A robot vision gluing method, controller and system

Abstract

The application relates to a robot vision gluing method. The gluing method comprises the following steps: S1, acquiring a gluing product image captured by a vision camera; S2, acquiring a product contour corresponding to the image generated by an image matching processor; determining two positioning points on the product contour, and determining a product matching point according to the two positioning points; S3, acquiring product matching point image coordinates according to image coordinates, and calculating robot matching points corresponding to the product matching points in an image-robot coordinate calibration file according to the image coordinates; S4, establishing a matching point coordinate system in a three-dimensional digital model with the robot matching points as the origin by using a three-dimensional digital model of a product to be glued, and calculating product contour three-dimensional coordinates of the product contour in the matching point coordinate system; S5, offset correcting the product contour coordinates to obtain corrected product contour three-dimensional coordinates, and generating a gluing track; and S6, gluing according to the gluing track. The gluing method has low cost, is convenient to operate, and is suitable for gluing work on products with radii and curved surfaces.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gluing, in particular to a robot vision gluing method, a controller and a system. BACKGROUND

[0002] In recent years, with the rapid development of industrial automation equipment, using robots to replace manual gluing has become a trend. Robot gluing is fast, delicate and of good quality. Therefore, robot gluing is widely used in manufacturing industries such as the automobile industry, the hardware industry, the ceramic industry and the machinery industry. There are mainly the following methods for existing robot gluing.

[0003] 1. Three-dimensional vision method, which uses a three-dimensional vision camera to collect product point clouds, processes product point cloud images with software, calculates product gluing trajectories with trajectory calculation algorithms, and realizes automatic vision gluing. This method is complex to operate and costly, and is not suitable for products in low-end industries, nor is it suitable for dynamic following gluing on assembly lines.

[0004] 2. Image coordinate conversion method, which collects product color images through a color vision camera, extracts the edges of objects in the images, converts RGB images into grayscale images, vectorizes feature points through edge tracking algorithms, converts pixel coordinate points into robot coordinate points, and realizes automatic robot gluing. This method uses complex image processing techniques to obtain feature points of objects in images, and is complex to operate. Moreover, this method is only suitable for generating gluing trajectories for planar products, cannot generate products with height changes, has great limitations, and requires high configuration.

[0005] 3. Vision positioning product center point method, which takes the center point of a product image as the origin, calculates the offset of the remaining gluing points according to software, and generates the actual gluing trajectory. This method does not realize autonomous acquisition of product gluing trajectories by industrial robots, requires tedious robot gluing trajectory teaching operations for the first use, and cannot realize flexible production of robots. SUMMARY

[0006] The present application aims to overcome the shortcomings and deficiencies in the prior art and provide a robot vision gluing method.

[0007] The present application is implemented through the following technical solutions: a robot vision gluing method, comprising the following steps:

[0008] S1: obtaining an image of a product to be glued and image coordinates of each pixel point in the image captured by a vision camera;

[0009] S2: obtaining a product contour corresponding to the image of the product to be glued generated by the image matching processor; determining two positioning points P0 and P1 on the product contour, and determining a product matching point P according to the two positioning points P0 and P1 d ;

[0010] S3: obtaining image coordinate information of the product matching point P d according to the image coordinates, and calculating a robot matching point P d corresponding to the product matching point P robot in a robot tool coordinate system according to the image coordinate information by using a preset image-robot coordinate calibration file

[0011] S4: using a three-dimensional numerical model of the product to be glued, establishing a matching point coordinate system in the three-dimensional numerical model with the robot matching point P robot as a coordinate origin, and calculating product contour three-dimensional coordinates of the product contour of the product to be glued in the matching point coordinate system

[0012] S5: performing product contour point offset correction on the product contour three-dimensional coordinates to obtain corrected product contour three-dimensional coordinates, and generating a gluing track

[0013] S6: controlling the gluing jig to glue the product to be glued according to the gluing track.

[0014] Compared with the prior art, the robot vision gluing method does not need to design a special jig for different products, thereby avoiding high jig design cost; meanwhile, only a vision camera and a three-dimensional numerical model are used to realize vision positioning and gluing track generation of the product to be glued, which is more cost-saving, more convenient to operate, and more suitable for use of products in low-end industries; in addition, since the gluing point coordinates of the product to be glued are derived according to the three-dimensional numerical model, the method is suitable not only for gluing of planar products, but also for gluing of products with radii and curved surfaces.

[0015] Further, the image-robot coordinate calibration file is obtained by the following method:

[0016] placing a calibration board in the shooting range of the vision camera, shooting the calibration board by the vision camera to obtain two-dimensional image coordinates of the calibration board, and selecting nine feature point coordinates therefrom to record as an image coordinate matrix P image ,

[0017]

[0018] moving the mechanical arm of the robot so that the glue nozzle of the gluing jig is aligned with the calibration point on the calibration board, and recording the robot tool coordinates P robot-x, P robot-y ,

[0019]

[0020] coordinate conversion processing is performed on the robot tool coordinates:

[0021] convert the image coordinate matrix into a square matrix,

[0022] P image-temp = P image T · P image ,

[0023] respectively calculate the conversion matrix T result-x , T result-y ,

[0024] T result-y = P image-temp -1 · P image T · P robot-y ,

[0025] T result-x = image-temp -1 · P image T · P robot-x ,

[0026] splice the two matrices and denote them as T result ,

[0027] T result = [T result-x T result-y ],

[0028] save the coordinate conversion matrix T result in the image-robot coordinate calibration file, and obtain an image-robot coordinate calibration file for conversion between the image coordinate system and the robot tool coordinate system,

[0029] use the image-robot coordinate calibration file to calculate the robot matching point P d corresponding to the product matching point P robot ,

[0030] P robot = T resut · P image .

[0031] Further, in the step S2, the image matching processor adopts an image matching technology, the visual camera is in communication connection with the image matching processor; two positioning points P0, P1 are selected as two symmetrical points on the product to be glued, and the product matching point P d is the center point of the line connecting the two positioning points P0, P1.

[0032] Further, the step S4 is specifically as follows: the three-dimensional digital model of the product to be glued is directly imported into a three-dimensional digital model processor, the robot matching point P robot is set as the X, Y coordinate origin of the three-dimensional digital model of the product to be glued, and the Z coordinate origin is set as 0, with (P robot , 0) as the coordinate origin, the product contour three-dimensional coordinates of the product to be glued in the matching point coordinate system are recalculated, and the product contour three-dimensional coordinates of the product to be glued in the matching point coordinate system are exported.

[0033] The three-dimensional digital model processor can adopt any three-dimensional digital model processing technology when performing three-dimensional digital model processing.

[0034] Further, the step S5 is specifically as follows:

[0035] The product contour three-dimensional coordinates of the product to be glued in the matching point coordinate system exported in the step S5 are processed by a correction processor to obtain corrected product contour three-dimensional coordinates.

[0036] The corrected product contour three-dimensional coordinates of the product to be glued are exported and transmitted to the robot, the robot automatically generates a final gluing track, and stores the gluing track in the robot.

[0037] In the correction performed by the correction processor, the product contour points are offset corrected according to actual process parameters required for gluing.

[0038] Further, the step S6 is specifically as follows: the robot first judges whether the product to be glued has entered the working range of the robot, if it is judged that the product to be glued has entered the working range of the robot, the robot performs a gluing operation according to the stored gluing track.

[0039] Further, the robot judges whether the product to be glued has entered the working range of the robot according to the number of pulses transmitted by the encoder, and in the process of generating the image-robot coordinate calibration file, the encoder is used to record the working range of the robot, and the specific steps are as follows:

[0040] The encoder records the number n0 of pulses of the photographing position T of the visual camera,

[0041] The calibration board is moved to the working range of the robot. The encoder records the number of pulses at the start of following point F1 as n1 and the number of pulses at the stop of following point F2 as n2. n0, n1 and n2 are sent to the robot and recorded in the robot.

[0042] The movement distance of the product to be coated is calculated by the pulse number difference. The pulse number difference between the starting follow point F1 and the visual camera's shooting position T is denoted as Δn1 = n1 - n0, and the pulse number difference between the stopping follow point F2 and the current visual camera shooting position T is denoted as Δn2 = n2 - n0. If the pulse number difference of the encoder is within the range of Δn1 to Δn2, it indicates that the product to be coated has entered the working range of the robot.

[0043] The present invention also provides a robot vision adhesive application controller, including an image acquisition module, a product matching point acquisition module, a robot matching point acquisition module, a three-dimensional digital model coordinate acquisition module, an adhesive application trajectory generation module, and an adhesive application module;

[0044] The image acquisition module is used to acquire images of the product to be coated with adhesive taken by the vision camera and the image coordinates of each pixel in the image;

[0045] The product matching point acquisition module is used to acquire the product outline corresponding to the image of the product to be coated with adhesive generated by the image matching processor; determine two positioning points P0 and P1 on the product outline; and determine the product matching point P based on the two positioning points P0 and P1. d ;

[0046] The robot matching point acquisition module is used to acquire the product matching point P based on the image coordinates. d Based on the image coordinate information, the matching point P with the product is calculated using a preset image-robot coordinate calibration file. d The corresponding robot matching point P in the robot tool coordinate system robot ;

[0047] The three-dimensional digital model coordinate acquisition module is used to obtain robot matching point P in the three-dimensional digital model of the product to be coated. robot Establish a matching point coordinate system with the origin as the coordinate origin, and calculate the three-dimensional coordinates of the product outline of the product to be coated in the matching point coordinate system.

[0048] The adhesive application trajectory generation module is used to correct the product outline point offset in the three-dimensional coordinates of the product outline, obtain the corrected three-dimensional coordinates of the product outline, and generate the adhesive application trajectory.

[0049] The glue application module is used to control the glue application fixture to apply glue to the product to be glued according to the glue application trajectory.

[0050] The application also provides a robot visual gluing system, comprising a conveying belt, a visual camera, a robot, a gluing jig and a controller.

[0051] The conveying belt is arranged on a workbench, and the controller controls movement of the conveying belt.

[0052] The visual camera is arranged directly above the conveying belt, and a lens of the visual camera faces a conveying surface of the conveying belt. The visual camera is in communication connection with the controller, the controller controls photographing of the visual camera, and the visual camera transmits the photographed photos to the controller.

[0053] The robot comprises a robot controller and a mechanical arm. The robot controller is used to control movement of the mechanical arm, and the robot controller is in communication connection with the controller to receive gluing coordinate information transmitted by the controller.

[0054] The gluing jig is arranged on the mechanical arm, and the mechanical arm is used to control movement of the gluing jig.

[0055] The encoder is arranged on the conveying belt and moves synchronously with the conveying belt. The conveying belt is in communication connection with the robot controller to record a moving distance of the conveying belt.

[0056] The controller is the above-mentioned controller.

[0057] In order to better understand and implement, the application is described in detail below with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 FIG. 1 is a structural schematic diagram of a robot visual gluing system according to an embodiment of the application;

[0059] Figure 2 FIG. 2 is a structural schematic diagram of an inner part of a dark box in the robot visual gluing system according to an embodiment of the application;

[0060] Figure 3 FIG. 3 is a structural schematic diagram of a gluing jig in the robot visual gluing system according to an embodiment of the application;

[0061] Figure 4 FIG. 4 is a schematic diagram of a robot visual gluing controller according to an embodiment of the application;

[0062] Figure 5 FIG. 5 is a flow chart of a robot visual gluing method according to an embodiment of the application.

[0063] In the figure: 10-image acquisition module; 20-product matching point acquisition module; 30-robot matching point acquisition module; 40-three-dimensional model coordinate acquisition module; 50-glue coating track generation module; 60-glue coating module; 1-conveyor belt; 2-vision camera; 21-dark box; 3-robot; 4-glue coating jig; 5-encoder; 6-controller. DETAILED DESCRIPTION

[0064] The present application utilizes a vision camera and a three-dimensional model of a product to be glued to form a robot vision glue coating method and controller suitable for products with curvature and curved surfaces; and combines a glue coating device and a glue coating process to form a robot vision glue coating system.

[0065] Please refer to Figure 1 and Figure 2 , the robot vision glue coating system comprises a conveyor belt 1, a vision camera 2, a robot 3, a glue coating jig 4, an encoder 5 and a controller 6. The controller 6 controls the movement of the conveyor belt 1, the vision camera 2 is in communication connection with the controller 6, the controller 6 controls the vision camera 2 to take pictures to acquire images, the vision camera 2 then communicates the taken images back to the controller 6, the glue coating jig 4 is arranged on the robot 3, the controller 6 controls the movement of the robot 3, the robot 3 controls the movement of the glue coating jig 4, and the encoder 5 moves with the conveyor belt 1 and is in communication connection with the robot 3.

[0066] Specifically, the conveyor belt 1 conveys the product to be glued placed thereon to the position directly below the vision camera 2 for shooting. The conveyor belt 1 is arranged on a workbench, and is driven by a driving motor, and the controller 6 is in communication connection with the driving motor to control the movement of the conveyor belt 1.

[0067] The vision camera 2 shoots the product to be glued and transmits the shot images and image coordinate information of each pixel point in the images to the controller 6. The vision camera 2 is arranged directly above one end of the conveyor belt 1 with the lens of the vision camera 2 facing the conveying surface of the conveyor belt 1. The vision camera 2 is preferably an industrial area array camera, which can acquire two-dimensional image information of the product to be glued, so that the acquired images are more intuitive and the price is not high. In order to obtain clear and stable images of the product to be glued, a dark box 21 is arranged directly above one end of the conveyor belt 1, the vision camera 2 is fixedly installed in the dark box 21, and a light source is also arranged in the dark box 21, the light emitted by the light source irradiates on the product to be glued to provide light source for the vision camera 2 to shoot. Further, the light source is a line light source, and two line light sources provide clear, uniform and stable light source for the product to be glued.

[0068] The robot 3 moves according to the glue trajectory information of the product to be glued transmitted by the controller 6, and drives the glue fixture 4 mounted thereon to glue according to the glue trajectory. The robot 3 comprises a robot controller and a mechanical arm, the robot controller is used to control the movement of the mechanical arm, and the glue fixture 4 is arranged on the mechanical arm, and the mechanical arm is used to control the movement of the glue fixture 4.

[0069] The robot 3 is mounted on a base on one side of the conveying belt 1, so that the product to be glued transmitted by the conveying belt 1 can be transmitted into the working range of the mechanical arm of the robot 3. The movement of the mechanical arm is driven by a robot driving motor, the robot driving motor is in communication connection with the robot controller, and the robot controller controls the working posture of the mechanical arm of the robot 3 by controlling the robot driving motor.

[0070] Please refer to Figure 3 The glue fixture 4 is mounted at the end of the mechanical arm, and the glue fixture 4 is connected with the end shaft of the mechanical arm of the robot 3 through a connecting flange, so that the glue fixture 4 can realize 360° rotation, and through controlling the movement of the mechanical arm, the glue fixture 4 can realize up-down, forward-backward and left-right movement, so that the glue fixture 4 can realize multi-dimensional movement.

[0071] The glue fixture 4 comprises a glue gun, the glue gun comprises a glue dispensing valve, a glue cavity and a glue spraying part, an inlet hole is arranged between the glue dispensing valve and the glue cavity, two gas inlets are arranged on the glue dispensing valve, whether the inlet hole is filled with glue into the glue cavity is controlled through the glue dispensing valve, the glue spraying part is arranged at the end of the glue cavity, one end of the glue spraying part is connected with the glue cavity, and the other end is a glue outlet nozzle, the glue outlet nozzle sprays glue on the product to be glued. Since the glue fixture 4 can realize multi-dimensional movement, the glue gun arranged on the glue fixture 4 can also realize multi-dimensional movement, so that the glue gun can realize glue coating work of any trajectory.

[0072] The glue gun is controlled by a point glue controller, the point glue controller is connected with the robot controller in communication, and the robot controller controls the glue gun by controlling the point glue controller. Specifically, the air inlet of the point glue controller is connected with a 0.8Mpa air source pipeline, and the two air outlets are connected with the two air inlets of the point glue valve in a pipeline, for controlling the opening and closing of the glue; meanwhile, a glue supply pressure barrel for storing glue is also provided, which is also connected with the 0.8Mpa air source, and the glue supply pressure barrel is connected with the glue inlet hole through a plurality of glue supply pipes. The point glue controller controls whether the glue reaching the glue inlet hole enters the glue cavity by controlling whether the two air inlets of the point glue valve are supplied with air. After the glue enters the glue cavity, it is sprayed out from the glue outlet nozzle through the glue spraying part.

[0073] The encoder 5 is arranged on an encoder follower, the encoder follower is arranged on one side of the conveying belt 1 and can move synchronously with the conveying belt 1, so that the encoder 5 can move synchronously with the conveying belt 1, and the encoder 5 determines the movement of the conveying belt 1 through the recorded pulse number. The encoder 5 is connected with the robot controller in communication, and the encoder 5 transmits the recorded pulse signal to the robot controller, so that the robot 3 can accurately determine the position of the product to be glued and realize precise gluing.

[0074] Please refer to Figure 4 and Figure 5 , the controller 6 includes an image acquisition module 10, a product matching point acquisition module 20, a robot matching point acquisition module 30, a three-dimensional model coordinate acquisition module 40, a glue application trajectory generation module 50 and a glue application module 60.

[0075] The image acquisition module 10 is used to execute step S1: acquiring the image of the product to be glued taken by the vision camera and the image coordinates of each pixel point in the image.

[0076] The product matching point acquisition module 20 is used to execute step S2: acquiring the product contour corresponding to the image of the product to be glued generated by the image matching processor; determining two positioning points P0, P1 on the product contour, and determining the product matching point P according to the two positioning points P0, P1. d .

[0077] Specifically, the image acquired in step S1 is sent to the image matching processor, and the image is processed by the image matching processor to generate the product contour of the product to be glued; then the product contour of the product to be glued transmitted by the image matching processor is received, and two positioning points P0, P1 are determined on the product contour, and the product matching point P is determined according to the two positioning points P0, P1. d .

[0078] wherein two positioning points P0, P1 are selected as two symmetrical points on the product to be glued, and the product matching point P d is a center point of a line connecting the two positioning points P0, P1, and the product matching point P d is a center point of the product to be glued; the image matching processor adopts an existing image matching technology, and the image processor is arranged in the controller 6, and the visual camera 2 is in communication connection with the controller 6.

[0079] The robot matching point acquisition module 30 is configured to execute step S3: acquiring image coordinate information of the product matching point P d according to the image coordinates, and calculating a robot matching point P d corresponding to the product matching point P robot in a robot tool coordinate system according to the image-robot coordinate calibration file.

[0080] The image-robot coordinate calibration file is obtained by the following method.

[0081] Firstly, a calibration plate is placed on the conveying belt 1, and the calibration plate is transmitted to the shooting range of the visual camera 2, and the visual camera 2 shoots the calibration plate to obtain two-dimensional image coordinates of the calibration plate, and nine feature point coordinates are selected and recorded as an image coordinate matrix P image ,

[0082]

[0083] The mechanical arm of the robot 3 is moved so that the glue nozzle of the glue fixture 4 is aligned with the calibration point on the calibration plate, and the robot tool coordinates P robot-x of the calibration point at this time are recorded, P robot-y ,

[0084]

[0085] A coordinate converter is arranged in the controller 6 to perform coordinate conversion processing on the robot tool coordinates. Specifically, firstly, the image coordinate matrix is converted into a square matrix,

[0086] P image-temp = P image T ·P image ,

[0087] Then, the conversion matrices T result-x and T result-y of X and Y are calculated respectively,

[0088] T result-y = Pimage-temp -1 ·P image T ·P robot-y ,

[0089] T result-x = P image-temp -1 ·P image T ·P robot-x ,

[0090] Then the two matrices are spliced and recorded as T result ,

[0091] T result = [T result-x T result-y ],

[0092] Finally, the coordinate conversion matrix T result is saved in the image-robot coordinate calibration file, and the image-robot coordinate calibration file for conversion between the image coordinate system and the robot tool coordinate system is obtained.

[0093] Using the image-robot coordinate calibration file, the product matching point P d corresponding to the robot matching point P robot in the robot tool coordinate system is calculated.

[0094] P robot = T resul t·P image .

[0095] At the same time, the encoder 5 records the pulse number n0 of the current photographing position T of the vision camera 2,

[0096] The conveyor belt 1 is moved to move the calibration board into the mechanical arm working range of the robot 3, the encoder 5 records the pulse number n1 of the starting following point F1 and the pulse number n2 of the stopping following point F2, sends n0, n1 and n2 to the robot 3, and records in the robot controller;

[0097] The difference between the two pulse numbers is the number of pulses required for the product to be glued to move from the field of view of the vision camera 2 to the working range of the robot 3. The movement distance of the product to be glued can be calculated by the pulse difference. The pulse difference between the start following point F1 and the current photographing position T of the vision camera 2 is Δn1=n1-n0, and the pulse difference between the stop following point F2 and the current photographing position T of the vision camera 2 is Δn2=n2-n0. Therefore, when the pulse number difference of the encoder 5 is within the range of Δn1-Δn2, it indicates that the product to be glued has entered the working range of the mechanical arm of the robot 3. Therefore, the robot controller determines whether the product to be glued has entered the working range according to the recorded working range.

[0098] The three-dimensional model coordinate acquisition module 40 is configured to perform step S4: using the three-dimensional model of the product to be glued, and matching points P robot The matching point coordinate system is established as the coordinate origin, and the product contour three-dimensional coordinates of the product to be glued in the matching point coordinate system are calculated.

[0099] Specifically, the built three-dimensional model of the product to be glued is directly imported into a three-dimensional model processor, and the robot matching point P robot The X and Y coordinate origins of the three-dimensional model of the product to be glued are set, and the Z coordinate origin is set to 0. The product contour three-dimensional coordinates of the product to be glued in the matching point coordinate system are recalculated with (P robot , 0) as the coordinate origin. The product contour three-dimensional coordinates of the product to be glued in the matching point coordinate system are exported.

[0100] The three-dimensional model processor is pre-stored in the controller 6, and any existing three-dimensional model processing technology can be used for processing.

[0101] The glue application trajectory generation module 50 is configured to perform step S5: performing product contour point offset correction on the product contour three-dimensional coordinates to obtain corrected product contour three-dimensional coordinates, and generating a glue application trajectory.

[0102] Specifically, the product contour three-dimensional coordinates of the product to be glued in the matching point coordinate system exported in step S4 are processed by the correction processor to obtain corrected product contour three-dimensional coordinates. The corrected product contour three-dimensional coordinates of the product to be glued are exported and transmitted to the robot controller, and the final glue application trajectory is automatically generated in the robot controller and stored in the robot controller.

[0103] The correction processor is arranged in the controller 6, and the product profile point is offset corrected according to the actual process parameters of glue application during the correction of the correction processor.

[0104] The glue application module 60 is used to perform step S6: controlling the glue application fixture to apply glue to the product to be glued according to the glue application trajectory.

[0105] The robot 3 determines whether the product to be glued has entered the working range of the robot 3, i.e. the position range F1 to F2 recorded by the robot controller, according to the number of pulses transmitted by the encoder 5. If it is determined that the product to be glued has entered the range, the robot controller controls the mechanical arm to move the glue outlet nozzle of the glue application fixture 4 to the glue application starting point.

[0106] The robot controller controls the glue outlet of the glue gun on the glue application fixture 4 by controlling the point glue controller. The robot controller sends a pressure increasing signal to the point glue controller, and the point glue controller controls the 0.8Mpa air source to increase the pressure of the glue supply pressure barrel. The glue in the glue supply pressure barrel is transmitted to the glue inlet hole of the glue gun after being pressurized. At the same time, the point glue controller controls the 0.8Mpa air source to control the point glue valve to be in an open state through two air outlet pipes. The glue at the glue inlet hole enters the glue cavity and is sprayed out of the glue outlet nozzle of the glue spraying part.

[0107] Finally, the robot controller completes the glue application work by controlling the glue application trajectory of the glue application fixture 4, thereby realizing the visual glue application of the robot 3.

[0108] Since the obtained glue application trajectory of the product to be glued is generated based on the corrected product profile three-dimensional coordinates, and the glue application fixture 4 can realize multi-dimensional movement under the control of the robot 3, the glue application method of the present application can realize three-dimensional glue application work of the product, and is suitable for glue application of some products with curved surfaces and curved surfaces.

[0109] Compared with the prior art, the robot visual gluing method provided by the application adopts a visual camera 2 to take a picture of the product to be glued and positioned, determines a robot matching point through coordinate conversion, re-calculates the three-dimensional coordinates of the product contour with the robot matching point as the origin in a three-dimensional model of the product to be glued, corrects the product contour point of the product contour coordinates according to the on-site gluing requirement process, obtains the corrected three-dimensional coordinates of the product contour, generates a gluing track, and the robot 3 performs a gluing operation according to the gluing track. The robot visual gluing method does not need to design a special fixture for different products, thereby avoiding high fixture design costs. Meanwhile, only the visual camera 2 and the three-dimensional model are used to realize the visual positioning of the product to be glued and the generation of the gluing track, thereby saving costs, being more convenient to operate, and being more suitable for the use of products in the low-end industry. In addition, since the gluing track of the product to be glued is generated based on the corrected three-dimensional coordinates of the product contour, the method is not only suitable for the gluing of planar products, but also suitable for the gluing of products with radii and curved surfaces.

[0110] Based on the same inventive concept, the present application also provides an electronic device, which can be a server, a desktop computing device or a mobile computing device (for example, a laptop computer, a handheld computing device, a tablet computer, a netbook, etc.) or a terminal device. The device includes one or more processors and a memory, wherein the processor is configured to execute a program to implement the robot visual gluing method of the embodiments of the application; and the memory is configured to store a computer program executable by the processor.

[0111] Based on the same inventive concept, the present application also provides a computer readable storage medium, which corresponds to the robot visual gluing method of the above-mentioned embodiments, and the computer readable storage medium has a computer program stored thereon, wherein the program is executed by a processor to implement the steps of the robot visual gluing method described in any of the above-mentioned embodiments.

[0112] The application can take the form of a computer program product accessible from a storage medium (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code for use by or in connection with a computer. Computer-usable storage media include permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0113] The above-described embodiments only express several embodiments of the present application, which are described in detail and specifically, but cannot be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, and the present application also intends to include these modifications and improvements.

Claims

1. A robot vision-based adhesive application method, characterized in that: Includes the following steps: S1: Obtain the image of the product to be coated with glue captured by the vision camera and the image coordinates of each pixel in the image; S2: Obtain the product outline corresponding to the image of the product to be coated, generated by the image matching processor; determine two positioning points P0 and P1 on the product outline, and determine the product matching point P based on the two positioning points P0 and P1. d ; S3: Obtain the product matching point P based on the image coordinates. d Based on the image coordinate information, the matching point P with the product is calculated using a preset image-robot coordinate calibration file. d The corresponding robot matching point P in the robot tool coordinate system robot ; S4: Using the three-dimensional digital model of the product to be coated, the robot matching point P is located in the three-dimensional digital model. robot Establish a matching point coordinate system with the origin as the coordinate origin, and calculate the three-dimensional coordinates of the product outline of the product to be coated in the matching point coordinate system. S5: Correct the product outline point offset by the three-dimensional coordinates of the product outline to obtain the corrected three-dimensional coordinates of the product outline and generate the glue application trajectory. S6: Control the gluing fixture to apply glue to the product to be glued according to the gluing trajectory; The image-robot coordinate calibration file was obtained in the following way: A calibration board is placed within the field of view of the vision camera. The vision camera captures an image of the calibration board to obtain its two-dimensional image coordinates. The coordinates of nine feature points are selected and recorded as an image coordinate matrix. P image , ， Move the robot's robotic arm so that the glue dispensing nozzle of the glue applicator is aligned with the calibration point on the calibration plate, and record the robot tool coordinates of the calibration point on the calibration plate at this time. P robot-x , P robot-y , ， ， Perform coordinate transformation on the robot tool coordinates: Convert the image coordinate matrix into a square matrix. ， Calculate the transformation matrix T for X and Y respectively. result-x T result-y , ， ， Concatenate these two matrices, denoted as T. result , ， Transform the coordinate matrix T result Save the image-robot coordinate calibration file to obtain the image-robot coordinate calibration file that transforms the image coordinate system to the robot tool coordinate system. Using the image-robot coordinate calibration file, calculate the matching point P with the product. d The corresponding robot matching point P in the robot tool coordinate system robot , 。 2. The robot vision-based adhesive application method according to claim 1, characterized in that: In step S2, the image matching processor employs image matching technology, and the visual camera is communicatively connected to the image matching processor; the two positioning points P0 and P1 are selected as two symmetrical points on the product to be coated with adhesive, and the product matching point P... d It is the center point of the line connecting two positioning points P0 and P1.

3. The robot vision-based adhesive application method according to claim 2, characterized in that: Step S4 specifically involves: directly importing the 3D digital model of the product to be coated into the 3D digital model processor, and matching the robot point P. robot The origin of the X and Y coordinates of the three-dimensional digital model of the product to be coated is set to 0, and the origin of the Z coordinate is set to 0. robot ,0) as the origin of the coordinate system, recalculate the three-dimensional coordinates of the product outline of the product to be coated in the matching point coordinate system, and export the three-dimensional coordinates of the product outline of the product to be coated in the matching point coordinate system. The three-dimensional digital model processor can employ any three-dimensional digital model processing technology when performing three-dimensional digital model processing.

4. The robot vision adhesive application method according to claim 3, characterized in that: Step S5 specifically involves: The three-dimensional coordinates of the product outline in the matching point coordinate system exported in step S5 are processed by the correction processor to obtain the corrected three-dimensional product outline coordinates. The three-dimensional coordinates of the product outline to be coated with glue are exported and transmitted to the robot. The robot automatically generates the final coating trajectory and stores it in the robot. During the correction process, the product contour points are offset and corrected according to the actual adhesive application process parameters.

5. The robot vision adhesive application method according to claim 4, characterized in that: Specifically, step S6 involves the robot first determining whether the product to be coated has entered the robot's working range. If the robot determines that the product to be coated has entered the robot's working range, then the robot performs the coating operation according to the stored coating trajectory.

6. The robot vision-based adhesive application method according to claim 5, characterized in that: The robot determines whether the product to be coated has entered its working range based on the number of pulses transmitted by the encoder. During the generation of the image-robot coordinate calibration file, the encoder is used to record the robot's working range. The specific steps are as follows: The encoder records the number of pulses at the position T captured by the visual camera. n 0, The calibration board is moved to the robot's working range, and the encoder records the number of pulses starting from the following point F1. n 1. The number of pulses to stop following point F2 n 2. n 0、 n 1 and n 2. Send to the robot and record in the robot; The movement distance of the product to be coated is calculated by the pulse count difference, and the pulse count difference between the starting tracking point F1 and the visual camera's image capture position T is denoted as... n 1 =n 1 -n 0, the difference in the number of pulses between the stop tracking point F2 and the current visual camera capture position T is denoted as . n 2 =n 2 -n 0, if the difference in the number of pulses of the encoder is within n 1~ n If the value is within the range of 2, it means that the product to be coated with adhesive has entered the working range of the robot.

7. A robot vision adhesive application controller for performing the robot vision adhesive application method as described in claim 1, characterized in that: It includes an image acquisition module, a product matching point acquisition module, a robot matching point acquisition module, a 3D digital model coordinate acquisition module, an adhesive application trajectory generation module, and an adhesive application module; The image acquisition module is used to acquire images of the product to be coated with adhesive taken by the vision camera and the image coordinates of each pixel in the image; The product matching point acquisition module is used to acquire the product outline corresponding to the image of the product to be coated with adhesive generated by the image matching processor; determine two positioning points P0 and P1 on the product outline; and determine the product matching point P based on the two positioning points P0 and P1. d ; The robot matching point acquisition module is used to acquire the product matching point P based on the image coordinates. d Based on the image coordinate information, the matching point P with the product is calculated using a preset image-robot coordinate calibration file. d The corresponding robot matching point P in the robot tool coordinate system robot ; The three-dimensional digital model coordinate acquisition module is used to obtain robot matching point P in the three-dimensional digital model of the product to be coated. robot Establish a matching point coordinate system with the origin as the coordinate origin, and calculate the three-dimensional coordinates of the product outline of the product to be coated in the matching point coordinate system. The adhesive application trajectory generation module is used to correct the product outline point offset in the three-dimensional coordinates of the product outline, obtain the corrected three-dimensional coordinates of the product outline, and generate the adhesive application trajectory. The glue application module is used to control the glue application fixture to apply glue to the product to be glued according to the glue application trajectory.

8. A robot vision-based adhesive application system, characterized in that: This includes conveyor belts, vision cameras, robots, gluing fixtures, encoders, and controllers; The conveyor belt is mounted on a workbench, and the controller controls the movement of the conveyor belt; The vision camera is positioned directly above the conveyor belt with its lens facing the conveyor surface. The vision camera is communicatively connected to the controller, which controls the vision camera to take pictures. The vision camera then transmits the pictures back to the controller. The robot includes a robot controller and a robotic arm. The robot controller is used to control the movement of the robotic arm. The robot controller is communicatively connected to the controller to receive the glue application coordinate information transmitted by the controller. The adhesive application fixture is mounted on the robotic arm, and the robotic arm is used to control the movement of the adhesive application fixture. The encoder is mounted on the conveyor belt and moves synchronously with the conveyor belt. The conveyor belt is communicatively connected to the robot controller to record the distance the conveyor belt moves. The controller is the controller described in claim 7.

Images