A frameless door glass assembly method, device, equipment and storage medium

Abstract

The application relates to the technical field of vehicle manufacturing, and discloses a frameless door glass assembling method, device, equipment and storage medium, the method comprising the following steps: acquiring images of a to-be-installed frameless door glass at different standard tool positions; generating a plurality of point cloud images according to the acquired images; collecting a plurality of groups of point cloud data from the generated point cloud images, and performing point cloud registration on the collected plurality of groups of point cloud data; and determining an installation position and a movement displacement of the to-be-installed frameless door glass according to a registration result. After the images of the frameless door glass at the different standard tool positions are acquired, the full-automatic installation position and the movement displacement of the frameless door glass are determined through point cloud processing, and then full-automatic installation of a frameless door glass assembling robot is realized; the entire process does not need manual operation of a plurality of sets of cooperative tooling, the assembling speed can be improved, the precision is high, and the method is suitable for frameless door glasses of various types.

Description

Technical Field

[0001] This invention relates to the field of vehicle manufacturing technology, and in particular to a method, apparatus, equipment and storage medium for assembling frameless car door glass. Background Technology

[0002] In the current automotive industry context, the assembly of frameless door glass involves three main aspects: high assembly precision requirements, high assembly efficiency requirements, and limited assembly space requirements. The current industry-standard solution involves manual assembly using specially calibrated tooling, which presents several problems:

[0003] The first problem is the slow speed of manual assembly. On-site workers need to first use a collaborative tooling to align with specific reference points on the car door, and after locking the collaborative tooling, use a manual suction cup to lift the frameless glass. At this point, another worker tightens the screws. In the entire workflow, it often happens that manual workers cannot keep up with the high speed of the assembly line in the final assembly workshop, causing the entire plant's conveyor line to stop. It is worth mentioning that the loss of conveyor line downtime in the final assembly workshop is generally calculated in seconds.

[0004] The second problem is that the collaborative tooling used by workers is difficult to adapt to all types of frameless glass doors. Given the current complex vehicle designs and market competition, rapid installation and adaptation for new models is a fiercely contested area for every automaker. The common solution now is to design and calibrate a set of collaborative tooling for each vehicle model.

[0005] To solve the second problem, manufacturers have to use multiple sets of collaborative tooling. However, for each type of frameless door, even a small modification requires a lot of time to design and calibrate new collaborative tooling. In addition, each set of collaborative tooling must be within the space that workers can operate in. This means that for the main assembly line, necessary workstations must be reserved for each set of collaborative tooling. For example, 10 different models would require 10 workstations. Workstation costs are a factor that manufacturers have to consider when designing.

[0006] In addition, while manual use of collaborative tooling can ensure accuracy based on the calibration of the tooling, in reality, the calibration often does not match the actual production of the car door. Line personnel need to make adjustments according to the actual situation and increase the compatibility by manually adding shims. In fact, this method has no accuracy whatsoever. Summary of the Invention

[0007] In view of this, the purpose of this invention is to provide a frameless car door glass assembly method, apparatus, equipment, and storage medium, which can realize fully automated installation of high-precision frameless car door glass by a robotic assembly system. The specific solution is as follows:

[0008] A method for assembling frameless car door glass includes:

[0009] Acquire images of the frameless door glass to be installed at different standard tooling positions;

[0010] Based on the acquired images, generate multiple point cloud maps;

[0011] Multiple sets of point cloud data are collected from the generated point cloud map, and point cloud registration is performed on the collected multiple sets of point cloud data;

[0012] Based on the registration results, the installation position and movement displacement of the frameless door glass to be installed are determined.

[0013] Preferably, in the frameless door glass assembly method provided in the embodiments of the present invention, the step of performing point cloud registration on the collected multiple sets of point cloud data includes:

[0014] Multiple registrations of the collected point cloud data sets were performed on the digital model point cloud to obtain multiple registration results;

[0015] Obtain the optimal registration result and its corresponding registration parameters from all the obtained registration results;

[0016] The step of determining the installation position and movement displacement of the frameless door glass to be installed based on the registration results includes:

[0017] Based on the optimal registration result and the corresponding registration parameters, the installation position and movement displacement of the frameless door glass to be installed are determined.

[0018] Preferably, in the frameless door glass assembly method provided in the embodiments of the present invention, the step of performing multiple registrations of the collected multiple sets of point cloud data to obtain multiple registration results includes:

[0019] The multiple sets of point cloud data collected are registered according to the preset initial registration parameters. The registered point cloud data is then compared with the point cloud data of the standard installation template to obtain the first registration result.

[0020] Subtract the pre-set single fixed subtraction value from the initial registration parameters to obtain new registration parameters. Perform registration operation based on the new registration parameters, and compare the registered point cloud data with the standard installation template point cloud data to obtain a second registration result.

[0021] Repeat the step of subtracting the fixed subtraction value from the previous registration parameter to obtain the next registration parameter until the obtained registration parameter is less than or equal to the set threshold registration parameter, then stop the registration operation to obtain multiple registration results.

[0022] Preferably, in the frameless door glass assembly method provided in the embodiments of the present invention, before acquiring images of the frameless door glass to be installed at different standard tooling positions, the method further includes:

[0023] Get images of frameless car door windows of different models and colors that have been installed;

[0024] Based on the images of frameless car door glass of different models and colors, multiple sets of point cloud data were obtained;

[0025] After preprocessing the obtained multiple sets of point cloud data, standard installation template point cloud data corresponding to different models and colors are obtained.

[0026] Preferably, in the frameless door glass assembly method provided in the embodiments of the present invention, after generating multiple point cloud maps, the method further includes:

[0027] The point cloud map is compared and identified with the point cloud data of all the standard installation templates to identify the model and color of the door component to be installed.

[0028] Based on the identified model and color of the frameless door glass to be installed, the point cloud data of the standard installation template corresponding to the frameless door glass to be installed is found.

[0029] This invention also provides a frameless car door glass assembly device, comprising:

[0030] The image acquisition module is used to acquire images of the frameless door glass to be installed at different standard tooling positions;

[0031] The point cloud image generation module is used to generate multiple point cloud images based on the acquired image.

[0032] The point cloud registration module is used to collect multiple sets of point cloud data from the generated point cloud map and to perform point cloud registration on the collected multiple sets of point cloud data.

[0033] The installation position determination module is used to determine the installation position and movement displacement of the frameless door glass to be installed based on the registration results.

[0034] Preferably, in the frameless car door glass assembly device provided in the embodiments of the present invention, the point cloud registration module is specifically used to perform multiple registrations of the collected multiple sets of point cloud data into digital model point clouds to obtain multiple registration results; and to obtain the optimal registration result and the corresponding registration parameters from all the obtained registration results.

[0035] The installation position determination module is specifically used to determine the installation position and movement displacement of the frameless door glass to be installed based on the optimal registration result and the corresponding registration parameters.

[0036] Preferably, in the frameless car door glass assembly device provided in the embodiments of the present invention, the point cloud registration module is specifically used to perform registration operations on the collected multiple sets of point cloud data according to the preset initial registration parameters, compare the registered point cloud data with the standard installation template point cloud data to obtain the first registration result; subtract the preset single fixed subtraction value from the initial registration parameters to obtain a new registration parameter, perform registration operations according to the new registration parameter, compare the registered point cloud data with the standard installation template point cloud data to obtain the second registration result; repeat the step of subtracting the single fixed subtraction value from the previous registration parameter to obtain the next registration parameter until the obtained registration parameter is less than or equal to the set threshold registration parameter, and stop the registration operation to obtain multiple registration results.

[0037] This invention also provides a frameless car door glass assembly device, including a processor and a memory, wherein the processor executes a computer program stored in the memory to implement the frameless car door glass assembly method provided in this invention.

[0038] This invention also provides a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the frameless door glass assembly method described above in this invention.

[0039] As can be seen from the above technical solution, the frameless car door glass assembly method provided by the present invention includes: acquiring images of the frameless car door glass to be installed at different standard tooling positions; generating multiple point cloud maps based on the acquired images; collecting multiple sets of point cloud data from the generated point cloud maps, and performing point cloud registration on the collected multiple sets of point cloud data; and determining the installation position and movement displacement of the frameless car door glass to be installed based on the registration result.

[0040] The frameless car door glass assembly method provided by this invention, after acquiring images of the frameless car door glass at different standard tooling positions, determines the fully automatic installation position and movement displacement of the frameless car door glass through point cloud processing, thereby realizing fully automatic installation by a frameless car door glass assembly robot. The entire process does not require manual labor or the use of multiple sets of collaborative tooling, which can improve assembly speed and accuracy, and is applicable to various models of frameless car door glass. In addition, this invention also provides corresponding devices, equipment, and computer-readable storage media for the frameless car door glass assembly method, further making the above method more practical. These devices, equipment, and computer-readable storage media have corresponding advantages. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0042] Figure 1 A flowchart illustrating the frameless car door glass assembly method provided in this embodiment of the invention;

[0043] Figure 2 This is a schematic diagram of the frameless car door glass assembly device provided in an embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] This invention provides a method for assembling frameless car door glass, such as... Figure 1 As shown, it includes the following steps:

[0046] S101. Obtain images of the frameless door glass to be installed at different standard tooling positions;

[0047] Specifically, after the frameless door glass component to be installed enters the standard fixture position, a camera (such as a 3D camera at the end of a scanning robot) can be used to photograph the component, obtaining an image of the frameless door glass component at that standard fixture position. The standard fixture position can be set based on experience, as long as it covers the entire door and glass.

[0048] S102. Generate multiple point cloud maps based on the acquired images;

[0049] S103. Collect multiple sets of point cloud data from the generated point cloud map, and perform point cloud registration on the collected multiple sets of point cloud data;

[0050] It should be noted that the point cloud data here can be data based on the standard coordinate system of the digital model file (referred to as the digital model coordinate system). The digital model coordinate system has high accuracy and does not change.

[0051] S104. Based on the registration results, determine the installation position and movement displacement of the frameless door glass to be installed.

[0052] After determining the installation position and movement displacement of the frameless door glass to be installed in step S104, the frameless door glass can be installed automatically using an industrial robot based on the determined installation position and movement displacement.

[0053] It should be noted that before executing step S101 to acquire images of the frameless door glass to be installed at different standard tooling positions, the positions of the assembly robot, scanning robot, measuring equipment, and tooling, which are already in place, can be systematically calibrated. After calibration, these devices cannot be moved. Among them, industrial robots such as assembly robots and scanning robots possess the characteristics of high load capacity, high precision, and low latency. The repeatability of conventional industrial robots is generally 0.02mm to 0.03mm. In practical applications, the assembly robot is equipped with a glass clamp, mainly responsible for clamping the glass and adjusting its position; the scanning robot is equipped with a scanner and a tightening gun, mainly responsible for scanning the feature points of the door and glass, determining the glass gripping position, the glass displacement adjustment amount, and the positioning and tightening of the screws on the door glass.

[0054] In the frameless car door glass assembly method provided in the embodiments of the present invention, after acquiring images of the frameless car door glass at different standard tooling positions, the fully automatic installation position and movement displacement of the frameless car door glass are determined by point cloud processing, thereby realizing the fully automatic installation of the frameless car door glass assembly robot. The entire process does not require manual use of multiple sets of collaborative tooling, which can improve the assembly speed and has high precision, and is applicable to various types of frameless car door glass.

[0055] Furthermore, in a specific implementation, in the frameless car door glass assembly method provided in the embodiments of the present invention, before executing step S101 to obtain images of the frameless car door glass to be installed at different standard tooling positions, the method may further include: first, obtaining images of frameless car door glass of different models and colors that have already been installed; then, obtaining multiple sets of point cloud data based on the obtained images of frameless car door glass of different models and colors; finally, after preprocessing the obtained multiple sets of point cloud data, obtaining standard installation template point cloud data corresponding to different models and colors.

[0056] Specifically, the above steps require completing the standard installation posture of frameless door glass of various models and colors, and taking 3D photos of the door and glass with the standard installation posture at multiple standard tooling positions to generate point cloud maps. Multiple sets of point cloud data are obtained from the point cloud maps. After preprocessing steps such as data collection, extraction at a certain ratio, and deletion of redundant and useless parts, the data is used as standard installation template point cloud data for subsequent use.

[0057] In specific implementation, in the frameless car door glass assembly method provided in the embodiments of the present invention, after generating multiple point cloud maps in step S102, the method may further include: first, comparing and identifying the point cloud maps with all standard installation template point cloud data to identify the model and color of the car door component to be installed; then, based on the identified model and color of the frameless car door glass to be installed, finding the standard installation template point cloud data corresponding to the frameless car door glass to be installed.

[0058] Specifically, the point cloud image generated in step S102 is compared and identified with the standard installation template point cloud data to determine the model and color of the frameless door component to be installed. Since all standard installation template point cloud data are classified according to model and color, after determining the model and color of the frameless door component to be installed, the standard template point cloud data to be compared can be determined.

[0059] Furthermore, in a specific implementation, in the frameless door glass assembly method provided in the embodiments of the present invention, step S103, which involves point cloud registration of the collected multiple sets of point cloud data, may specifically include: first, performing multiple registrations of the collected multiple sets of point cloud data using digital model point clouds to obtain multiple registration results; then, obtaining the optimal registration result and its corresponding registration parameters from all the obtained registration results. This can further improve the registration accuracy.

[0060] In practice, the above steps involve multiple registrations of the collected point cloud data sets to obtain multiple registration results. Specifically, this may include: registering the collected point cloud data sets according to pre-set initial registration parameters; comparing the registered point cloud data with standard installation template point cloud data to obtain the first registration result; subtracting a pre-set fixed subtraction value from the initial registration parameters to obtain a new registration parameter; performing registration based on the new registration parameter; comparing the registered point cloud data with standard installation template point cloud data to obtain the second registration result; repeating the step of subtracting a fixed subtraction value from the previous registration parameter to obtain the next registration parameter until the obtained registration parameter is less than or equal to a set threshold registration parameter, at which point the registration operation is stopped, thus obtaining multiple registration results.

[0061] Specifically, firstly, based on experience, set the initial registration parameters A0, B0, C0, D0, etc., and perform the registration operation according to these parameters. Then, compare the registered point cloud with the point cloud data of the standard installation template to obtain the first registration result X0, which is the point cloud overlap rate (0~1).

[0062] Then, based on experience, set fixed subtraction values ​​a, b, c, d... for each registration parameter. Subtract the corresponding fixed subtraction values ​​from the previous registration parameters to obtain new registration parameters A1, B1, C1, D1... Perform the registration operation based on these parameters, compare the registered point cloud with the standard installation template point cloud data, and obtain the second registration result X1.

[0063] Then, repeat the previous step to obtain the registration parameter A. n B n C n D n ...Based on this parameter, a registration operation is performed, and the registered point cloud is compared with the point cloud data of the standard installation template to obtain the (n+1)th registration result X. n .

[0064] Until registration parameter A n B n C n D n ...≤ set threshold registration parameter A 00 B 00 C 00 D 00 ...stop repeating and registering operations, compare the previous registration results, and select the optimal (maximum) registration result and its corresponding registration parameters.

[0065] The specific settings for the registration parameters are shown in Table 1 below:

[0066] Table 1

[0067] Standard deviation 1mm-100mm 10mm 1mm 30mm Sampling interval 0.1mm-10mm 0.1mm 0.1mm 0.1mm Template weights 0-100 100 5 80 Confidence threshold 0-1 0.5 0.1 0.8 Number of iterations 1-100 30 5 50 Minimum standard deviation 0.1mm-1mm 0.1mm 0.1mm 0.3mm

[0068] In practice, the optimal point cloud overlap rate can be 90%. The point cloud registration algorithm of this invention is significantly optimized according to the actual situation, resulting in high registration accuracy.

[0069] Based on this, step S104 determines the installation position and movement displacement of the frameless door glass to be installed according to the registration result. Specifically, it may include: determining the installation position and movement displacement of the frameless door glass to be installed according to the optimal registration result and the corresponding registration parameters.

[0070] Specifically, based on the point cloud data under the optimal registration result and the completed system calibration, the coordinates of the measuring equipment are transformed into the digital model coordinate system. At this time, the actual overall system (i.e. the actual scene) is unified with the system used in the digital model, and the assembly robot can assemble the frameless door glass to be installed according to the data in the digital model coordinate system.

[0071] Based on the same inventive concept, this invention also provides a frameless car door glass assembly device. Since the principle of this device in solving the problem is similar to that of the aforementioned frameless car door glass assembly method, the implementation of this device can refer to the implementation of the frameless car door glass assembly method, and the repeated parts will not be described again.

[0072] In specific implementation, the frameless door glass assembly device provided in this embodiment of the invention, such as... Figure 2 As shown, it specifically includes:

[0073] Image acquisition module 11 is used to acquire images of the frameless door glass to be installed at different standard tooling positions;

[0074] The point cloud image generation module 12 is used to generate multiple point cloud images based on the acquired images;

[0075] The point cloud registration module 13 is used to collect multiple sets of point cloud data from the generated point cloud map and to perform point cloud registration on the collected multiple sets of point cloud data.

[0076] The installation position determination module 14 is used to determine the installation position and movement displacement of the frameless door glass to be installed based on the registration results.

[0077] In the frameless car door glass assembly device provided in the embodiments of the present invention, the interaction of the above four modules can be used to obtain images of the frameless car door glass at different standard tooling positions. Then, point cloud processing is used to determine the fully automatic installation position and movement displacement of the frameless car door glass, thereby realizing the fully automatic installation of the frameless car door glass assembly robot. The whole process does not require manual use of multiple sets of collaborative tooling, which improves the assembly speed and has high precision. It is suitable for various types of frameless car door glass.

[0078] In a specific implementation, in the frameless car door glass assembly device provided in the embodiments of the present invention, the point cloud registration module 13 can be used to perform multiple registrations of digital model point clouds on multiple sets of collected point cloud data to obtain multiple registration results; and obtain the optimal registration result and the corresponding registration parameters from all the obtained registration results.

[0079] Specifically, the point cloud registration module 13 can be used to perform registration operations on multiple sets of collected point cloud data according to the preset initial registration parameters, compare the registered point cloud data with the standard installation template point cloud data to obtain the first registration result; subtract the preset single fixed subtraction value from the initial registration parameters to obtain a new registration parameter, perform registration operations according to the new registration parameter, compare the registered point cloud data with the standard installation template point cloud data to obtain the second registration result; repeat the step of subtracting the single fixed subtraction value from the previous registration parameter to obtain the next registration parameter until the obtained registration parameter is less than or equal to the set threshold registration parameter, and stop the registration operation to obtain multiple registration results.

[0080] In a specific implementation, in the frameless car door glass assembly device provided in the embodiments of the present invention, the installation position determination module 14 can be used to determine the installation position and moving displacement of the frameless car door glass to be installed based on the optimal registration result and the corresponding registration parameters.

[0081] In a specific implementation, in the frameless car door glass assembly device provided in the embodiments of the present invention, the image acquisition module 11 can also be used to acquire images of frameless car door glass of different models and colors that have been installed.

[0082] The point cloud image generation module 12 can also be used to generate multiple first point cloud images based on the acquired images of frameless car door glass of different models and colors.

[0083] The point cloud registration module 13 can also be used to obtain multiple sets of point cloud data from the generated first point cloud map; it may also include:

[0084] The template data acquisition module is used to preprocess multiple sets of point cloud data to obtain standard installation template point cloud data corresponding to different models and colors.

[0085] In specific implementations, the frameless car door glass assembly device provided in the embodiments of the present invention may further include:

[0086] The comparison and recognition module is used to compare and recognize the point cloud map with the point cloud data of all standard installation templates to identify the model and color of the door component to be installed.

[0087] The template query module is used to find the standard installation template point cloud data corresponding to the frameless door glass to be installed based on the identified model and color of the frameless door glass to be installed.

[0088] For more detailed information on the working process of each of the above modules, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0089] Accordingly, this invention also discloses a frameless car door glass assembly device, including a processor and a memory; wherein, when the processor executes the computer program stored in the memory, it implements the frameless car door glass assembly method disclosed in the foregoing embodiments.

[0090] For more detailed information on the above methods, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0091] Furthermore, the present invention also discloses a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the aforementioned frameless door glass assembly method.

[0092] For more detailed information on the above methods, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0093] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatuses, devices, and storage media disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0094] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0095] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0096] In summary, the frameless car door glass assembly method provided by this invention includes: acquiring images of the frameless car door glass to be installed at different standard tooling positions; generating multiple point cloud images based on the acquired images; collecting multiple sets of point cloud data from the generated point cloud images and performing point cloud registration on the collected sets of point cloud data; and determining the installation position and movement displacement of the frameless car door glass to be installed based on the registration result. After acquiring images of the frameless car door glass at different standard tooling positions, the above method determines the fully automatic installation position and movement displacement of the frameless car door glass through point cloud processing, thereby realizing fully automatic installation by a frameless car door glass assembly robot. The entire process does not require manual labor or the use of multiple sets of collaborative tooling, which can improve assembly speed and accuracy, and is applicable to various models of frameless car door glass. Furthermore, this invention also provides corresponding devices, equipment, and computer-readable storage media for the frameless car door glass assembly method, further enhancing the practicality of the above method. These devices, equipment, and computer-readable storage media have corresponding advantages.

[0097] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0098] The frameless car door glass assembly method, apparatus, equipment, and storage medium provided by the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for assembling frameless car door glass, characterized in that, include: Acquire images of frameless door glass of different models and colors that have been installed; generate multiple first point cloud images based on the acquired images of frameless door glass of different models and colors; obtain multiple sets of point cloud data from the generated first point cloud images; After preprocessing the obtained multiple sets of point cloud data, standard installation template point cloud data corresponding to different models and colors are obtained; Acquire images of the frameless door glass to be installed at different standard tooling positions; generate multiple point cloud maps based on the acquired images; The generated point cloud map is compared and identified with all standard installation template point cloud data to identify the model and color of the frameless door glass to be installed; based on the identified model and color of the frameless door glass to be installed, the corresponding standard installation template point cloud data is found. Multiple sets of point cloud data are collected from the generated point cloud map, and the collected sets of point cloud data are registered according to the preset initial registration parameters. The registered point cloud data is compared with the found standard installation template point cloud data to obtain the first registration result. The preset single fixed subtraction value is subtracted from the initial registration parameters to obtain the new registration parameters. Based on the new registration parameters, a registration operation is performed. The registered point cloud data is compared with the point cloud data of the found standard installation template to obtain a second registration result. The step of subtracting the fixed subtraction value from the previous registration parameter to obtain the next registration parameter is repeated until the obtained registration parameter is less than or equal to the set threshold registration parameter. The registration operation is then stopped to obtain multiple registration results. The optimal registration result and its corresponding registration parameters are obtained from all the obtained registration results. Based on the optimal registration result and the corresponding registration parameters, the installation position and moving displacement of the frameless door glass to be installed are determined. Based on the determined installation position and movement displacement of the frameless car door glass to be installed, an industrial robot is used to automatically install the frameless car door glass.

2. A frameless car door glass assembly device, characterized in that, include: The image acquisition module is used to acquire images of frameless car door glass of different models and colors that have been installed; it is also used to acquire images of frameless car door glass to be installed at different standard tooling positions. The point cloud image generation module is used to generate multiple first point cloud images based on the acquired images of frameless car door glass of different models and colors; it is also used to generate multiple point cloud images based on the acquired images. The point cloud registration module is used to obtain multiple sets of point cloud data from the generated first point cloud map; The template data acquisition module is used to preprocess multiple sets of point cloud data to obtain standard installation template point cloud data corresponding to different models and colors. The comparison and recognition module is used to compare and recognize the generated point cloud map with the point cloud data of all standard installation templates, and to identify the model and color of the frameless door glass to be installed. The template query module is used to find the standard installation template point cloud data corresponding to the frameless door glass to be installed based on the identified model and color of the frameless door glass to be installed. The point cloud registration module is further configured to collect multiple sets of point cloud data from the generated point cloud map, and perform registration operations on the collected multiple sets of point cloud data according to a preset initial registration parameter. The registered point cloud data is then compared with the found standard installation template point cloud data to obtain a first registration result. A new registration parameter is obtained by subtracting a preset fixed subtraction value from the initial registration parameter, and registration is performed based on the new registration parameter. The registered point cloud data is then compared with the found standard installation template point cloud data to obtain a second registration result. The process of subtracting the fixed subtraction value from the previous registration parameter to obtain the next registration parameter is repeated until the obtained registration parameter is less than or equal to a preset threshold registration parameter, at which point the registration operation stops, resulting in multiple registration results. Finally, the optimal registration result and its corresponding registration parameter are obtained from all the obtained registration results. The installation position determination module is used to determine the installation position and moving displacement of the frameless car door glass to be installed based on the optimal registration result and the corresponding registration parameters; and to use an industrial robot to automatically install the frameless car door glass according to the determined installation position and moving displacement.

3. A frameless car door glass assembly device, characterized in that, It includes a processor and a memory, wherein the processor executes a computer program stored in the memory to implement the frameless door glass assembly method as described in claim 1.

4. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the frameless door glass assembly method as described in claim 1.

Images