Mobile inkjet printing equipment and system

By designing mobile inkjet printing equipment and systems, using piezoelectric oscillator and ultraviolet curing lamp technology, combined with vision modules and deep learning network models, the automation and aesthetics of train car inkjet printing are achieved, and the problems of cumbersome manual operations and uneven inkjet printing are solved in the existing technology.

CN117901554BActive Publication Date: 2025-06-06SHANDONG JITIE MASCH EQUIP GRP CO LTD +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410106291.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-25
Publication Date
2025-06-06
Estimated Expiration
2044-01-25

AI Technical Summary

Technical Problem

The existing train car inkjet coding technology has problems such as cumbersome manual operations, diverse template production, uneven and unsightly inkjet coding, and the dot matrix inkjet coding of the automated inkjet coding does not meet specific standards, which is suspected of tampering with the shape of the font.

Method used

A mobile inkjet printing equipment and system is designed, including a mobile car, a robotic arm and a inkjet printer. The inkjet printer is equipped with a nozzle and an ink supply device. The ink is extruded with a piezoelectric oscillator to make the ink spray out in a mist, and combined with an ultraviolet curing lamp to ensure uniform curing of the ink. The system also integrates a vision module and a PLC electronic control module, and realizes fully automated coding through 3D positioning algorithms and deep learning network models.

Benefits of technology

The continuity and uniformity of the ink-coded font or shape are achieved, manpower is reduced, the aesthetics and accuracy of the ink-coded font is ensured, and the uneven and inclined fonts are avoided, forming an efficient and automated ink-coded system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117901554B_ABST
    Figure CN117901554B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of coding in train carriages, and discloses a mobile coding device and system, which includes a mobile vehicle, a mechanical arm, and a coding machine; the coding machine is installed on the mechanical arm, the mechanical arm is installed on the mobile vehicle, the mobile vehicle is used to drive the mechanical arm to travel, the mechanical arm is used to drive the coding machine to rise and fall, and the coding machine is used to code the outer surface of the carriage; the coding machine includes a nozzle and an ink supply device, the nozzle is provided with a coding crystal and a piezoelectric oscillator, the piezoelectric oscillator is in contact with the coding crystal, and the piezoelectric oscillator is used to squeeze the coding crystal to spray the ink in a mist form. The present invention has the effect of automatic coding, reducing manual labor, and making the graphic ink sprayed out uniform and continuous.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of railway carriage coding, and in particular to a mobile coding device and system. Background Art

[0002] When railway train carriages are repaired in the factory or section, they need to be repainted with the type, model, license plate number and performance markings of the repaired vehicle at specific locations on the vehicle body according to different requirements in order to meet the technical requirements for vehicle inspection after maintenance.

[0003] The old method of inkjet coding is to stick a template where the coding is needed and spray it manually with a paint sprayer. This method will form a uniform color paint, but different words require multiple templates to be made and they need to be sprayed manually. Later, automatic inkjet printers appeared, which do not require molds and can directly spray out the desired shape or font. However, the automatic inkjet printers spray out dot matrix with very large spacing. The human eye can see that it is a combination of dots, which is not beautiful and does not meet the specific painting standards. There is a suspicion of tampering with the font. Summary of the invention

[0004] In order to reduce manual labor and make the ink sprayed uniform, the present invention provides a mobile inkjet printing device and system.

[0005] The present invention provides a mobile inkjet printer and system that adopts the following technical solutions:

[0006] In a first aspect, a mobile inkjet printer comprises a mobile vehicle, a mechanical arm and an inkjet printer;

[0007] The inkjet printer is installed on a mechanical arm, the mechanical arm is installed on a mobile vehicle, the mobile vehicle is used to drive the mechanical arm to travel, the mechanical arm is used to drive the inkjet printer to rise and fall, and the inkjet printer is used to print on the outer surface of the carriage;

[0008] The inkjet printer includes a nozzle and an ink supply device. The nozzle is provided with an inkjet crystal and a piezoelectric oscillator. The piezoelectric oscillator is in contact with the inkjet crystal. The piezoelectric oscillator is used to squeeze the inkjet crystal to spray the ink in a mist form.

[0009] By adopting the above technical scheme, the piezoelectric oscillator is used to squeeze the coding transistor, and the ink is sprayed out in a mist form, so that the font or shape of the coding becomes continuous, and the coding transistor is used. The coding transistor is densely distributed, which improves the phenomenon of discontinuous shape after coding. By controlling the operation of the piezoelectric oscillator, manual labor is reduced, and the sprayed ink can be made uniform and continuous.

[0010] Furthermore, the ink supply device includes an ink box and an ink supply tube. The ink box is installed on the mobile vehicle. One end of the ink supply tube is connected to the ink box, and the other end is connected to the coding transistor. The ink supply tube is distributed around the robotic arm.

[0011] Furthermore, the lower end of the mobile vehicle is provided with an electrically controlled universal wheel, the mechanical arm is provided with an electrically controlled lifting device for lifting and lowering, the mechanical arm is installed with an industrial 3D camera, and the industrial 3D camera and the inkjet printer are both arranged at the end of the mechanical arm.

[0012] Furthermore, the inkjet printer is provided with an ultraviolet curing lamp in the horizontal direction.

[0013] By adopting the above technical solution, if the ink cannot be completely solidified immediately after being sprayed, a UV curing lamp is installed on one side of the inkjet printer. When the ink is sprayed onto the carriage, the UV curing lamp solidifies the ink, making the sprayed fonts or graphics neat and beautiful.

[0014] In a second aspect, a mobile inkjet coding system includes the following modules: a visual module and a PLC electronic control module; the visual module is connected to the PLC electronic control module;

[0015] The visual module includes a visual recognition unit and a visual positioning unit. The industrial 3D camera is used to record the image data of the carriage. The visual recognition unit identifies the size of the carriage based on the visual SLAM algorithm and the weighted image learning algorithm. The visual positioning unit locates the position of the coding on the surface of the carriage based on the 3D positioning algorithm.

[0016] Furthermore, the visual SLAM algorithm is specifically:

[0017] Extract pixel coordinates (x, y) from multiple images taken by industrial 3D cameras, and obtain coordinate shape vectors and texture information vectors from the pixel coordinates (x, y). newModel is the coordinate shape vector, T newModel is the texture information vector;

[0018]

[0019] The mathematical expression of the 3D model is M = (S newModel , T newModel );

[0020] m is the number of pixel coordinates, i is the i-th pixel coordinate;

[0021] is the average coordinate shape model; s i is the PCA part of the image, that is, the eigenvector of the covariance matrix of the principal component analysis arranged in descending order of eigenvalues; α i is the coordinate shape coefficient;

[0022] is the average texture information model, t i is the eigenvector of the covariance matrix arranged horizontally according to the eigenvalues ​​of the principal component analysis; β i is the texture information coefficient.

[0023] Furthermore, the weighted image learning algorithm is specifically:

[0024] Based on the 3D model, the front view, side view and top view of the car are generated. According to the idea of ​​greedy algorithm, the minimum spanning tree is searched in the weighted connected graph of the three images, that is, all vertices in the image are connected and the sum of the weights of all edges is minimized. The size of the car is obtained according to the weights.

[0025] Furthermore, the 3D positioning algorithm is specifically:

[0026] According to the nonlinear least squares method, the carriage image is divided into several points (P n , P n '), set the reprojection parameter of the inkjet starting point θ=(θ x ,θ y ,θ z ,θ α ,θ β ,θ γ );

[0027] Find several points randomly on the carriage, mark the position coordinates P' of the points, and divide the carriage image into several coordinate points P;

[0028] P n After the rotation transformation R x (θ α )·R y (θ β )·R z (θ γ )·P n , after translation transformation (θ x ,θ y ,θ z ) T , we get the point m(θ,P i )=(θ x ,θ y ,θ z ) T +R x (θ α )·R y (θ β )·R z (θ γ )·P i, and then the point m(θ,P i ) Perform perspective projection to obtain point m' on the surface of the carriage;

[0029] The proportional factor is s = (s x ,s y );

[0030] s x =s y =f, f is the focal length of the industrial 3D camera, (c x , c y )=(0,0);

[0031] Finally, all the values ​​are brought into m'(P i ,θ), and the nonlinear least square method is used to obtain θ=(θ x ,θ y ,θ z ,θ α ,θ β ,θ γ ) to locate the position of the code on the surface of the carriage.

[0032] Furthermore, the visual module transmits the located inkjet printer position on the carriage surface to the PLC electronic control module, and the PLC electronic control module controls the robotic arm to lift the inkjet printer to a calibrated position.

[0033] Furthermore, the visual module transmits the carriage size to the PLC electronic control module, and the PLC electronic control module calculates the data required for coding under the carriage size according to the deep learning network model, and the PLC electronic control module controls the piezoelectric oscillator to pressurize the coding transistor according to the data required for coding;

[0034] The deep learning network model calculates that the data required for coding under the carriage size is specifically:

[0035] The dimensions of a carriage include its length, width and height. The volume of the carriage is calculated based on the length, width and height, and then converted into deadweight tonnage. Based on the appearance of the carriage, it is determined whether the carriage is a covered car or an open car.

[0036] By adopting the above technical solution, since the mobile vehicle may not be level with the carriage after it moves near the carriage, the position where the coding needs to be sprayed on the carriage is located through the 3D positioning algorithm, and the horizontal and vertical positions where the coding needs to be sprayed are obtained, which effectively ensures that the coding is facing the carriage to avoid the tilt of the coding. The size of the carriage is identified according to the visual SLAM algorithm and the weighted image learning algorithm, and the data required for coding is obtained by the deep learning network model, forming a fully automated coding equipment, which effectively reduces manual labor.

[0037] In summary, the present invention has the following beneficial technical effects:

[0038] 1. The piezoelectric oscillator squeezes the inkjet transistor and sprays the ink in a mist form, making the font or shape of the inkjet code continuous. The dense distribution of the inkjet transistor improves the discontinuous shape after inkjet coding. By controlling the operation of the piezoelectric oscillator, the manpower is reduced and the ink sprayed can be uniform and continuous.

[0039] 2. If the ink cannot be completely solidified immediately after being sprayed, install a UV curing lamp on one side of the inkjet printer. When the ink is sprayed onto the carriage, the UV curing lamp will solidify the ink, making the sprayed fonts or graphics neat and beautiful;

[0040] 3. Since the mobile vehicle may not be level with the carriage after it moves near the carriage, the 3D positioning algorithm is used to locate the position where the code needs to be sprayed on the carriage, and the horizontal and vertical positions where the code needs to be sprayed are obtained, which effectively ensures that the code is facing the carriage and avoids the tilt of the code. The size of the carriage is identified according to the visual SLAM algorithm and the weighted image learning algorithm, and then the deep learning network model is used to obtain the data that needs to be sprayed, forming a fully automated coding equipment, which effectively reduces manual labor. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The present invention is a mobile inkjet coding device and system overall equipment diagram.

[0042] Figure 2 The present invention discloses a system flow chart of a mobile inkjet coding device and system.

[0043] Explanation of the accompanying drawings: 1. Mobile vehicle; 2. Robotic arm; 3. Ultraviolet curing lamp; 4. Printhead; 5. Ink box; 6. Ink supply tube; 7. Electric universal wheel; 8. Industrial 3D camera. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0045] The embodiment of the invention discloses a mobile inkjet coding device and system.

[0046] Reference Figure 1 , a mobile inkjet printer device, comprising a mobile vehicle 1, a mechanical arm 2, an inkjet printer and an ultraviolet curing lamp 3;

[0047] The inkjet printer and the ultraviolet curing lamp 3 are both installed on the mechanical arm 2, and the mechanical arm 2 is installed on the mobile vehicle 1. The mobile vehicle 1 is used to drive the mechanical arm 2 to travel, and the mechanical arm 2 is used to drive the inkjet printer to rise and fall. The inkjet printer is used to spray codes on the outer surface of the carriage.

[0048] An electrically controlled universal wheel 7 is provided at the lower end of the mobile vehicle 1, which can drive the mobile vehicle 1 to travel. An electrically controlled lifting device for lifting is provided on the robotic arm 2. An industrial 3D camera 8 is also installed on the robotic arm 2. The industrial 3D camera 8 and the inkjet printer are both arranged at the end of the robotic arm 2.

[0049] It should be noted that the mechanical arm 2 and the electric universal wheel 7 are both prior art. The working principle of the mechanical arm 2 refers to the Chinese authorized patent with patent number CN109562523B and patent name “mechanical arm mechanism”. The working principle of the electric universal wheel 7 refers to the Chinese authorized patent with patent number CN208247864U and patent name “electric universal wheel”.

[0050] The inkjet printer includes a nozzle 4 and an ink supply device. The nozzle 4 is provided with a coding transistor and a piezoelectric oscillator. The nozzle 4 is also provided with an ink tank for temporarily storing ink. The coding transistors are placed in a row. The piezoelectric oscillator is in contact with the coding transistors. The piezoelectric oscillator is used to squeeze the coding transistors to spray the ink in a mist form and fall on the surface of the printed object to form a 36-48PL small dot matrix, thereby forming text, numbers or graphics. The ultraviolet curing lamp 3 and the nozzle 4 are on the same horizontal line.

[0051] Since the ink cannot be completely solidified immediately after being sprayed, an ultraviolet curing lamp 3 is installed on one side of the nozzle 4. When the ink is sprayed onto the carriage, the ultraviolet curing lamp 3 solidifies the ink without ink flowing downward, making the sprayed fonts or graphics neat and beautiful.

[0052] The ink supply device is divided into an ink circuit and an air circuit. The ink circuit includes an ink tank 5 and an ink supply tube 6. The air circuit is used to form air pressure to supply ink to the nozzle 4. The ink tank 5 is installed on the mobile vehicle 1. One end of the ink supply tube 6 is connected to the ink tank 5 and the other end is connected to the coding transistor. The ink supply tube 6 is distributed around the mechanical arm 2. An electric pump is arranged in the ink tank 5. One end of the electric pump is connected to the air circuit to balance the air pressure. When there is a demand for ink supply, the ink is pumped out by the electric pump and transferred to the nozzle 4 through the ink supply tube 6. The piezoelectric oscillator squeezes the coding transistor and sprays the ink in a mist form, which falls on the surface of the printed object to form a 36-48PL small dot matrix. The dense small dot matrix makes the font or shape of the coding continuous. After the spraying is completed, the crystal in the nozzle returns to its original state. Due to the negative pressure formed by the air circuit, the ink flows into the ink tank inside the nozzle 4, and new ink enters the nozzle to continue to complete the printing. The dense distribution of inkjet transistors improves the phenomenon of discontinuous shape after inkjet printing. The remote control of the piezoelectric oscillator operation reduces manual labor and makes the ink sprayed even and continuous.

[0053] Reference Figure 2 In order to reduce manpower and realize automatic coding, a mobile coding system is embedded in the coding equipment. The coding system has built-in pipeline algorithm and arithmetic application algorithm, and is highly integrated with the hardware for modular design. It should be noted that both the pipeline algorithm and the arithmetic application algorithm use existing algorithms.

[0054] The inkjet coding system includes the following modules: a visual module and a PLC electronic control module; the visual module is connected to the PLC electronic control module.

[0055] The visual module includes a visual recognition unit and a visual positioning unit. The industrial 3D camera 8 is used to record the image data of the carriage. The visual recognition unit identifies the size of the carriage based on the visual SLAM algorithm and the weighted image learning algorithm. The visual positioning unit locates the position of the inkjet coding on the surface of the carriage based on the 3D positioning algorithm.

[0056] After the mobile vehicle 1 moves near the carriage, the nozzle 4 may not be level with the carriage, so a point on the carriage is first found as a reference point, and the position where the code needs to be sprayed on the carriage is located through the 3D positioning algorithm, so as to obtain the horizontal and vertical position where the code needs to be sprayed, which effectively ensures that the code is facing the carriage and avoids the tilt of the code.

[0057] The specific 3D positioning algorithm is:

[0058] According to the nonlinear least squares method, the carriage image is divided into several points (P n , P n '), set the reprojection parameter of the inkjet starting point θ=(θ x ,θ y ,θ z ,θα ,θ β ,θ γ );

[0059] Find several points randomly on the carriage, mark the position coordinates P' of the points, and divide the carriage image into several coordinate points P;

[0060] P n After the rotation transformation R x (θ α )·R y (θ β )·R z (θ γ )·P n , after translation transformation (θ x ,θ y ,θ z ) T , we get the point m(θ,P i )=(θ x ,θ y ,θ z ) T +R x (θ α )·R y (θ β )·R z (θ γ )·P i , and then the point m(θ,P i ) Perform perspective projection to obtain point m' on the surface of the carriage;

[0061] The proportional factor is s = (s x ,s y );

[0062] s x =s y =f, f is the focal length of the industrial 3D camera, (c x , c y )=(0,0);

[0063] Finally, all the values ​​are brought into m'(P i ,θ), and the nonlinear least square method is used to obtain θ=(θ x ,θ y ,θ z ,θ α ,θ β ,θ γ ) to locate the position of the code on the surface of the carriage.

[0064] The visual module transmits the located inkjet printer position on the carriage surface to the PLC electronic control module, and the PLC electronic control module controls the robot arm 2 to adjust the position of the inkjet printer to be level with the calibrated position.

[0065] The specific visual SLAM algorithm is:

[0066] Extract pixel coordinates (x, y) from multiple images taken by industrial 3D cameras, and obtain coordinate shape vectors and texture information vectors from the pixel coordinates (x, y). newModel is the coordinate shape vector, T newModel is the texture information vector;

[0067]

[0068] The mathematical expression of the 3D model is M = (S newModel , T newModel );

[0069] m is the number of pixel coordinates, i is the i-th pixel coordinate;

[0070] is the average coordinate shape model; s i is the PCA part of the image, that is, the eigenvector of the covariance matrix of the principal component analysis arranged in descending order of eigenvalues; α i is the coordinate shape coefficient;

[0071] is the average texture information model, t i is the eigenvector of the covariance matrix arranged horizontally according to the eigenvalues ​​of the principal component analysis; β i is the texture information coefficient.

[0072] The visual SLAM algorithm extracts the image recorded in the industrial 3D camera 8 into multiple coordinate points, and finally forms a 3D model that can be recognized by a computer.

[0073] The weighted image learning algorithm is specifically:

[0074] Based on the 3D model, the front view, side view and top view of the car are generated. According to the greedy algorithm, the minimum spanning tree is searched in the weighted connected graph of the three images, that is, all vertices in the image are connected and the sum of the weights of all edges is minimized, and the size of the car is obtained according to the weights. The greedy algorithm is an existing algorithm and can be found on Baidu Encyclopedia.

[0075] The vision module passes the car size to the PLC electronic control module. The PLC electronic control module calculates the data that needs to be coded under the car size based on the deep learning network model. The PLC electronic control module controls the piezoelectric oscillator to pressurize the coding transistor according to the data that needs to be coded.

[0076] The deep learning network model calculates that the data that needs to be coded for this carriage size is:

[0077] The dimensions of a carriage include its length, width and height. The volume of the carriage is calculated based on the length, width and height, and then converted into deadweight tonnage. The conversion between cubic meters and tons is based on the formulas mass = volume x density and volume = mass ÷ density. The density is calculated using the density of water.

[0078] And based on the appearance of the car, it can be determined whether it is a covered car. A covered car is marked with P, and an open car is marked with C.

[0079] The car body size is identified based on the visual SLAM algorithm and the weighted image learning algorithm, and then the deep learning network model is used to obtain the data that needs to be coded, forming a fully automated coding equipment, which effectively reduces manual labor.

[0080] The implementation principle of a mobile inkjet printer and system in the embodiment of the present invention is as follows: the mobile vehicle 1 brings the inkjet printer to the front of the carriage, the industrial 3D camera 8 takes a photo of the carriage for analysis, and locates the position where the carriage needs to be coded by the 3D positioning algorithm, and obtains the horizontal and vertical position where the coding needs to be coded, effectively ensuring that the coding is facing the carriage, avoiding the tilt phenomenon of the coding, and identifying the size of the carriage according to the visual SLAM algorithm and the weighted image learning algorithm, and then using the deep learning network model to obtain the data required for coding. The system program is first processed and distributed by the PLC electronic control module, and then a series of electrical signals are output to each coding transistor in the nozzle through the drive board. The high-frequency deformation of the coding transistor causes the ink to be ejected from the nozzle 4 and fall on the surface of the printed object, forming a 36-48PL small dot matrix, thereby forming text, numbers or graphics. Then, the coding transistor in the nozzle 4 returns to its original state, and due to the negative pressure formed by the air path, the ink enters the ink tank inside the nozzle 4, and the new ink enters the nozzle to continue to complete the printing. The dense distribution of inkjet transistors improves the phenomenon of discontinuous shape after inkjet printing. By controlling the operation of the piezoelectric oscillator, it reduces manual labor and makes the ink sprayed even and continuous, forming a fully automated inkjet equipment.

[0081] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A mobile inkjet printer, characterized in that: Including mobile vehicle, robotic arm and inkjet printer; The inkjet printer is installed on a mechanical arm, the mechanical arm is installed on a mobile vehicle, the mobile vehicle is used to drive the mechanical arm to travel, the mechanical arm is used to drive the inkjet printer to rise and fall, and the inkjet printer is used to print on the outer surface of the carriage; The inkjet printer comprises a nozzle and an ink supply device, wherein a coding crystal and a piezoelectric oscillator are arranged in the nozzle, wherein the piezoelectric oscillator is in contact with the coding crystal, and the piezoelectric oscillator is used to squeeze the coding crystal to spray the ink in a mist form; The ink supply device comprises an ink box and an ink supply tube, wherein the ink box is installed on the mobile vehicle, one end of the ink supply tube is connected to the ink box, and the other end is connected to the inkjet transistor, and the ink supply tube is distributed around the robot arm; The lower end of the mobile vehicle is provided with an electrically controlled universal wheel, the mechanical arm is provided with an electrically controlled lifting device for lifting, the mechanical arm is installed with an industrial 3D camera, and the industrial 3D camera and the inkjet printer are both arranged at the end of the mechanical arm; It also includes the following modules: a visual module and a PLC electronic control module, wherein the visual module is connected to the PLC electronic control module; The visual module includes a visual recognition unit and a visual positioning unit. The industrial 3D camera is used to record the image data of the carriage. The visual recognition unit identifies the size of the carriage based on the visual SLAM algorithm and the weighted image learning algorithm. The visual positioning unit locates the position of the coding on the surface of the carriage based on the 3D positioning algorithm. The visual SLAM algorithm is specifically: Extract pixel coordinates (x, y) from multiple images taken by industrial 3D cameras, and obtain coordinate shape vectors and texture information vectors from the pixel coordinates (x, y). newModel is the coordinate shape vector, T newModel is the texture information vector; The mathematical expression of the 3D model is M = (S newModel , T newModel ); m is the number of pixel coordinates, i is the i-th pixel coordinate; is the average coordinate shape model; s i is the PCA part of the image, that is, the eigenvector of the covariance matrix of the principal component analysis arranged in descending order of eigenvalues; α i is the coordinate shape coefficient; is the average texture information model, t i is the eigenvector of the covariance matrix arranged horizontally according to the eigenvalues ​​of the principal component analysis; β i is the texture information coefficient; The 3D positioning algorithm is specifically: According to the nonlinear least squares method, the carriage image is divided into several points (P n , P n '), set the reprojection parameter of the inkjet starting point θ=(θ x ,θ y ,θ z ,θ α ,θ β ,θ γ ); Find several points randomly on the carriage, mark the position coordinates P' of the points, and divide the carriage image into several coordinate points P; P n After rotation transformation R x (θ α )·R y (θ β )·R z (θ γ )·P n , after translation transformation (θ x ,θ y ,θ z ) T , we get the point m(θ,P i )=(θ x ,θ y ,θ z ) T +R x (θ α )·R y (θ β )·R z (θ γ )·P i , and then the point m(θ,P i ) Perform perspective projection to obtain point m' on the surface of the carriage; The proportional factor is s = (s x ,s y ); s x =s y =f, f is the focal length of the industrial 3D camera, (c x , c y )=(0,0); Finally, all the values ​​are brought into m'(P i ,θ), and the nonlinear least square method is used to obtain θ=(θ x ,θ y ,θ z ,θ α ,θ β ,θ γ ) to locate the position of the code on the surface of the carriage; The visual module transmits the located inkjet printer position on the carriage surface to the PLC electronic control module, and the PLC electronic control module controls the mechanical arm to lift the inkjet printer to the calibrated position.

2. A mobile inkjet printer according to claim 1, characterized in that: The inkjet printer is provided with an ultraviolet curing lamp in the horizontal direction.

3. The mobile inkjet printer according to claim 1, characterized in that: The weighted image learning algorithm is specifically: Based on the 3D model, the front view, side view and top view of the car are generated. According to the idea of ​​greedy algorithm, the minimum spanning tree is searched in the weighted connected graph of the three images, that is, all vertices in the image are connected and the sum of the weights of all edges is minimized. The size of the car is obtained according to the weights.

4. The mobile inkjet printer according to claim 1, characterized in that: The visual module transmits the carriage size to the PLC electronic control module, and the PLC electronic control module calculates the data that needs to be coded under the carriage size according to the deep learning network model. The PLC electronic control module controls the piezoelectric oscillator to pressurize the coding transistor according to the data required for coding; The deep learning network model calculates that the data required for coding under the carriage size is specifically: The dimensions of a carriage include its length, width and height. The volume of the carriage is calculated based on the length, width and height, and then converted into deadweight tonnage. Based on the appearance of the carriage, it is determined whether the carriage is a covered car or an open car.

Citation Information

Patent Citations

  • robotic arm mechanism

    CN109562523B

  • Electronic universal wheel

    CN208247864U

  • Spraying system and spraying method

    CN112078253A

  • Automobile curved part ink-jet printing device based on multi-axis manipulator and control system

    CN116394664A