Sunroof production assembly control method and system based on a robot collaboration network

Abstract

The application provides a sunroof production assembly control method and system based on a robot cooperation network, relates to the technical field of intelligent control, and comprises the following steps: obtaining a customized production sunroof part, performing alignment pre-assembly, and obtaining a pre-assembled sunroof part; starting a distributed cooperation robot; constructing a force cooperation control model, obtaining cooperation pressure parameters and cooperation suction force parameters; a pressure robot above performs pressure control on the pre-assembled sunroof part according to the cooperation pressure parameters, a suction force robot below performs suction force control on the pre-assembled sunroof part according to the cooperation suction force parameters, and a completed sunroof part is obtained. The application solves the technical problem that when multiple robots work together, a global coordination mechanism is lacking, robot movement is prone to mismatch, and the sunroof assembly efficiency is affected. Through the cooperation of the double-press assembly robots, high-precision and high-efficiency assembly of the sunroof assembly is realized, and the product quality and production efficiency of the sunroof assembly are improved.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology, and in particular to a method and system for controlling the production and assembly of sunroofs based on a robot collaborative network. Background Technology

[0002] Sunroofs have complex structures involving various materials (such as glass, metal frames, and rubber seals), requiring high assembly precision. This necessitates ensuring accurate alignment of the glass and frame, uniform sealing of the seals, and stable connection of the power mechanism. With the development of industrial automation and intelligent manufacturing, robotics has been introduced into sunroof assembly. However, this typically employs an independent control mode, where each robot performs its own task, lacking global scheduling and information sharing mechanisms. This leads to conflicting motion trajectories or asynchronous operations, reducing overall production efficiency.

[0003] In summary, existing technologies suffer from the technical problem that the lack of a global coordination mechanism when multiple robots work together can easily lead to mismatched robot movements, affecting the efficiency of sunroof assembly. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for controlling the production and assembly of sunroofs based on a robot collaborative network, in order to solve the technical problem in the prior art that the lack of a global coordination mechanism when multiple robots work together easily leads to mismatched robot movements, which affects the assembly efficiency of sunroofs.

[0005] In view of the above problems, this application provides a method and system for controlling the production and assembly of sunroofs based on a robot collaborative network.

[0006] In a first aspect, this application provides a sunroof production assembly control method based on a robot collaborative network. This method is implemented through a sunroof production assembly control system based on a robot collaborative network. The method includes: acquiring a custom-made sunroof component; aligning and pre-assembling the component according to its assembly position to obtain a pre-assembled sunroof component; activating distributed collaborative robots, including an upper pressure robot and a lower suction robot, which are synchronously controlled through a distributed control architecture; constructing a force collaboration control model; acquiring collaborative pressure parameters and collaborative suction parameters based on the model; the upper pressure robot applying pressure control to the pre-assembled sunroof component according to the collaborative pressure parameters; and the lower suction robot applying suction control to the pre-assembled sunroof component according to the collaborative suction parameters to obtain the assembled sunroof component.

[0007] Optionally, the force collaboration control model is connected to the assembly path planning module. It acquires assembly images of the pre-assembled sunroof component through a vision extraction unit; performs pixel recognition on the assembly images to obtain a first pixel dataset and a second pixel dataset; acquires first unit coverage pixel data corresponding to the upper pressure robot and second unit coverage pixel data corresponding to the lower suction robot; the assembly path planning module performs assembly path planning on the first pixel dataset and the second pixel dataset based on the first unit coverage pixel data and the second unit coverage pixel data to obtain an assembly planning path; and controls the upper pressure robot and the lower suction robot according to the assembly planning path.

[0008] Optionally, the assembly path planning module uses the first unit coverage pixel data as the pixel step size to separate the first pixel dataset to obtain a first group of path nodes; the assembly path planning module uses the second unit coverage pixel data as the pixel step size to separate the second pixel dataset to obtain a second group of path nodes; the first group of path nodes is connected by trajectory to obtain a first planned path, the second group of path nodes is connected by trajectory to obtain a second planned path, and the first planned path and the second planned path are combined to generate an assembly planning path, wherein the first planned path and the second planned path are not synchronized.

[0009] Optionally, mean pixel data is determined based on the first unit coverage pixel data and the second unit coverage pixel data; the mean pixel data is used as the pixel step size to separate the first pixel dataset and the second pixel dataset to obtain a first group of path nodes and a second group of path nodes; the path nodes in the first group of path nodes or the second group of path nodes are connected according to a preset trajectory to obtain a synchronous planning path; and an assembly planning path is generated based on the synchronous planning path.

[0010] Optionally, the upper pressure robot includes a six-degree-of-freedom (DOF) pressurizing robotic arm, a pressurizing platform, a pressure control module, and a force sensor. The collaborative pressure parameters are sent to the six-DOF pressurizing robotic arm, and the pressure control module controls the six-DOF pressurizing robotic arm to drive the pressurizing platform to control the pressure of the pre-assembled sunroof component. The force sensor monitors the pressure and uploads the obtained pressure sensing data to the pressure control module, which then provides force feedback to the six-DOF pressurizing robotic arm.

[0011] Optionally, the suction robot below includes a negative pressure generator, a vacuum adsorption platform, and a suction control module; the cooperative suction parameters are sent to the suction control module, and the suction control module controls the suction of the suction cups on the vacuum adsorption platform according to the negative pressure generator.

[0012] Optionally, the suction cups on the vacuum adsorption platform are arranged in a honeycomb array.

[0013] Optionally, a force sample dataset of the distributed collaborative robot is obtained, wherein the force sample dataset includes a pressure sample dataset of the upper pressure robot and a suction sample dataset of the lower suction robot, a difference dataset corresponding to the pressure sample dataset and the suction sample dataset, and label information characterizing assembly errors; parameters of the impedance control model are defined, including the initial stiffness, initial damping, and initial mass parameters of the upper pressure robot and the lower suction robot; the defined impedance control model is trained based on the force sample dataset, and the force collaborative control model is output when the model converges.

[0014] Optionally, material and geometric information of the sunroof component are collected; force protection analysis is performed based on the material and geometric information, a force protection threshold is output, and the force protection threshold is used as a constraint condition for training the impedance control model to obtain a force cooperative control model.

[0015] Secondly, this application also provides a sunroof production assembly control system based on a robot collaborative network, used to execute the sunroof production assembly control method based on a robot collaborative network as described in the first aspect. The sunroof production assembly control system based on a robot collaborative network includes: a sunroof component acquisition module, used to acquire customized sunroof components, align and pre-assemble them according to their assembly positions, and acquire pre-assembled sunroof components; a robot activation module, used to activate distributed collaborative robots, including an upper pressure robot and a lower suction robot, which are synchronously controlled through a distributed control architecture; a control model construction module, used to construct a force collaboration control model, and acquire collaborative pressure parameters and collaborative suction parameters based on the force collaboration control model; and a sunroof component assembly module, used for the upper pressure robot to perform pressure control on the pre-assembled sunroof component according to the collaborative pressure parameters, and the lower suction robot to perform suction control on the pre-assembled sunroof component according to the collaborative suction parameters, to acquire the assembled sunroof component.

[0016] One or more technical solutions provided in this application have at least the following beneficial effects:

[0017] By acquiring custom-made sunroof components, pre-assembling them according to their assembly positions, and obtaining pre-assembled sunroof components, a distributed collaborative robot system is activated. This system includes an upper pressure robot and a lower suction robot, which are synchronously controlled through a distributed control architecture. A force collaboration control model is constructed, and collaborative pressure and suction parameters are obtained based on this model. The upper pressure robot applies pressure control to the pre-assembled sunroof component according to the collaborative pressure parameters, and the lower suction robot applies suction control to the pre-assembled sunroof component according to the collaborative suction parameters, resulting in the assembled sunroof component. In other words, by setting up dual-pressure assembly robots—one pressure robot above the sunroof and one suction robot below—the two robots flexibly cooperate to perform assembly after the sunroof is aligned, achieving precise collaboration and improving the product quality and production efficiency of sunroof assembly.

[0018] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the sunroof production assembly control method based on a robot collaborative network according to this application.

[0021] Figure 2 This is a schematic diagram of the sunroof production assembly control system based on a robot collaborative network, as described in this application.

[0022] Figure labeling: 11, Sunroof component acquisition module; 12, Robot startup module; 13, Control model construction module; 14, Sunroof component assembly module. Detailed Implementation

[0023] This application provides a sunroof production assembly control method and system based on a robot collaborative network, solving the technical problem in existing technologies where the lack of a global coordination mechanism when multiple robots work together easily leads to mismatched robot movements, affecting sunroof assembly efficiency. By setting up a dual-pressure assembly robot system, with one pressure robot distributed on the sunroof and one suction robot distributed below, the two robots flexibly cooperate to perform assembly after the sunroof is aligned, achieving precise coordination and improving product quality and production efficiency in sunroof assembly.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a sunroof production assembly control method based on a robot collaborative network, wherein the sunroof production assembly control method based on a robot collaborative network is executed by a sunroof production assembly control system based on a robot collaborative network, and the sunroof production assembly control method based on a robot collaborative network specifically includes the following steps:

[0026] S100: Obtain the customized sunroof component, align and pre-assemble it according to the assembly position of the sunroof component, and obtain the pre-assembled sunroof component.

[0027] Specifically, sunroof components are customized according to the specific vehicle model required by the order and customer needs. Different vehicle models require sunroofs of different specifications, therefore it is necessary to determine the sunroof parts that need to be produced. These sunroof parts will have data regarding their assembly positions included during the manufacturing process. The sunroof part is a major component of the sunroof, including glass, metal frame, sealing strips, motor, etc. During assembly, these components need to be precisely matched to ensure final airtightness and stability. The sunroof parts are transported to the assembly line, moved to the assembly position for alignment and preliminary fixation, resulting in pre-assembled sunroof parts. In other words, the sunroof parts are placed in the assembly position and initially secured using temporary fixing devices. Pre-assembly ensures precise alignment of the sunroof parts, improving assembly accuracy and efficiency, and laying the foundation for subsequent formal assembly.

[0028] S200: Activate the distributed collaborative robot, which includes an upper pressure robot and a lower suction robot, and the upper pressure robot and the lower suction robot are synchronously controlled through a distributed control architecture.

[0029] Specifically, a dual-pressure assembly robot system is used, consisting of an upper pressure robot and a lower suction robot, one positioned above the sunroof and the other below. Once the sunroof is aligned, the two robots work together flexibly to perform assembly. The distributed collaborative robot system is activated, meaning the two robots coordinate with each other through a distributed control system to jointly execute complex tasks. The robot control center reads the sunroof component's specifications and assigns tasks to the two collaborative robots, including the upper pressure robot and the lower suction robot. The upper pressure robot applies pressure from above to ensure a tight fit between the sunroof component and the vehicle body or other parts, while the lower suction robot provides suction and fixation from below to ensure the sunroof component does not slip or shift during assembly.

[0030] Through a distributed control architecture, distributed collaborative robots are synchronously controlled to ensure coordinated movements when performing tasks. The central control center sends synchronous control commands to both robots. The upper pressure robot applies downward pressure according to the command, while the lower suction robot simultaneously applies upward suction. Force sensors and other sensors monitor the movements and applied forces of both robots in real time to ensure they operate according to preset parameters. The upper pressure robot and lower suction robot apply pressure and suction simultaneously according to commands to ensure the sunroof component is smoothly and precisely assembled onto the car body.

[0031] For example, different robots were used to assemble sunroofs. The test data are as follows: The alignment error of the traditional single robot assembly was ±1.2mm, the assembly success rate was 87%, the installation time was 10s, and the maximum force was 550N; the alignment error of the distributed collaborative robot without synchronous control was ±0.8mm, the assembly success rate was 92%, the installation time was 7s, and the maximum force was 520N; the alignment error of the distributed collaborative robot with synchronous control was ±0.2mm, the assembly success rate was 99.5%, the installation time was 5s, and the maximum force was 490N. The test results show that the assembly accuracy of the distributed collaborative robot is 83% higher than that of the single robot, the installation time is shortened by 50%, the risk of glass breakage is reduced, and the assembly stability is improved.

[0032] Furthermore, this application S200 includes:

[0033] The force collaboration control model is connected to the assembly path planning module. It acquires assembly position images of the pre-assembled sunroof component through a vision extraction unit to obtain assembly images. Pixel recognition is performed on the assembly images to obtain a first pixel dataset and a second pixel dataset. First unit coverage pixel data corresponding to the upper pressure robot and second unit coverage pixel data corresponding to the lower suction robot are acquired respectively. The assembly path planning module performs assembly path planning on the first and second pixel datasets based on the first and second unit coverage pixel data to obtain an assembly planning path. The upper pressure robot and the lower suction robot are controlled according to the assembly planning path.

[0034] Specifically, the force collaboration control model is mainly responsible for calculating and adjusting the forces applied by the robot in real time during the assembly process (such as the pressure applied by the upper pressure robot and the suction force generated by the lower suction robot), ensuring that the sunroof component is subjected to uniform force during assembly and that it will not be displaced or damaged due to excessive or insufficient force in certain areas. The assembly path planning module uses visually acquired data to analyze the edge pixel data of the pre-assembled sunroof component and, based on the robot's unit coverage pixel data, plans an optimal assembly path that includes both the geometric trajectory of the robot's movement and the assembly accuracy requirements to be achieved at each key node.

[0035] The visual extraction unit first acquires images of the pre-assembly position of the sunroof component. Through edge detection and pixel recognition, it generates upper surface edges (first pixel dataset) and lower surface edges (second pixel dataset), respectively. Simultaneously, it acquires unit coverage pixel data from the robot ends (upper pressure robot and lower suction robot). The assembly path planning module uses the unit coverage pixel data to segment the first and second pixel datasets according to a preset pixel step size, generating two sets of discrete path nodes. These two sets of nodes are then connected to generate the first and second planned paths.

[0036] The vision extraction unit acquires images of the pre-assembled sunroof components at their assembly locations, resulting in an assembly image of the pre-assembled sunroof components. The vision extraction unit is a visual sensing unit used to acquire image information, including devices such as industrial cameras, 3D laser scanners, and depth cameras. Images of the sunroof components are acquired using an industrial camera or 3D laser scanner. The assembly image refers to the visual data of the sunroof assembly area captured by the vision extraction unit, including key information such as the sunroof glass, mounting base, and sealing strips.

[0037] Image processing algorithms were used to perform pixel recognition on the assembly image, extracting a first pixel dataset (pixels at the upper surface edge of the sunroof assembly) and a second pixel dataset (pixels at the lower surface edge of the sunroof assembly). First, the image was preprocessed to remove noise and irrelevant information, enhancing edge features for subsequent pixel recognition. Edge detection algorithms (such as Canny edge detection) were used to identify the edges of the upper and lower surfaces of the sunroof component. The identified edge pixels were then classified into the first pixel dataset (upper surface edges) and the second pixel dataset (lower surface edges).

[0038] Based on the robot's working range and assembly requirements, determine the respective operating areas for the upper and lower robots. Calculate the number of pixels per unit area based on the area of ​​each operating area and the image resolution. Obtain the first unit coverage pixel data for the upper pressure robot, i.e., the pixel data per unit area within its operating area. Obtain the second unit coverage pixel data for the lower suction robot, i.e., the pixel data per unit area within its operating area.

[0039] The assembly path planning module separates the first pixel dataset and the second pixel dataset based on the unit coverage pixel data, and then generates the assembly planning path based on the generated path nodes. Here, a synchronous or asynchronous strategy can be selected according to requirements. In the synchronous strategy, the upper and lower robots perform assembly actions simultaneously along the same path. In the asynchronous strategy, the lower suction robot starts first to complete the adsorption and positioning, while the upper pressure robot starts after a delay to apply pressure, avoiding mutual interference.

[0040] Taking the asynchronous strategy as an example, the first unit of coverage pixel data is used as the step size to segment the first pixel dataset, resulting in the first set of path nodes. These represent the key positions where the upper pressure robot needs to stop or operate during the assembly process. Similarly, the assembly path planning module uses the first unit of coverage pixel data as the pixel step size to separate the second pixel dataset, resulting in the second set of path nodes. These represent the key positions where the lower suction robot needs to stop or operate during the assembly process. Using trajectory connection techniques, such as Bézier curve fitting, the first set of path nodes is connected into a continuous, smooth trajectory, i.e., the first planned path; similarly, the second set of path nodes is connected into the second planned path. Finally, the two paths are combined to generate the complete assembly planning path. The first and second planned paths are asynchronous, meaning that the two robots will operate according to different schedules and paths during the assembly process.

[0041] The acquired assembly planning path is transmitted to the robot control center and sent to the corresponding upper pressure robot and lower suction robot. The robots execute the assembly task according to their respective motion commands, ensuring accurate alignment and balanced force distribution on the upper and lower surfaces of the sunroof component during assembly. For example, after generating the assembly planning path using a preset trajectory planning algorithm, experimental results show that: when executed synchronously, the average assembly error between the upper and lower robots is controlled within ±0.25mm; using an asynchronous strategy (lower robot first, upper robot delayed by 0.5 seconds), the overall assembly success rate reaches 99.5%, and the glass breakage rate is reduced to 0.1%. Through the vision extraction unit and assembly path planning module, high-precision, automated assembly of sunroof components can be achieved, significantly improving assembly accuracy and speed, and significantly enhancing the efficiency and quality of the sunroof production line.

[0042] Furthermore, this application also includes the following steps:

[0043] The assembly path planning module uses the first unit coverage pixel data as the pixel step size to separate the first pixel dataset to obtain a first group of path nodes; the assembly path planning module uses the second unit coverage pixel data as the pixel step size to separate the second pixel dataset to obtain a second group of path nodes; the first group of path nodes is connected by trajectory to obtain a first planned path, the second group of path nodes is connected by trajectory to obtain a second planned path, and the first planned path and the second planned path are combined to generate an assembly planning path, wherein the first planned path and the second planned path are not synchronized.

[0044] Specifically, the assembly path planning module first receives image data of the upper surface (first pixel dataset) and lower surface (second pixel dataset) acquired from the vision extraction unit. Using a first unit coverage pixel data (e.g., 50 pixels) as the pixel step size, the first pixel dataset is separated to obtain a first set of path nodes. For example, if the upper surface edge is 1000 pixels long in the image, segmenting it at a 50-pixel step size will yield approximately 20 discrete path nodes. Using a second unit coverage pixel data (e.g., 40 pixels) as the pixel step size, the second pixel dataset is separated to obtain a second set of path nodes. Path nodes refer to discrete key location points obtained by segmenting the image edge data according to a preset step size. These nodes reflect the geometric features of the skylight edge, providing a basis for subsequent trajectory fitting. The first set of path nodes includes key location points extracted from the first pixel dataset, and the second set of path nodes includes key location points extracted from the second pixel dataset.

[0045] The first set of path nodes is connected to form the first planned path. Trajectory connection algorithms are typically used to smoothly connect discrete path nodes into a continuous motion trajectory, such as Bézier curve fitting and spline curve interpolation. These algorithms transform discontinuities between nodes into smooth curves, reducing sharp turns and vibrations during motion. Bézier curve fitting is characterized by its smoothness and ease of control, and is often used in robot motion trajectory planning, resulting in a motion path with good continuity and flexibility. Spline curve interpolation uses spline functions (such as cubic spline interpolation) to interpolate discrete data points, generating a smooth curve.

[0046] Based on specific assembly requirements and node distribution, a suitable smoothing algorithm is selected. Taking Bézier curves as an example, the starting point, ending point, and several intermediate control points are selected based on the first set of path nodes. Then, the coordinates of each point on the entire curve are calculated using the Bézier curve formula. For example, assuming the selected nodes are P0, P1, ..., P... 19 By adjusting the control points, a curve can be generated, ensuring that the curve passes through or closely approximates the original node at each node. The resulting smooth curve serves as the first planned path, guiding the movement of the upper pressure robot. For example, in a car sunroof assembly experiment, an industrial camera was used to capture images of the sunroof's upper surface. After image processing, the pixel data of the upper surface edge was obtained, and 20 discrete nodes were separated with a step size of 50 pixels. Using a Bezier curve fitting algorithm, the generated first planned path showed an average path error of ±0.2mm (deviation from the actual edge measurement), a smooth motion curve, and a curve fitting degree exceeding 95%.

[0047] Similarly, following the steps described above, the second set of path nodes are connected to generate a continuous second planned path. The first planned path guides the upper pressure robot to perform force application actions, while the second planned path guides the lower suction robot to perform adsorption and positioning actions. Since the two sets of data originate from different surfaces of the skylight (upper and lower surfaces), the generated two planned paths may have slight differences in shape and node distribution. The first and second planned paths are combined to form a complete assembly planned path. The first and second planned paths are asynchronous, employing different timing sequences or start times during assembly execution, and do not move completely simultaneously, thus avoiding mutual interference between the two robots during operation.

[0048] The asynchronous planning method mainly involves independently separating path nodes from the pixel data of the upper and lower surfaces, generating separate planned paths, and then combining them into an assembly planned path. Since the two paths are formed independently, an asynchronous strategy is adopted. This method is convenient for optimizing paths for different surfaces, but it requires careful handling of the time coordination between the two paths to avoid interference and assembly errors.

[0049] For example, in the sunroof assembly test, the visual extraction unit collected data on the upper surface edge with a length of 1000 pixels and the lower surface edge with a length of 800 pixels. For the upper surface data, a first unit coverage pixel data step size of 50 pixels was used to obtain approximately 20 path nodes. For the lower surface data, a second unit coverage pixel data step size of 40 pixels was used to obtain approximately 20 path nodes. After using the Bezier curve fitting algorithm, the average assembly error of the first planned path was measured to be ±0.25mm, while the average error of the second planned path was approximately ±0.3mm. The lower suction robot started first to complete the adsorption, and after a delay of 0.5 seconds, the upper pressure robot began to apply pressure. After actual testing, the overall assembly success rate reached 99.2%, and due to the asynchronous strategy, the local stress concentration caused by synchronous force application was effectively reduced, and the glass breakage rate was reduced from the traditional 1.5% to 0.1%. By adopting an asynchronous execution strategy, the upper pressure robot and the lower suction robot are not synchronized during the assembly process, thereby effectively avoiding mechanical interference and operational conflicts and ensuring assembly stability.

[0050] Furthermore, this application also includes the following steps:

[0051] Based on the first unit coverage pixel data and the second unit coverage pixel data, the mean pixel data is determined; the mean pixel data is used as the pixel step size to separate the first pixel dataset and the second pixel dataset to obtain a first group of path nodes and a second group of path nodes; the path nodes in the first group of path nodes or the second group of path nodes are connected according to a preset trajectory to obtain a synchronous planning path, and an assembly planning path is generated based on the synchronous planning path.

[0052] Specifically, the average of the first unit coverage pixel data and the second unit coverage pixel data is calculated to obtain the mean pixel data. This mean pixel data is then used as the pixel stride to separate the first and second pixel datasets, resulting in a first set of path nodes and a second set of path nodes. Unlike the previous methods, the separation stride for the two pixel sets is the same here. Using the mean pixel data as a uniform pixel stride takes into account the working characteristics of both the upper and lower parts while ensuring scale consistency between the two sets of data during segmentation.

[0053] In robot assembly path planning, a preset trajectory refers to the robot's motion path pre-defined during the design phase. This is typically based on the shape, size, and assembly requirements of the vent component, as well as the robot's motion capabilities and accuracy. Based on the shape, size, and assembly requirements of the vent component, the trajectory the robot needs to follow during assembly is analyzed. Based on the analysis results, one or more preset trajectories are designed; these can be straight lines, curves, arcs, etc., depending on the specific assembly requirements. Before practical application, the preset trajectory is validated to ensure its feasibility and effectiveness in the actual assembly process.

[0054] Choose either the first or second group of path nodes and connect them according to the preset trajectory to obtain a synchronized planning path. Due to the use of a uniform step size, the generated path has good continuity and consistency, and the robot can move strictly in sync along this path, ensuring consistent assembly actions. Based on the synchronized planning path, the final assembly planning path is generated and transmitted to the robot control center, enabling the upper pressure robot and the lower suction robot to operate synchronously according to this path, ensuring the synchronization of the assembly process. The final motion trajectory generated based on the synchronized planning path is used to guide the robot's precise movement during assembly, ensuring consistent alignment of the upper and lower edges when assembling the sunroof component.

[0055] The synchronous planning method calculates the mean of the first and second unit coverage pixel data, unifies the pixel step size, segments the two pixel datasets, and selects only one set of nodes for trajectory connection to generate a single synchronous planning path. This method simplifies the path planning process, enabling robots to move in strict synchronization, thereby improving assembly consistency and overall system stability, and is suitable for applications requiring a high degree of consistency in the assembly process. By using the mean pixel data to unify the step size and separating the upper and lower surface datasets, it ensures that the distribution of the two sets of path nodes is consistent. The resulting synchronous planning path enables the robot to maintain high precision during movement, allowing the two robots to work more closely together, thus improving the overall performance of the assembly line.

[0056] Furthermore, this application also includes the following steps:

[0057] The upper pressure robot includes a six-degree-of-freedom (DOF) pressurizing robotic arm, a pressurizing platform, a pressure control module, and a force sensor. The collaborative pressure parameters are sent to the six-DOF pressurizing robotic arm, and the pressure control module controls the six-DOF pressurizing robotic arm to drive the pressurizing platform to control the pressure on the pre-assembled sunroof component. The force sensor monitors the pressure and uploads the obtained pressure sensing data to the pressure control module, which then provides force feedback to the six-DOF pressurizing robotic arm based on the pressure control module's input.

[0058] Specifically, the overhead pressure robot, used to apply pressure above the sunroof during sunroof assembly, mainly consists of a six-DOF pressurizing robotic arm, a pressurizing platform, a pressure control module, and force sensors. The six-DOF pressurizing robotic arm is a robotic arm with six degrees of freedom (i.e., translation in three spatial directions and rotation around the X, Y, and Z axes), enabling flexible attitude adjustment and positioning. The pressurizing platform is typically mounted at the end of the robotic arm to convert the arm's output motion into uniform pressure on the sunroof component. The pressure control module receives and processes pressure parameters from the host computer or collaborative system, controlling the movement of the robotic arm to ensure the pressurizing platform applies the predetermined pressure. Force sensors, mounted on the pressurizing platform or robotic arm, monitor the actual pressure applied to the sunroof component in real time and feed the data back to the pressure control module.

[0059] The robot control center sends the collaborative pressure parameters obtained from the force collaboration control model, based on the assembly task requirements, to the six-degree-of-freedom (DOF) pressurizing robotic arm and the pressure control module. The pressure control module then sends control commands to the six-DOF pressurizing robotic arm based on these parameters. The six-DOF pressurizing robotic arm drives the pressurizing platform to apply pressure to the pre-assembled sunroof component. In other words, driven by the control commands, the pressurizing robotic arm drives the pressurizing platform to apply pressure to the pre-assembled sunroof component, ensuring a tight fit between the sunroof component and the assembly base or other components. Simultaneously, force sensors begin real-time monitoring of the actual pressure applied to the pre-assembled sunroof component, obtaining pressure sensing data.

[0060] The monitored pressure sensor data is uploaded to the pressure control module for pressure feedback. The pressure control module compares the actual pressure value with the target pressure parameter. If there is a deviation between the actual pressure and the target, the movement of the robotic arm is adjusted through closed-loop control, thereby achieving force feedback control and ensuring that the final applied pressure is consistent with the collaborative pressure parameter. Through closed-loop control, the upper pressure robot can ensure that precise and stable pressure is applied to the sunroof component during assembly, improving assembly accuracy and reliability. This not only increases production efficiency but also reduces human error and improves the overall quality of the product.

[0061] Furthermore, this application also includes the following steps:

[0062] The suction robot below includes a negative pressure generator, a vacuum adsorption platform, and a suction control module; the cooperative suction parameters are sent to the suction control module, and the suction control module controls the suction of the suction cups on the vacuum adsorption platform according to the negative pressure generator.

[0063] The suction cups on the vacuum adsorption platform are arranged in a honeycomb array.

[0064] Specifically, the suction robot is used during sunroof assembly to fix pre-assembled sunroof components from below using suction force. It mainly consists of a negative pressure generator, a vacuum suction platform, and a suction control module. The negative pressure generator is a device that generates negative pressure (vacuum) to provide suction for the vacuum suction platform. It lowers atmospheric pressure to create the necessary negative pressure environment for suction. The vacuum suction platform is the working platform mounted on the suction robot, equipped with suction cups for direct contact and fixation of the sunroof components. The suction control module receives cooperative suction parameters and precisely controls the suction cups on the vacuum suction platform to generate appropriate suction based on the negative pressure provided by the generator. It typically receives external commands and adjusts the suction generated by the negative pressure generator. On the vacuum suction platform, the suction cups are arranged in a honeycomb pattern, ensuring even distribution within the assembly area. This increases the contact area, improves suction uniformity and stability, disperses localized negative pressure for better overall suction, and allows for flexibility when suctioning sunroof components of different shapes and sizes.

[0065] Based on the assembly task and the force collaboration control model, the collaborative suction parameters are obtained and sent to the suction control module of the lower suction robot. Upon receiving the parameters, the suction control module sends a command to the negative pressure generator to adjust the negative pressure output. The negative pressure generator then begins operation, generating negative pressure and transmitting it through pipelines to the suction cups on the vacuum adsorption platform. The suction cups on the platform are arranged in a honeycomb array, ensuring uniform distribution throughout the assembly area and providing balanced suction force. When the suction force reaches the set value, the suction cups begin to adsorb the sunroof component. The suction control module ensures that the suction cups adsorb the sunroof component with uniform suction force, preventing deformation or damage due to uneven suction. Once the sunroof component is firmly adsorbed, the lower suction robot can work in conjunction with the upper pressure robot, operating according to the planned assembly path to complete the assembly of the sunroof component.

[0066] The suction control module monitors the adsorption status of the suction cups and adjusts the output of the negative pressure generator based on feedback information to ensure that the adsorption force of each suction cup reaches the set target, thereby stably fixing the pre-assembled sunroof component. The collected actual negative pressure data is fed back to the suction control module to achieve closed-loop control, ensuring that the actual suction force remains consistent with the cooperative suction force parameters. For example, on a car sunroof assembly line, a suction robot below generates negative pressure through a negative pressure generator and distributes it to the honeycomb-arranged suction cups on a vacuum adsorption platform. The cooperative suction force parameters are set to maintain a negative pressure of at least 60 kPa at the suction cups. After receiving the target suction force parameters, the suction control module adjusts the output of the negative pressure generator using a PID algorithm. In the experiment, the initial negative pressure was 55 kPa, which stabilized within the range of 60 ± 1 kPa after closed-loop adjustment. Using a honeycomb-arranged suction cup pattern improves the uniformity of adsorption across the entire adsorption area, ensuring that the sunroof component does not shift due to poor local adsorption during assembly. The results show that the data transmission delay is less than 10 ms and the control response time is 0.3 s. Through precise suction control, the lower suction robot can work closely with the upper pressure robot to achieve high-precision assembly of sunroof components, which not only improves production efficiency but also reduces the uncertainty of manual operation and ensures the consistency of product quality.

[0067] Furthermore, this application S300 includes:

[0068] Obtain the force sample dataset of the distributed collaborative robot, wherein the force sample dataset includes the pressure sample dataset of the upper pressure robot and the suction sample dataset of the lower suction robot, the difference dataset corresponding to the pressure sample dataset and the suction sample dataset, and the label information characterizing the assembly error; define the parameters of the impedance control model, including the initial stiffness, initial damping, and initial mass parameters of the upper pressure robot and the lower suction robot; train the defined impedance control model according to the force sample dataset, and output the force collaborative control model when the model converges.

[0069] Specifically, the system acquires force sample datasets for the distributed robots, including pressure sample datasets from the upper pressure robot and suction sample datasets from the lower suction robot. This consists of data on the pressure values ​​(in Newtons) applied by the upper pressure robot during assembly, and data on the suction forces generated by the lower suction robot during assembly. Additionally, it includes a difference dataset between the pressure and suction sample datasets, reflecting the coordinated force differences between the two robots during assembly. Labeling information characterizing assembly errors involves quantifying and annotating errors generated during assembly (such as alignment errors, deformation caused by uneven force, etc.), identifying the assembly error conditions corresponding to the force sample datasets, such as assembly deviation and assembly speed.

[0070] Using the force sensors built into each robot (such as the force sensor of the upper pressure robot and the negative pressure sensor of the lower suction robot), the applied pressure and suction force are recorded in real time during the assembly process. The collected raw data is preprocessed, including filtering, normalization, and outlier removal. The difference between the corresponding data of the upper and lower robots is calculated (e.g., pressure value minus suction value or the ratio of the two), generating difference data reflecting the degree of coordination between the two. Each data sample is labeled according to assembly error detection (e.g., using vision to detect assembly deviations) for supervised learning.

[0071] Initial stiffness, damping, and mass parameters are defined for the upper pressure robot and the lower suction robot, respectively. Initial stiffness affects the robot's response to force; higher stiffness causes the robot to decelerate rapidly when encountering resistance. Initial damping reduces system oscillations and improves stability. Initial mass affects the robot's dynamic response; a larger mass results in slower acceleration and deceleration. Impedance control is a control strategy designed to provide the robot with good compliance and force control performance when in contact with the workpiece. Its core idea is to treat the robot system as a mass-damped-stiffness system, adjusting model parameters (such as stiffness, damping, and mass) to achieve an active response to external forces, thereby achieving stable assembly and precise force application. The initial stiffness, initial damping, and initial mass parameters are the initial parameters of the impedance control model, used to define the robot's initial dynamic characteristics.

[0072] An impedance control model is established and trained using a force sample dataset (including pressure samples, suction samples, difference data, and label information). A regression model is constructed using a deep learning framework, with model parameters (stiffness, damping, and mass) as variables to be optimized. Gradient descent (or the Adam optimizer) is used iteratively to minimize the mean squared error (MSE) between the predicted and actual assembly errors. The MSE is used as the loss function to calculate the difference between the model's predicted assembly error and the actual assembly error. The goal is to minimize the MSE, meaning the optimized model can accurately predict the assembly error. In each training step, the gradient of the loss function with respect to the model parameters (including stiffness, damping, and mass) is calculated using the backpropagation algorithm, and the parameters are then updated to reduce the loss. The training process is repeated until the loss function decreases to a preset convergence threshold, indicating that the model parameters have stabilized and the model has converged.

[0073] When the training process meets the convergence condition (e.g., MSE is below a certain threshold or the training loss changes very little over several consecutive epochs), the trained force cooperation control model is output. This model can adjust the cooperative force between the upper pressure robot and the lower suction robot in real time during actual assembly, achieving precise and dynamic force feedback control.

[0074] For example, in a car sunroof assembly test, the following data was collected by force sensors: Upper pressure robot pressure sample data: [490,495,500,505,510] N; Lower suction robot suction sample data: [58,59,60,61,62] kPa; Difference dataset (after unit conversion): [432,435,440,445,448] N; Assembly error label (obtained through visual inspection): [0.3,0.25,0.2,0.22,0.2] mm. Initial impedance parameters were set as follows: Upper: stiffness 1000 N / m, damping 50 N·s / m, mass 200 kg; Lower: stiffness 800 N / m, damping 40 N·s / m, mass 150 kg. After 2000 training iterations, the model's mean square error on the validation set decreased to ±0.05 mm, indicating that the model had converged. By training and adjusting the impedance control model, the stability and accuracy of the robot assembly process can be improved, thereby enhancing assembly precision and efficiency.

[0075] Furthermore, this application also includes the following steps:

[0076] The material and geometric information of the sunroof component is collected; force protection analysis is performed based on the material and geometric information, and a force protection threshold is output. The force protection threshold is used as a constraint condition for training the impedance control model to obtain a force cooperative control model.

[0077] Specifically, using sensors, industrial cameras, laser scanners, and product design documents, the material properties (such as elastic modulus, yield strength, and fracture toughness) and geometric parameters (such as thickness, dimensions, and curvature) of the sunroof components are obtained. Material information refers to the physical and chemical properties of the materials constituting the sunroof, such as material type (e.g., steel, aluminum, glass, plastic), strength, elastic modulus, and fracture toughness, which determine the sunroof component's load-bearing capacity and safety range under stress. Geometric information refers to the sunroof component's dimensions, shape, thickness, area, and other geometric parameters, directly affecting the stress distribution and deformation characteristics of the sunroof component under stress.

[0078] The sunroof component is simulated using finite element method (FEM) software (such as ANSYS or ABAQUS). The stress state is analyzed based on material and geometric information. In other words, the material and geometric information of the sunroof component is input into the existing FEM software for simulation, mimicking the assembly process. The simulation results determine the maximum allowable force value of the sunroof component during assembly or under stress, and assess the maximum and minimum forces that the sunroof component can safely withstand at different assembly steps to avoid damage. For example, the safe operating force of the sunroof component under the maximum allowable stress is calculated to be 500 N (or other units). This safe operating force is used as a force protection threshold and a constraint during subsequent training to ensure that the model's predictions and control outputs do not exceed this threshold during optimization.

[0079] During model training, a force protection threshold is used as a constraint. A loss function or regularization term is designed to ensure that the model does not produce predicted outputs exceeding the safety threshold when adjusting stiffness, damping, and mass parameters. Then, following the detailed training process described above, the model parameters are continuously adjusted to minimize the mean square error, ensuring the predicted output matches the actual assembly error while satisfying the force protection threshold constraint. When the training loss reaches the preset convergence criterion and all predicted outputs meet the safety constraints, the final force cooperation control model is output. Through training, the model learns how to coordinate the force outputs of two robots without exceeding the force protection threshold to complete the assembly task. The model obtained after training convergence is the force cooperation control model, which provides cooperative pressure and cooperative suction parameters. This approach ensures that the trained force cooperation control model not only efficiently completes the assembly task but also protects the sunroof component from damage during assembly.

[0080] S300: Construct a force cooperation control model, and obtain cooperation pressure parameters and cooperation suction parameters based on the force cooperation control model.

[0081] Specifically, based on the collected force sample dataset (including upper pressure, lower suction, the difference between the two, and assembly error labels), a force cooperation control model based on impedance control is constructed through the aforementioned detailed process. The model uses physical parameters such as stiffness, damping, and mass as variables to be optimized, and performs parameter fitting using deep learning (e.g., regressive neural networks) to reflect the relationship between force and error during assembly. Using the trained force cooperation control model, the currently collected force data (e.g., actual applied pressure, suction, etc.) and assembly status information are input in real time during the actual assembly process. The model infers based on the input data and outputs the optimal cooperative pressure parameters and cooperative suction parameters.

[0082] For example, suppose the following data is collected during a sunroof assembly test: Upper pressure data samples: 490N, 495N, 500N, 505N, 510N; Lower suction data samples: converted suction data (approximately 60kPa); Assembly error labels: 0.3mm, 0.25mm, 0.2mm, 0.22mm, 0.2mm. A regression neural network model is constructed, initially setting the upper robot stiffness to 1000N / m, damping to 50N·s / m, and mass to 200kg, and the lower stiffness to 800N / m, damping to 40N·s / m, and mass to 150kg. After 2000 epochs of training, the model's mean square error on the validation set is reduced to approximately ±0.05mm. Once training converges, the model can output cooperative pressure parameters (approximately 500N) and cooperative suction parameters (approximately 60kPa) based on the current real-time force data, guiding the robot to adjust pressure and suction. By constructing a force-cooperative control model, the stability and safety of the robot assembly process can be achieved, thereby improving the automation and intelligence level of the assembly process.

[0083] S400: The upper pressure robot controls the pressure of the pre-assembled sunroof component according to the cooperative pressure parameters, and the lower suction robot controls the suction of the pre-assembled sunroof component according to the cooperative suction parameters, thereby obtaining the assembled sunroof component.

[0084] Specifically, the collaborative pressure parameters are sent to the upper pressure robot, and the collaborative suction parameters are sent to the lower suction robot. The upper pressure robot adjusts its degree-of-freedom pressurizing arm and pressurizing platform according to the collaborative pressure parameters to precisely apply the predetermined pressure, ensuring the upper surface of the sunroof component fits tightly against the assembly base. Simultaneously, the lower suction robot controls its negative pressure generator and the honeycomb suction cups on its vacuum adsorption platform according to the collaborative suction parameters to achieve adsorption and fixation of the lower surface of the sunroof component. The uniform distribution of suction helps prevent the sunroof component from shifting or being damaged due to uneven force during the pressurization process.

[0085] Throughout the pressure application and suction process, both robots monitor the actual force in real time using their respective sensors (such as force sensors and negative pressure sensors) and feed the data back to their respective control modules. The control modules make fine adjustments to the collaborative pressure and suction based on the feedback data to ensure that the actual applied pressure and suction remain stable near the predetermined target, thereby ensuring that the sunroof component can maintain precise alignment under force without exceeding safe operating conditions.

[0086] Once the upper robot applies pressure and the lower robot performs its suction and fixation actions in coordination, the pre-assembled sunroof component is firmly fixed to the assembly base, thus completing the assembly. Throughout the process, the upper and lower robots, under the coordinated action of forces, ensure that all components of the sunroof are evenly stressed and accurately aligned, achieving high assembly precision and stability. For example, the upper pressure robot applies pressure according to the cooperative pressure parameters (e.g., 500N) output by the force cooperation control model, while the lower suction robot controls its suction according to the cooperative suction parameters (e.g., 60kPa). In the experiment, during the initial pressure application phase, the pressure applied by the upper robot was measured to be 498N, which stabilized at 500N±2N after closed-loop adjustment; the initial negative pressure of the lower suction robot was 59kPa, which stabilized at 60kPa±1kPa after adjustment. The results show that after the coordinated application of pressure and suction, the pre-assembled sunroof component can be firmly and evenly fixed, with the assembly error controlled within ±0.25mm.

[0087] By coordinating forces through the upper and lower robots, the sunroof components are ensured to be subjected to balanced forces, avoiding local overpressure or insufficient suction, thereby reducing the risk of material damage and breakage. The upper and lower robots can quickly adjust the pressure and suction parameters to ensure that they are always kept within the predetermined target range during the assembly process, resulting in a high degree of automation in the assembly process, a significant reduction in assembly time, and a reduction in rework caused by assembly errors, thus significantly improving overall production efficiency.

[0088] In summary, the sunroof production assembly control method based on robot collaborative networks provided in this application has the following beneficial effects:

[0089] By acquiring custom-made sunroof components, pre-assembling them according to their assembly positions, and obtaining pre-assembled sunroof components, a distributed collaborative robot system is activated. This system includes an upper pressure robot and a lower suction robot, which are synchronously controlled through a distributed control architecture. A force collaboration control model is constructed, and collaborative pressure and suction parameters are obtained based on this model. The upper pressure robot applies pressure control to the pre-assembled sunroof component according to the collaborative pressure parameters, and the lower suction robot applies suction control to the pre-assembled sunroof component according to the collaborative suction parameters, resulting in the assembled sunroof component. In other words, by setting up dual-pressure assembly robots—one pressure robot above the sunroof and one suction robot below—the two robots flexibly cooperate to perform force collaboration assembly after the sunroof is aligned, achieving precise coordination and improving the product quality and production efficiency of sunroof assembly.

[0090] Example 2: Based on the same inventive concept as the sunroof production and assembly control method based on a robot collaborative network in Example 1, this application also provides a sunroof production and assembly control system based on a robot collaborative network. Please refer to the appendix. Figure 2 The sunroof production and assembly control system based on a robot collaborative network includes:

[0091] The sunroof component acquisition module 11 is used to acquire customized sunroof components, align and pre-assemble them according to their assembly positions, and acquire pre-assembled sunroof components. The robot start-up module 12 is used to start distributed collaborative robots, including an upper pressure robot and a lower suction robot, which are synchronously controlled through a distributed control architecture. The control model construction module 13 is used to construct a force collaboration control model and acquire collaborative pressure parameters and collaborative suction parameters based on the force collaboration control model. The sunroof component assembly module 14 is used for the upper pressure robot to perform pressure control on the pre-assembled sunroof component according to the collaborative pressure parameters, and the lower suction robot to perform suction control on the pre-assembled sunroof component according to the collaborative suction parameters, to acquire the assembled sunroof component.

[0092] Furthermore, the robot start module 12 in the sunroof production and assembly control system based on a robot collaborative network is also used for:

[0093] The force collaboration control model is connected to the assembly path planning module. It acquires assembly position images of the pre-assembled sunroof component through a vision extraction unit to obtain assembly images. Pixel recognition is performed on the assembly images to obtain a first pixel dataset and a second pixel dataset. First unit coverage pixel data corresponding to the upper pressure robot and second unit coverage pixel data corresponding to the lower suction robot are acquired respectively. The assembly path planning module performs assembly path planning on the first and second pixel datasets based on the first and second unit coverage pixel data to obtain an assembly planning path. The upper pressure robot and the lower suction robot are controlled according to the assembly planning path.

[0094] Furthermore, the robot start module 12 in the sunroof production and assembly control system based on a robot collaborative network is also used for:

[0095] The assembly path planning module uses the first unit coverage pixel data as the pixel step size to separate the first pixel dataset to obtain a first group of path nodes; the assembly path planning module uses the second unit coverage pixel data as the pixel step size to separate the second pixel dataset to obtain a second group of path nodes; the first group of path nodes is connected by trajectory to obtain a first planned path, the second group of path nodes is connected by trajectory to obtain a second planned path, and the first planned path and the second planned path are combined to generate an assembly planning path, wherein the first planned path and the second planned path are not synchronized.

[0096] Furthermore, the robot start module 12 in the sunroof production and assembly control system based on a robot collaborative network is also used for:

[0097] Based on the first unit coverage pixel data and the second unit coverage pixel data, the mean pixel data is determined; the mean pixel data is used as the pixel step size to separate the first pixel dataset and the second pixel dataset to obtain a first group of path nodes and a second group of path nodes; the path nodes in the first group of path nodes or the second group of path nodes are connected according to a preset trajectory to obtain a synchronous planning path, and an assembly planning path is generated based on the synchronous planning path.

[0098] Furthermore, the sunroof production assembly control system based on a robot collaborative network also includes an upper pressure feedback module, which is further used for:

[0099] The upper pressure robot includes a six-degree-of-freedom (DOF) pressurizing robotic arm, a pressurizing platform, a pressure control module, and a force sensor. The collaborative pressure parameters are sent to the six-DOF pressurizing robotic arm, and the pressure control module controls the six-DOF pressurizing robotic arm to drive the pressurizing platform to control the pressure on the pre-assembled sunroof component. The force sensor monitors the pressure and uploads the obtained pressure sensing data to the pressure control module, which then provides force feedback to the six-DOF pressurizing robotic arm based on the pressure control module's input.

[0100] Furthermore, the sunroof production assembly control system based on a robot collaborative network also includes a lower suction feedback module, which is further used for:

[0101] The suction robot below includes a negative pressure generator, a vacuum adsorption platform, and a suction control module; the cooperative suction parameters are sent to the suction control module, and the suction control module controls the suction of the suction cups on the vacuum adsorption platform according to the negative pressure generator.

[0102] Furthermore, the sunroof production and assembly control system based on a robot collaborative network also includes a vacuum adsorption module, which is further used for:

[0103] The suction cups on the vacuum adsorption platform are arranged in a honeycomb array.

[0104] Furthermore, the control model construction module 13 in the sunroof production and assembly control system based on a robot collaborative network is also used for:

[0105] Obtain the force sample dataset of the distributed collaborative robot, wherein the force sample dataset includes the pressure sample dataset of the upper pressure robot and the suction sample dataset of the lower suction robot, the difference dataset corresponding to the pressure sample dataset and the suction sample dataset, and the label information characterizing the assembly error; define the parameters of the impedance control model, including the initial stiffness, initial damping, and initial mass parameters of the upper pressure robot and the lower suction robot; train the defined impedance control model according to the force sample dataset, and output the force collaborative control model when the model converges.

[0106] Furthermore, the control model construction module 13 in the sunroof production and assembly control system based on a robot collaborative network is also used for:

[0107] The material and geometric information of the sunroof component is collected; force protection analysis is performed based on the material and geometric information, and a force protection threshold is output. The force protection threshold is used as a constraint condition for training the impedance control model to obtain a force cooperative control model.

[0108] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1 The sunroof production assembly control method and specific examples based on robot collaborative networks in Embodiment 1 are also applicable to the sunroof production assembly control system based on robot collaborative networks in this embodiment. Through the foregoing detailed description of the sunroof production assembly control method based on robot collaborative networks, those skilled in the art can clearly understand the sunroof production assembly control system based on robot collaborative networks in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here. As for the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant details can be found in the method section.

[0109] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0110] Obviously, those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for controlling the production and assembly of sunroofs based on a robot collaborative network, characterized in that, include: Obtain the customized sunroof component, align and pre-assemble it according to the assembly position of the sunroof component, and obtain the pre-assembled sunroof component; The distributed collaborative robot is activated, which includes an upper pressure robot and a lower suction robot, and the upper pressure robot and the lower suction robot are synchronously controlled through a distributed control architecture; Construct a force cooperation control model, and obtain cooperation pressure parameters and cooperation suction parameters based on the force cooperation control model; The upper pressure robot controls the pressure of the pre-assembled sunroof component according to the cooperative pressure parameters, and the lower suction robot controls the suction of the pre-assembled sunroof component according to the cooperative suction parameters, so as to obtain the assembled sunroof component. The activation of the distributed collaborative robot, comprising an upper pressure robot and a lower suction robot, wherein the upper pressure robot and the lower suction robot are synchronously controlled through a distributed control architecture, including: The force collaboration control model is connected to the assembly path planning module, and the assembly position image of the pre-assembled sunroof component is acquired through the visual extraction unit to obtain the assembly image. Pixel recognition is performed on the assembly image to obtain a first pixel dataset and a second pixel dataset. The first pixel dataset consists of the upper surface edge pixels of the sunroof assembly, and the second pixel dataset consists of the lower surface edge pixels of the sunroof assembly. The first unit coverage pixel data corresponding to the upper pressure robot and the second unit coverage pixel data corresponding to the lower suction robot are obtained respectively. The assembly path planning module performs assembly path planning on the first pixel dataset and the second pixel dataset based on the first unit coverage pixel data and the second unit coverage pixel data to obtain the assembly planning path. Control the upper pressure robot and the lower suction robot according to the assembly planning path.

2. The sunroof production assembly control method based on robot collaborative network as described in claim 1, characterized in that, The assembly path planning module performs assembly path planning on the first pixel dataset and the second pixel dataset based on the first unit coverage pixel data and the second unit coverage pixel data, including: The assembly path planning module uses the first unit coverage pixel data as the pixel step size to separate the first pixel dataset to obtain the first group of path nodes. The assembly path planning module uses the second unit coverage pixel data as the pixel step size to separate the second pixel dataset to obtain a second group of path nodes; A first planned path is obtained by connecting the first group of path nodes, and a second planned path is obtained by connecting the second group of path nodes. The first planned path and the second planned path are then used to generate an assembly planned path, wherein the first planned path and the second planned path are not synchronized.

3. The sunroof production assembly control method based on robot collaborative network as described in claim 1, characterized in that, The assembly path planning module performs assembly path planning on the first pixel dataset and the second pixel dataset based on the first unit coverage pixel data and the second unit coverage pixel data, and further includes: The mean pixel data is determined based on the first unit coverage pixel data and the second unit coverage pixel data; The mean pixel data is used as the pixel step size to separate the first pixel dataset and the second pixel dataset, resulting in a first group of path nodes and a second group of path nodes. Connect the path nodes in the first group or the second group according to the preset trajectory to obtain the synchronous planning path, and generate the assembly planning path based on the synchronous planning path.

4. The sunroof production assembly control method based on robot collaborative network as described in claim 1, characterized in that, The above-ground pressure robot includes a six-degree-of-freedom pressurizing robotic arm, a pressurizing platform, a pressure control module, and a force sensor; The collaborative pressure parameters are sent to the six-degree-of-freedom pressurizing robotic arm, and the pressure control module controls the six-degree-of-freedom pressurizing robotic arm to drive the pressurizing platform to control the pressure of the pre-assembled sunroof component; The force sensor monitors the pressure and uploads the obtained pressure sensing data to the pressure control module. The pressure control module then provides force feedback to the six-degree-of-freedom pressurized robotic arm.

5. The sunroof production assembly control method based on robot collaborative network as described in claim 1, characterized in that, The suction robot below includes a negative pressure generator, a vacuum adsorption platform, and a suction control module; The cooperative suction parameters are sent to the suction control module, which controls the suction of the suction cups on the vacuum adsorption platform according to the negative pressure generator.

6. The sunroof production assembly control method based on robot collaborative network as described in claim 5, characterized in that, The suction cups on the vacuum adsorption platform are arranged in a honeycomb array.

7. The sunroof production assembly control method based on robot collaborative network as described in claim 1, characterized in that, The sunroof production assembly control method based on robot collaborative network also includes: Obtain the force sample dataset of the distributed collaborative robot, wherein the force sample dataset includes the pressure sample dataset of the upper pressure robot and the suction sample dataset of the lower suction robot, the difference dataset corresponding to the pressure sample dataset and the suction sample dataset, and the label information characterizing the assembly error; Define the parameters of the impedance control model, including the initial stiffness, initial damping, and initial mass parameters of the upper pressure robot and the lower suction robot; The defined impedance control model is trained based on the force sample dataset, and the force cooperative control model is output when the model converges.

8. The sunroof production assembly control method based on robot collaborative network as described in claim 7, characterized in that, The acquisition of force-cooperative control models also includes: Collect the material and geometric information of the sunroof component; Force protection analysis is performed based on the material and geometric information, and a force protection threshold is output. The force protection threshold is used as a constraint condition for training the impedance control model to obtain a force cooperative control model.

9. A sunroof production and assembly control system based on a robot collaborative network, characterized in that, The steps for implementing the sunroof production assembly control method based on a robot collaborative network according to any one of claims 1 to 8, wherein the sunroof production assembly control system based on a robot collaborative network includes: The sunroof component acquisition module is used to acquire customized sunroof components, align and pre-assemble them according to their assembly positions, and acquire pre-assembled sunroof components. A robot startup module is used to start a distributed collaborative robot, which includes an upper pressure robot and a lower suction robot. The upper pressure robot and the lower suction robot are synchronously controlled through a distributed control architecture. The control model construction module is used to construct a force cooperation control model and obtain cooperation pressure parameters and cooperation suction parameters based on the force cooperation control model. The sunroof assembly module is used for the upper pressure robot to control the pressure of the pre-assembled sunroof component according to the cooperative pressure parameters, and the lower suction robot to control the suction of the pre-assembled sunroof component according to the cooperative suction parameters, so as to obtain the assembled sunroof component.

Images