse positioning unit and the color image feature recognition and adjustment unit is as follows: S1, due to (X t , Y t ) reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions; S2, therefore, is substituted into the color image feature recognition and adjustment unit, and a new X coordinate value X of the robot end effector is output. nnew and the new robot end effector Y coordinate value Y nnew , that is (X nnew , Y nnew ): X nnew =X m +X t ; Y nnew =Y m +YY t ; S3, then, compare (X nnew , Y nnew ) and (X t , Y t ) is the distance difference: 1. If (X nnew , Y nnew ) and (X t , Y t ) coordinate values ​​are consistent, the industrial robot can achieve fast positioning consistency; 2. If (X nnew , Y nnew ) and (X t , Y t ) has a difference in the coordinate value, further positioning adjustment is performed through the comprehensive positioning adjustment module until (X nnew , Y nnew ) and (X t , Y t ) keep the coordinate values ​​consistent.

[0010] Optionally, the integrated positioning adjustment unit and X t With Y t Coordinate transformation of .

[0011] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention adopts high-resolution, high-sensitivity industrial cameras and lenses to improve the quality of image acquisition and optimize the visual system, and designs a stable light source system to reduce ambient light interference, ensure the consistency of lighting conditions, and develop or adopt advanced image processing algorithms and positioning algorithms to improve the accuracy of target recognition and positioning.

[0012] 2. The present invention can combine mechanical positioning with visual positioning, and automatically adjust the position of the robot's end effector through an intelligent algorithm to adapt to the inconsistency of the workpiece position, forming an intelligent positioning strategy, and monitoring the workpiece position changes in real time during the processing process. The robot's motion trajectory is adjusted through a dynamic compensation algorithm, which helps to ensure processing accuracy and achieve dynamic compensation.

[0013] The present invention develops an automated calibration tool to reduce manual intervention, improve calibration efficiency and accuracy, and studies a universal calibration algorithm applicable to a variety of shapes and materials, which helps to reduce calibration complexity.

[0014] The present invention introduces environmental perception technology to monitor changes in the production environment in real time, automatically adjusts visual system parameters to adapt to environmental changes, and considers the influence of environmental factors in the design of the visual system and algorithm, thereby improving the robustness and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of the method for rapid visual positioning of industrial robots; Figure 2 This is a schematic diagram of the equipment used in this image acquisition module.

[0016] Figure 3 It is a structural schematic diagram of the comprehensive positioning adjustment module of the present invention; Figure 4 This is a schematic diagram of an algorithm formula for rapid positioning of the coordinate axes of the new robot end effector of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] The present industrial robot rapid visual positioning method is different from the existing industrial robot rapid visual positioning method. The existing industrial robot rapid visual positioning method uses the traditional mechanical positioning method, and the traditional mechanical positioning method is difficult to achieve micron-level or even nanometer-level positioning accuracy, resulting in unstable positioning accuracy. The position of the workpiece in each processing may be inconsistent, which increases the difficulty of visual positioning and easily leads to processing errors. In high-precision measurement, complex calibration and calibration algorithms are required, but the calibration process is cumbersome and prone to errors, affecting the positioning accuracy. The stability and reliability of the existing visual positioning technology are easily affected in complex and changeable production environments, resulting in Positioning failure or reduced accuracy. This algorithm unit adopts advanced image processing algorithms and positioning algorithms to improve the accuracy of target recognition and positioning. It can adjust the robot's motion trajectory through dynamic compensation algorithms, which helps to ensure processing accuracy and achieve dynamic compensation. It can develop automated calibration tools to reduce manual intervention and improve calibration efficiency and accuracy. Research on universal calibration algorithms suitable for a variety of shapes and materials will help reduce calibration complexity. It also introduces environmental perception technology to monitor production environment changes in real time and automatically adjust visual system parameters to adapt to environmental changes. The impact of environmental factors is considered in the design of the visual system and algorithm to improve the robustness and stability of the system.

[0019] For examples, see Figures 1 to 4 , this implementation provides a fast visual positioning method for an industrial robot, the fast visual positioning method is as follows: Data acquisition, collecting depth image data and color image data of the target object through the image acquisition module; Data processing: extract and preprocess the collected depth image data and color image data through the image processing module, and use the target detection and recognition module to analyze the processed depth image data and color image data to identify the feature information of the target object that needs to be processed; Data positioning and adjustment, based on the target detection and recognition module, the comprehensive positioning adjustment module performs comprehensive adjustment of rapid visual positioning. The comprehensive positioning adjustment module includes a depth image visual coarse positioning unit, a color image feature recognition and adjustment unit, and a comprehensive positioning adjustment unit; Data upload: upload the collected image data to the industrial robot control system terminal; The equipment used in the image acquisition module includes a visual sensor, which includes a depth camera and a color camera for collecting image data; The equipment used in the image processing module includes image processing hardware, which includes GPU and DSP, which are used to accelerate the image processing process.

[0020] The depth image data collected by the visual sensor depth camera is also called the distance image, which records the distance information from each point in the scene to the camera plane. The color image data collected by the visual sensor color camera is an image composed of three color channels: red, green and blue. Each channel contains 8 bits of data, so it can represent 256 color intensities. The color RGB image can provide richer visual information, including color, brightness, and contrast. The conversion and processing from depth image data to color image data is a complex process involving multiple steps and technologies. It realizes information fusion or conversion through image fusion, mapping of depth information to color space, or color image reconstruction assisted by depth image. In this process, a variety of image processing technologies are used to optimize image quality and extract useful information.

[0021] In this embodiment, the system cooperates with the three algorithm units to improve the positioning accuracy of the industrial robot fast visual positioning method, solve the problem of workpiece position inconsistency, and enhance environmental adaptability on the basis of simplifying the calibration process. and (X t , Y t ) The three operation results can help improve the quality and efficiency of bearing processing and promote the intelligent development of manufacturing industry. G is the rough positioning distance value of the target workpiece. It is mainly used to obtain the distance between the center of the target workpiece and the center of the camera depth sensor field of view through depth image data. For target workpieces such as bearings, this is the first step to achieve precise positioning. The Euclidean distance formula is used to calculate the straight-line distance between two points, that is, the distance between the target object and the camera, which provides the approximate position information of the target object in space. To adjust the angle value, based on the known position of the target object, combined with the features in the color image data, including the radius of the target hole and the current position of the robot's end effector, the robot needs to adjust the angle value. , so that the robot end effector can more accurately point to the processing position of the target object, so as to more accurately align the target, X t is the X coordinate adjustment value, Y t is the Y coordinate adjustment value, (X t , Y t ) combines G and , and the preset distance from the robot end effector to the target plane, output the precise adjustment amount of the robot end effector in the X and Y directions (X t , Y t ), and taking into account the fine-tuning in the depth direction, the robot end effector is accurately positioned in three-dimensional space, and (X t , Y t ) can also affect the feedback to G and The calculation makes the three algorithms of this system have a high correlation and entanglement, so that the overall algorithm system can perform automatic feedback and optimization according to the actual situation to be closer to reality.

[0022] See also Figures 1 to 4 , the calculation formula of the depth image visual coarse positioning unit is as follows: ; in: G is the rough positioning distance value of the target workpiece; X o is the X coordinate value of the visual center, Y o is the Y coordinate value of the visual center, (X o , Y o ) reflects the coordinate position of the center of the visual sensor's field of view; X g is the X coordinate value of the target workpiece, Y g is the Y coordinate value of the target workpiece, (X g , Y g ) reflects the coordinate position of the center of the target workpiece; The positioning process of the depth image visual coarse positioning unit is as follows: The depth image data extracted by the image processing module is used to obtain the distance between the center of the output target workpiece and the center of the visual sensor field of view, that is, the rough positioning distance value G of the target workpiece. The depth image data includes the X coordinate value X of the visual center. o , Visual center Y coordinate value Y o 、Target workpiece X coordinate value X g and the target workpiece Y coordinate value Y g .

[0023] In this embodiment: First, in this algorithm unit, the visual center X coordinate value X o The Y coordinate value v of the visual center is the fixed coordinate detected by the visual sensor at the center of the field of view, which is usually determined when the system is initialized. The X coordinate value X g and the target workpiece Y coordinate value Y g The center coordinates of the target object are identified from the depth image data through image processing algorithms, including edge detection and contour extraction algorithms, and the coordinate position of the center of the target workpiece (X g , Y g ) and the coordinate position of the center of the visual sensor’s field of view (X o , Y o) can quickly determine the coarse positioning distance value G of the target workpiece, which is the first step to achieve precise positioning and provides important basic data for subsequent calculations. By measuring the distance, the spatial range that needs to be searched in subsequent processing can be significantly reduced, thereby improving the overall positioning efficiency and reducing the search space.

[0024] The deep image vision coarse positioning unit uses simple geometric principles to directly calculate the distance through coordinate differences, avoiding complex image processing and feature matching processes, ensuring the efficiency of fast visual positioning, and because the calculation process is simple, the results can be obtained quickly, meeting the real-time requirements of industrial robots.

[0025] See also Figures 1 to 4 , the calculation formula of the color image feature recognition and adjustment unit is as follows: ; in: To adjust the angle value; X m is the X coordinate value of the robot end effector, Y m is the Y coordinate value of the robot end effector, (X m , Y m ) reflects the coordinate position of the end effector of the industrial robot; PT is the deviation angle adjustment value, and PT reflects the degree of adjustment of the deviation angle; MR is the target hole radius value; BR is the standard circular hole radius value; The recognition and adjustment process of the color image feature recognition and adjustment unit is as follows: Based on the target workpiece position (X) output by the depth image visual rough positioning unit g , Y g ), combined with the characteristic information of the target object identified and analyzed in the color image data, the characteristic information of the target object includes the target circular hole radius value MR and the current position of the robot end effector (X m , Y m ), and the degree of angle deviation under the standard circular hole radius value BR standard, output the angle that the industrial robot end effector needs to adjust, that is, the adjustment angle value .

[0026] In this embodiment, first X m is the X and Y coordinates of the robot end effector mThe Y coordinate value of the robot end effector is the current coordinate of the robot end effector, which can be obtained through the robot control system. The target hole radius value MR is the target hole radius identified by the color image processing algorithm. The standard hole radius value BR is the preset average and standard hole radius, which is used to compare and adjust with the actual radius. The angle is adjusted according to the depth distance and the deviation of the target hole radius from the average radius, where is a constant used to adjust the sensitivity of angle changes. The inverse tangent function is used to calculate the relative angle between the target object and the robot end effector, and fine-tuning is performed based on the target workpiece coarse positioning distance value G and the hole radius deviation (MR-BR). The output is the adjusted angle value. ; This unit improves the accuracy and robustness of positioning by combining the position information in the depth image with the feature information in the color image, achieving the effect of feature fusion.

[0027] During the recognition and adjustment process of this unit, the feature information in the color image data is used for recognition, including shape, color, and texture, to achieve the purpose of feature recognition, thereby improving the accuracy and reliability of target object detection; and through the calculated angle adjustment amount, the robot is guided to make precise angle adjustments to ensure that the end effector can accurately point to the target position.

[0028] See also Figures 1 to 4 , the calculation formula of the comprehensive positioning adjustment unit is as follows: ; in: X t is the X coordinate adjustment value, Y t is the Y coordinate adjustment value, (X t , Y t ) reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions; WT is the depth adjustment value; S is the processing depth value; The adjustment process of the comprehensive positioning adjustment unit is as follows: Comprehensively consider the target workpiece position (X g , Y g ), the adjustment angle value output by the color image feature recognition and adjustment unit , and the processing depth value S, the multi-dimensional output reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions, that is, (X t , Y t ).

[0029] In this embodiment, the algorithm unit first calculates Xt With Y t The coordinate transformation is the adjustment angle value output by the color image feature recognition and adjustment unit On this basis, the coordinate transformation is performed by using the rotation matrix or simple trigonometric function relationship on the two-dimensional plane. Specifically, and X g -X m and Y g -Y m Transform to get the adjusted coordinate difference and WT is the depth adjustment value, where S is the processing depth value, which is the preset distance from the robot end effector to the target plane and is set according to the specific processing task. It is derived based on the inverse calculation of the Pythagorean theorem and is used to make fine adjustments in the processing depth direction to ensure that the end effector approaches the target at the correct depth; This algorithm unit combines the processing depth S, adjustment angle value and the rough positioning distance value of the target workpiece, calculate the precise adjustment amount of the robot end effector in the X and Y directions (X t , Y t ), these adjustments are used to guide the robot to perform the final precise movement. It is worth noting that by calculating the adjustment amounts and controlling the robot to move, a closed-loop control system can be formed. During the movement process, the position of the robot can be monitored and adjusted in real time to ensure that it can finally reach the predetermined position accurately.

[0030] In actual application, the comprehensive positioning adjustment unit takes into account multiple factors including depth, angle, and preset distance when calculating the adjustment amount, ensuring the accuracy and reliability of the adjustment amount; through the closed-loop control system, the position of the robot can be monitored and adjusted in real time, improving the positioning accuracy and stability.

[0031] See also Figures 1 to 4 , based on the X coordinate adjustment value X output by the comprehensive positioning adjustment unit t , Y coordinate adjustment value Y t , that is (X t , Y t ), the influence on the rapid positioning of the depth image visual coarse positioning unit and the color image feature recognition and adjustment unit is as follows: S1, due to (X t , Y t ) reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions; S2, therefore, is substituted into the color image feature recognition and adjustment unit, and a new X coordinate value X of the robot end effector is output. nnew and the new robot end effector Y coordinate value Y nnew , that is (X nnew , Y nnew ): X nnew =X m +X t ; Y nnew =Y m +YY t ; S3, then, compare (X nnew , Y nnew ) and (X t , Y t ) is the distance difference: 1. If (X nnew , Y nnew ) and (X t , Y t ) coordinate values ​​are consistent, the industrial robot can achieve fast positioning consistency; 2. If (X nnew , Y nnew ) and (X t , Y t ) has a difference in the coordinate value, further positioning adjustment is performed through the comprehensive positioning adjustment module until (X nnew , Y nnew ) and (X t , Y t ) keep the coordinate values ​​consistent.

[0032] In this embodiment, the algorithm unit is based on the X coordinate adjustment value X output by the comprehensive positioning adjustment unit. t , Y coordinate adjustment value Y t , that is (X t , Y t ), so according to (X t , Y t ) is the result value of the robot's current end effector position (X m , Y m ), the new X coordinate value X of the robot end effector is obtained nnew and the Y coordinate value of the robot end effector Y nnew , that is (X nnew , Y nnew ), (X nnew , Y nnew The specific output of ) is obtained through "X nnew =X m+X and Y nnew =Y m +YY t " is calculated, and the rapid visual positioning of the industrial robot is realized. It is worth noting that after the rapid visual positioning, (X nnew , Y nnew ) and (X t , Y t ) coordinate positions are compared. If the coordinate positions of the two are consistent, the industrial robot has achieved rapid visual positioning; if (X nnew , Y nnew ) and (X t , Y t )'s coordinate position has any deviation, whether near or far, it is considered that the industrial robot has achieved but has not achieved rapid visual positioning. It is necessary to substitute the deviation into the comprehensive positioning adjustment module for comprehensive positioning adjustment until (X nnew , Y nnew ) and (X t , Y t ) to achieve consistency in coordinate positions, thereby improving positioning accuracy and stability, achieving closed-loop control, being able to automatically adjust and adapt to changes in the external environment, and also accelerating processing speed and improving production efficiency.

[0033] In summary, in the specific implementation process, these three algorithm formulas together constitute an efficient and accurate industrial robot visual positioning system, which not only considers the position information of the target object (through depth image data), but also combines the specific characteristics of the target object (including the radius of the circular hole, through the color image), as well as the current position of the robot end effector and the preset processing parameters (such as distance). Through loop iteration and closed-loop control, the system can adjust the position of the robot end effector in real time to ensure processing accuracy and efficiency.

[0034] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for rapid visual positioning of an industrial robot, characterized in that: The fast visual positioning method is as follows: Data acquisition, collecting depth image data and color image data of the target object through the image acquisition module; Data processing: extract and preprocess the collected depth image data and color image data through the image processing module, and use the target detection and recognition module to analyze the processed depth image data and color image data to identify the feature information of the target object that needs to be processed; Data positioning and adjustment, based on the target detection and recognition module, the comprehensive positioning adjustment module performs comprehensive adjustment of rapid visual positioning. The comprehensive positioning adjustment module includes a depth image visual coarse positioning unit, a color image feature recognition and adjustment unit, and a comprehensive positioning adjustment unit; Data upload: upload the collected image data to the industrial robot control system terminal.

2. The method for rapid visual positioning of an industrial robot according to claim 1, characterized in that: The equipment used in the image acquisition module includes a visual sensor, which includes a depth camera and a color camera for collecting image data; The equipment used in the image processing module includes image processing hardware, and the image processing hardware includes a GPU and a DSP, which are used to accelerate the image processing process.

3. The method for rapid visual positioning of an industrial robot according to claim 2, characterized in that: The calculation formula of the depth image visual coarse positioning unit is as follows: ； in: G is the rough positioning distance value of the target workpiece; X o is the X coordinate value of the visual center, Y o is the Y coordinate value of the visual center, (X o , Y o ) reflects the coordinate position of the center of the visual sensor's field of view; X g is the X coordinate value of the target workpiece, Y g is the Y coordinate value of the target workpiece, (X g , Y g ) reflects the coordinate position of the center of the target workpiece; The positioning process of the depth image visual coarse positioning unit is as follows: The depth image data extracted by the image processing module is used to obtain the distance between the center of the output target workpiece and the center of the visual sensor field of view, that is, the rough positioning distance value G of the target workpiece. The depth image data includes the X coordinate value X of the visual center. o , Visual center Y coordinate value Y o 、Target workpiece X coordinate value X g and the target workpiece Y coordinate value Y g .

4. The method for rapid visual positioning of an industrial robot according to claim 3, characterized in that: The calculation formula of the color image feature recognition and adjustment unit is as follows: ； in: To adjust the angle value; X m is the X coordinate value of the robot end effector, Y m is the Y coordinate value of the robot end effector, (X m , Y m ) reflects the coordinate position of the end effector of the industrial robot; PT is the deviation angle adjustment value, and PT reflects the degree of adjustment of the deviation angle; MR is the target hole radius value; BR is the standard circular hole radius value; The recognition and adjustment process of the color image feature recognition and adjustment unit is as follows: Based on the target workpiece position (X) output by the depth image visual rough positioning unit g , Y g ), combined with the characteristic information of the target object identified and analyzed in the color image data, the characteristic information of the target object includes the target circular hole radius value MR and the current position of the robot end effector (X m , Y m ), and the degree of angle deviation under the standard circular hole radius value BR standard, output the angle that the industrial robot end effector needs to adjust, that is, the adjustment angle value .

5. The method for rapid visual positioning of an industrial robot according to claim 4, characterized in that: The calculation formula of the comprehensive positioning adjustment unit is as follows: ； in: X t is the X coordinate adjustment value, Y t is the Y coordinate adjustment value, (X t , Y t ) reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions; WT is the depth adjustment value; S is the processing depth value; The adjustment process of the comprehensive positioning adjustment unit is as follows: Comprehensively consider the target workpiece position (X g , Y g ), the adjustment angle value output by the color image feature recognition and adjustment unit , and the processing depth value S, the multi-dimensional output reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions, that is, (X t , Y t ).

6. The method for rapid visual positioning of an industrial robot according to claim 5, characterized in that: Based on the X coordinate adjustment value X output by the comprehensive positioning adjustment unit t , Y coordinate adjustment value Y t , that is (X t , Y t ), the influence on the rapid positioning of the depth image visual coarse positioning unit and the color image feature recognition and adjustment unit is as follows: S1, due to (X t , Y t ) reflects the adjustment degree of the industrial robot end effector in the X and Y coordinate directions; S2, therefore, is substituted into the color image feature recognition and adjustment unit, and a new X coordinate value X of the robot end effector is output. nnew and the new robot end effector Y coordinate value Y nnew , that is (X nnew , Y nnew ): X nnew =X m +X t ; AND nnew =And m +YY t ; S3, then, compare (X nnew , Y nnew ) and (X t , Y t ) is the distance difference: If (X nnew , Y nnew ) and (X t , Y t ) coordinate values ​​are consistent, the industrial robot can achieve fast positioning consistency; If (X nnew , Y nnew ) and (X t , Y t ) has a difference in the coordinate value, further positioning adjustment is performed through the comprehensive positioning adjustment module until (X nnew , Y nnew ) and (X t , Y t ) keep the coordinate values ​​consistent.

7. The method for rapid visual positioning of an industrial robot according to claim 5, characterized in that: The integrated positioning adjustment unit and X t With Y t Coordinate transformation of .

Images