Double-station robot control method based on clean panel sorting and terminal equipment

Abstract

The invention discloses a double-station robot control method and terminal equipment based on clean panel sorting, and the method comprises the steps: obtaining three-dimensional point clouds and two-dimensional images of different clean panels and protection plate sleeves, and carrying out the analysis to obtain material features; by acquiring the three-dimensional point cloud and the two-dimensional image, the spatial form and surface information of the material can be comprehensively and accurately captured, and a rich and accurate data basis is provided for subsequent feature analysis. In practical application, different panel types and panel sleeve materials are obviously different in three-dimensional structure and two-dimensional vision, and feature analysis results become important basis for identification and sorting. According to the double-station robot control method based on clean panel sorting and the terminal equipment, the space-time coordination model is adopted to optimize the double-station coordination path according to the robot working space and obtain the optimal coordination path, the robot working space can be fully utilized, the double-station action conflict is reduced, and the working efficiency is improved. And the double-station collaborative operation efficiency and fluency of the robot are improved.

Description

Technical Field

[0001] The present invention relates to the field of clean panel sorting, and in particular to a double-station robot control method and terminal equipment based on clean panel sorting. Background Art

[0002] In the field of industrial automation production, traditional processing methods for sorting precision and clean materials such as LCD panels, new energy photovoltaic panels, and integrated circuit PCB boards generally suffer from insufficient automation. Existing sorting systems often rely on manual visual inspection or basic mechanical devices for rough classification. The means of collecting material characteristics are limited, usually only able to obtain partial information such as size or weight, and it is difficult to fully reflect the three-dimensional morphological characteristics and surface texture details of the material. This lack of information directly leads to fluctuations in classification accuracy, especially when processing panels and protective cover plates made of mixed materials. Misjudgments or missed inspections are prone to occur, forcing the production line to frequently stop for review, seriously affecting overall operational efficiency.

[0003] Existing sorting equipment has obvious shortcomings in terms of collaborative operation and control accuracy. Dual-station robot systems generally use independent path planning methods and lack an optimization mechanism for the dynamic allocation of workspaces, which leads to overlapping motion trajectories of the robotic arms and frequent motion conflicts and path retractions. Due to the lack of trajectory constraints during the sorting process, the robot's end effector is prone to accidental collisions with materials or tooling, causing scratches on the clean panel surface or deformation of the protective plate cover. The traditional centralized control architecture concentrates data calculations on cloud servers, resulting in real-time response delays and decentralized operation terminal functions. Operators need to switch between multiple interfaces to complete parameter adjustments and status monitoring, further reducing the smoothness of system operation. The equipment protection design is insufficient, the control module is exposed to the production environment for a long time, and dust intrusion can easily cause electrical failures, becoming a potential hidden danger that restricts system stability. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application provides a dual-station robot control method and terminal device based on clean panel sorting. This method addresses the aforementioned centralized control architecture, which concentrates data operations on cloud servers, resulting in delayed real-time responses and fragmented terminal functions. Operators must switch between multiple interfaces to complete parameter adjustments and status monitoring, further reducing the smoothness of system operation. Inadequate equipment protection design, long-term exposure of the control module to the production environment, and dust intrusion can easily cause electrical failures, becoming a potential technical problem that restricts system stability.

[0005] To achieve the above objectives, the present application provides the following technical solutions: a dual-station robot control method based on clean panel sorting, comprising: Acquire 3D point clouds and 2D images of different clean panels and protective cover sheets, and analyze them to derive material characteristics. This allows for comprehensive and accurate capture of the material's spatial form and surface information, providing a rich and accurate data foundation for subsequent feature analysis. In practical applications, different panel types and cover sheet materials exhibit significant differences in 3D structure and 2D visual appearance, and these feature analysis results become crucial for identification and sorting. An intelligent classification model is used to identify and output panel types and panel sleeve materials based on the material characteristics, where the panel types include liquid crystal panels, new energy photovoltaic panels, and integrated circuit PCB boards. The intelligent classification model uses the powerful feature learning capabilities of deep learning to automatically extract the most discriminative information from material characteristics. Compared with traditional manual recognition or simple algorithms, its recognition accuracy and efficiency are greatly improved, which can effectively avoid sorting errors caused by manual misjudgment and significantly improve sorting reliability in large-scale production scenarios. Based on a preset sorting action sequence combined with separation trajectory constraints, the robot is collaboratively controlled to sort clean panels and protective cover plates. This combination of preset sorting action sequences and separation trajectory constraints enables the robot to operate in the most rational and safest manner during the sorting process. The preset action sequence represents the optimal operating process developed based on extensive experimentation and practical experience, while the separation trajectory constraints, established through the robot's kinematic and dynamic models, precisely plan the robot's motion path, ensuring collision-free collisions with panels and cover plates even during high-speed sorting. This effectively protects materials from damage and improves sorting efficiency. The robot's dual-station collaborative path is trained based on a preset material stacking method and a preset interlayer pressure threshold. The preset material stacking method and interlayer pressure threshold provide clear goals and standards for dual-station collaborative path training. Different stacking methods are suitable for different types of panels, while the interlayer pressure threshold is determined based on the panel material and thickness. This ensures that the stacked materials remain stable during transportation and storage, preventing damage to the panels due to uneven pressure. Furthermore, dual-station collaborative path training enables efficient coordination between the two workstations, improving overall work efficiency. A spatiotemporal collaboration model is used to optimize the dual-station collaborative path based on the robot's workspace, resulting in the optimal collaborative path. This model fully considers the geometric constraints and time factors of the robot's workspace. By optimizing the dual-station collaborative path, it can rationally arrange the work sequence and time nodes of the two stations, avoiding work conflicts and excessive waiting times. In complex working environments, this model enables the robot to operate efficiently within limited space, further improving the overall efficiency and quality of clean panel sorting.

[0006] Preferably, the step of obtaining three-dimensional point clouds and two-dimensional images of different clean panels and protective plate covers and analyzing them to obtain material characteristics includes: The three-dimensional point cloud and two-dimensional image are obtained through the visual positioning module; the visual positioning module can quickly and stably obtain the three-dimensional point cloud and two-dimensional image of the material, providing raw data for subsequent processing. Its high-precision positioning function can ensure that the acquired image information accurately reflects the actual state of the material. The three-dimensional point cloud and two-dimensional image are preprocessed, including denoising, enhancement and normalization; denoising can remove interference information introduced during the image acquisition process to make the image clearer; enhancement operation can highlight key features in the image to facilitate subsequent extraction; normalization processing unifies the format and range of image data, improves data consistency and comparability, and lays the foundation for accurate extraction of material features. Extract the three-dimensional shape and size features of the 3D point cloud, and the color and texture features of the 2D image. By extracting these features, we can comprehensively describe the appearance and structure of the material. The three-dimensional shape and size features can be used to determine the type and specification of the material, while the color and texture features can help distinguish between different materials, thus providing detailed and effective data support for intelligent classification.

[0007] Preferably, the intelligent classification model is used to identify and output the panel type and the panel sleeve material according to the material characteristics, including: The material features are input into a convolutional neural network model based on deep learning. The convolutional neural network model has powerful image and data feature extraction capabilities. Inputting the material features into the model can give full play to its advantages and automatically learn and explore the potential laws in the material features. The convolutional neural network model extracts and classifies the material features, outputting the panel type and panel material. Through multiple layers of convolution and pooling, the model gradually extracts higher-level, more discriminative features, then uses a classifier to accurately classify these features. Compared to traditional methods, this model has higher accuracy and stability in classifying complex material features, enabling rapid and precise identification of panel type and panel material.

[0008] Preferably, the method of collaboratively controlling the robot to sort the clean panels and protective plate covers according to a preset sorting action sequence combined with separation trajectory constraints includes: The preset sorting action sequence is pre-established according to different panel types and plate sleeve materials; the preset sorting action sequence is specially designed according to the characteristics of different panel types and plate sleeve materials. Each type of material has its most suitable sorting method. Following the preset action sequence can ensure the standardization and efficiency of the sorting process. The separation trajectory constraint is established using the robot's kinematic and dynamic models to prevent collisions between the robot and the clean panel and protective cover during the sorting process. These models accurately describe the robot's motion patterns and mechanical properties. The separation trajectory constraint, based on these models, monitors and adjusts the robot's path in real time during movement, ensuring a safe distance between the robot and the material. This effectively prevents collisions even during high-speed sorting or in complex environments.

[0009] Preferably, the training of the robot's dual-station collaborative path according to a preset material stacking method and a preset inter-layer pressure threshold includes: The preset material stacking methods include layered stacking and vertical stacking. The interlayer pressure threshold is set based on the material and thickness of the clean panels to ensure the stability of the clean panel stacking. Layered stacking and vertical stacking are suitable for different types and specifications of clean panels. Choosing the right stacking method can improve space utilization. The interlayer pressure threshold ensures uniform force on the panels during stacking, preventing deformation and damage due to excessive or insufficient pressure. By combining the preset stacking method and pressure threshold to train a dual-station collaborative path, the robot can precisely control force and position during the stacking process, achieving efficient and stable stacking operations.

[0010] A terminal device for controlling a double-station robot based on clean panel sorting adopts the above-mentioned double-station robot control method based on clean panel sorting, and includes a control module, an electric rotating base, an electric rotating module, a first telescopic module and a sorting module. The second telescopic module is arranged at the upper end of the electric rotating base, and the two sides of the first telescopic module are respectively rotatably connected to the sorting module and the second telescopic module. An electric rotating module is arranged between the second telescopic module and the electric rotating base, between the first telescopic module and the sorting module, and between the first telescopic module and the sorting module.

[0011] The outside of the control module is rotatably connected to a closed door, and the outside of the closed door is respectively provided with a perception control terminal, a robot control module and an I / O scheduling module, and the inner cavity of the control module is respectively provided with an edge computing terminal and a human-computer interaction terminal. The control module, serving as the core hub, rapidly processes collected data through an edge computing terminal. Combining this with a pre-set sorting algorithm, it sends instructions to the robot control module and I / O scheduling module, coordinating the operation of various components. The perception control terminal senses the external environment in real time and feeds this data back to the control module, creating a closed-loop control system. The motorized rotating base rotates under the control of the control module, driving the entire device to adjust its working position. The motorized rotating module enables flexible rotation between the connected components, coordinating the telescopic movement of the first and second telescopic modules to enable the sorting module to precisely reach its designated position.

[0012] The human-machine interaction terminal facilitates operator configuration of equipment parameters, monitoring of equipment operating status, and troubleshooting. Effectively, this structural design enables precise control and flexible scheduling of equipment. Through efficient data processing on the edge computing terminal, command response time is significantly shortened, enabling rapid adaptation to varying sorting task requirements and significantly improving clean panel sorting efficiency. The closed door effectively protects the precision components within the control module, preventing the intrusion of dust and other impurities, ensuring the stability and reliability of equipment operation. Preferably, a vacuum suction cup module and a monitoring module are respectively connected to both sides of the bottom of the sorting module, and a protective shell is provided on the outside of the monitoring module. When the sorting module reaches the clean panel, the vacuum cup module activates under the control module's command, creating negative pressure by pumping air to firmly grasp the clean panel and achieve stable grasping. The monitoring module monitors the suction status, the position and posture of the clean panel in real time, and feeds this data back to the control module so that it can adjust the sorting action in a timely manner. The protective shell protects the monitoring module from damage caused by collisions or contamination during operation.

[0013] The cooperation between the vacuum suction cup module and the monitoring module ensures stable grasping and accurate sorting of clean panels, reduces the risk of panel damage caused by unstable grasping, and improves the sorting success rate and product quality; the protective shell extends the service life of the monitoring module and reduces the maintenance cost of the equipment.

[0014] Preferably, a load-bearing seat is provided on the top of the electric rotating base, and one end of the load-bearing seat is connected to the electric rotating module. After receiving the rotation command from the control module, the electric rotating base drives the load-bearing base to rotate. The load-bearing base provides stable support for the electric rotating module and other components connected to it, and transmits the rotation action to the electric rotating module, thereby driving the entire sorting mechanism to adjust its working direction.

[0015] The setting of the load-bearing seat enhances the stability of the connection between the electric rotating base and the upper components, and can withstand the large load generated during the sorting process, ensuring that the equipment maintains stable operation during the rotation process, avoiding sorting errors caused by shaking, and improving the reliability and sorting accuracy of the equipment operation.

[0016] Preferably, the external rotation of the electric rotation module is connected to a cylinder, and the output end of the cylinder is connected to the telescopic end of the outside of the second telescopic module. The control module drives the second telescopic module to perform telescopic movements by controlling the extension and retraction of the cylinder. At the same time, the electric rotation module provides flexible rotation coordination for the extension and retraction of the second telescopic module, enabling the second telescopic module to achieve precise extension and retraction at different angles and directions.

[0017] The coordinated work of the cylinder and the electric rotation module greatly enhances the movement flexibility and controllability of the second telescopic module, making it able to adapt to complex working environments and diverse sorting tasks, improving the equipment's working range and operating accuracy, and ensuring that the sorting module can accurately reach clean panels at different heights and positions for sorting operations.

[0018] Preferably, a reinforcement base is provided at the bottom of the electric rotating base, and fixing plates are evenly distributed on the outside of the reinforcement base. The reinforced base is connected to the ground or other fixed plane through a fixing plate, which firmly fixes the equipment in the workplace and prevents the electric rotating base from being displaced or shaken during the rotation process.

[0019] The setting of the reinforced base and fixed plate significantly improves the overall stability of the equipment, enabling it to withstand the large forces and torques generated during the sorting process. Even in the case of high-speed rotation or frequent movements, the equipment can remain stable, ensuring the continuity and accuracy of the sorting work, while also reducing the risk of damage to the equipment due to unstable operation.

[0020] In summary, the present application provides a dual-station robot control method and terminal device based on clean panel sorting, which has the following beneficial effects: This dual-station robot control method and terminal device based on clean panel sorting can comprehensively and accurately extract multi-dimensional information such as the shape, size, and texture of the materials by acquiring three-dimensional point clouds and two-dimensional images of different clean panels and protective plate covers and analyzing the material characteristics, providing a reliable data basis for subsequent classification and sorting. An intelligent classification model is used to identify and output panel types and plate cover materials based on material characteristics. This method can efficiently and accurately automatically classify different types of panels and plate cover materials such as liquid crystal panels, new energy photovoltaic panels, and integrated circuit PCB boards, thereby improving classification efficiency and accuracy and reducing manual intervention. The robot is collaboratively controlled to sort clean panels and protective plate covers based on a preset sorting action sequence combined with separation trajectory constraints, making the robot's sorting action more accurate and orderly, avoiding collision or damage to materials during the sorting process, and ensuring the safety and reliability of the sorting operation. A spatiotemporal collaborative model is used to optimize the dual-station collaborative path according to the robot's workspace and obtain the optimal collaborative path. This can fully utilize the robot's workspace, reduce dual-station action conflicts, shorten path planning time, and improve the efficiency and smoothness of the robot's dual-station collaborative operation.

[0021] In this double-station robot control method and terminal device based on clean panel sorting, the external rotation of the control module is connected to the closed door, which can protect the internal components of the control module and prevent dust, debris, etc. from entering and affecting the operation of the equipment; a perception control terminal, a robot control module and an I / O scheduling module are respectively arranged outside the closed door, which facilitates the operator to perform perception control, robot motion adjustment and input and output scheduling on the equipment, thereby improving the convenience of equipment operation; an edge computing terminal is arranged in the inner cavity of the control module, which can realize local and rapid processing of data, reduce data transmission delay, and improve the data processing efficiency and response speed of the equipment; a human-computer interaction terminal is arranged to facilitate information exchange between the operator and the equipment, and realize operations such as monitoring the working status of the equipment and setting parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a front schematic diagram of the present invention; Figure 2 is an external schematic diagram of the control module of the present invention; Figure 3 is a schematic diagram of the exterior of the electric rotating base of the present invention; Figure 4 It is a plan view of the electric rotating base of the present invention; Figure 5 Schematic diagram of the external appearance of the sorting module of the present invention.

[0023] Description of reference numerals: 1. Control module; 11. Closed door; 12. Perception control terminal; 13. Robot control module; 14. I / O scheduling module; 15. Edge computing terminal; 16. Human-computer interaction terminal; 2. Electric rotating base; 21. Load-bearing seat; 3. Electric rotating module; 4. First telescopic module; 5. Sorting module; 51. Vacuum suction cup module; 52. Monitoring module; 6. Second telescopic module; 61. Cylinder. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0025] This application provides a technical solution, a dual-station robot control method based on clean panel sorting, including: Acquire 3D point clouds and 2D images of different clean panels and protective cover sheets, and analyze them to derive material characteristics. This allows for comprehensive and accurate capture of the material's spatial form and surface information, providing a rich and accurate data foundation for subsequent feature analysis. In practical applications, different panel types and cover sheet materials exhibit significant differences in 3D structure and 2D visual appearance, and these feature analysis results become crucial for identification and sorting. An intelligent classification model is used to identify and output panel types and panel sleeve materials based on material characteristics. Panel types include LCD panels, new energy photovoltaic panels, and integrated circuit PCB boards. The intelligent classification model uses the powerful feature learning capabilities of deep learning to automatically extract the most discriminative information from material characteristics. Compared with traditional manual recognition or simple algorithms, its recognition accuracy and efficiency are greatly improved, which can effectively avoid sorting errors caused by human misjudgment and significantly improve sorting reliability in large-scale production scenarios. Based on a preset sorting action sequence combined with separation trajectory constraints, the robot is collaboratively controlled to sort clean panels and protective cover plates. This combination of preset sorting action sequences and separation trajectory constraints enables the robot to operate in the most rational and safest manner during the sorting process. The preset action sequence represents the optimal operating process developed based on extensive experimentation and practical experience, while the separation trajectory constraints, established through the robot's kinematic and dynamic models, precisely plan the robot's motion path, ensuring collision-free collisions with panels and cover plates even during high-speed sorting. This effectively protects materials from damage and improves sorting efficiency. The robot's dual-station collaborative path is trained based on a preset material stacking method and a preset inter-layer pressure threshold. The preset material stacking method and inter-layer pressure threshold provide clear goals and standards for dual-station collaborative path training. Different stacking methods are suitable for different types of panels, while the inter-layer pressure threshold is determined based on the panel material and thickness. This ensures that the stacked materials remain stable during transportation and storage, preventing damage to the panels due to uneven pressure. Furthermore, dual-station collaborative path training enables efficient coordination between the two workstations, improving overall work efficiency. A spatiotemporal collaboration model is used to optimize the dual-station collaborative path based on the robot's workspace, resulting in the optimal collaborative path. This model fully considers the geometric constraints and time factors of the robot's workspace. By optimizing the dual-station collaborative path, it can rationally arrange the work sequence and time nodes of the two stations, avoiding work conflicts and excessive waiting times. In complex working environments, this model enables the robot to operate efficiently within limited space, further improving the overall efficiency and quality of clean panel sorting.

[0026] Obtain 3D point clouds and 2D images of different clean panels and protective panels, and analyze them to obtain material characteristics, including: The visual positioning module is used to obtain 3D point clouds and 2D images. The module can quickly and stably obtain 3D point clouds and 2D images of materials, providing raw data for subsequent processing. Its high-precision positioning function ensures that the acquired image information accurately reflects the actual status of the material. Preprocessing of 3D point clouds and 2D images includes denoising, enhancement and normalization. Denoising can remove interference information introduced during image acquisition to make the image clearer. Enhancement can highlight key features in the image to facilitate subsequent extraction. Normalization unifies the format and range of image data, improves data consistency and comparability, and lays the foundation for accurate extraction of material features.

[0027] Extract the three-dimensional shape and size features of the 3D point cloud, and the color and texture features of the 2D image. By extracting these features, we can comprehensively describe the appearance and structure of the material. The three-dimensional shape and size features can be used to determine the type and specification of the material, while the color and texture features can help distinguish between different materials, thus providing detailed and effective data support for intelligent classification.

[0028] An intelligent classification model is used to identify the output panel type and panel material based on material characteristics, including: The material features are input into a convolutional neural network model based on deep learning. The convolutional neural network model has powerful image and data feature extraction capabilities. Inputting material features into this model can give full play to its advantages and automatically learn and explore the potential patterns in material features. A convolutional neural network model extracts and classifies material features, outputting panel type and panel material. Through multiple layers of convolution and pooling, the model gradually extracts higher-level, more discriminative features, then uses a classifier to accurately classify these features. Compared to traditional methods, this model has higher accuracy and stability in classifying complex material features, enabling rapid and precise identification of panel type and panel material.

[0029] According to the preset sorting action sequence combined with the separation trajectory constraints, the robot is collaboratively controlled to sort clean panels and protective plate covers, including: The preset sorting action sequence is pre-established according to different panel types and plate cover materials; the preset sorting action sequence is specially designed according to the characteristics of different panel types and plate cover materials. Each type of material has its most suitable sorting method. Following the preset action sequence can ensure the standardization and efficiency of the sorting process. Separation trajectory constraints are established using the robot's kinematic and dynamic models to prevent collisions between the robot and the clean panel and protective cover during the sorting process. These models accurately describe the robot's motion and mechanical characteristics. These separation trajectory constraints monitor and adjust the robot's path in real time during motion, ensuring a safe distance between the robot and the material. This effectively prevents collisions even during high-speed sorting or in complex environments.

[0030] Based on the preset material stacking method and the preset inter-layer pressure threshold, the robot is trained on a dual-station collaborative path, including: Preset material stacking methods include layered stacking and vertical stacking. The interlayer pressure threshold is set based on the material and thickness of the clean panel to ensure the stability of the clean panel stacking. Layered stacking and vertical stacking are suitable for different types and specifications of clean panels. Choosing the right stacking method can improve space utilization. The interlayer pressure threshold setting ensures that the panels are evenly stressed during stacking, preventing deformation and damage due to excessive or insufficient pressure. By combining the preset stacking method and pressure threshold to train a dual-station collaborative path, the robot can precisely control the force and position during the stacking process, achieving efficient and stable stacking operations.

[0031] By acquiring three-dimensional point clouds and two-dimensional images of different clean panels and protective cover sheets and analyzing the material characteristics, the shape, size, texture and other multi-dimensional information of the materials can be comprehensively and accurately extracted, providing a reliable data basis for subsequent classification and sorting; the intelligent classification model is used to identify the output panel type and cover material according to the material characteristics, which can efficiently and accurately classify different types of panels and cover materials such as LCD panels, new energy photovoltaic panels and integrated circuit PCB boards, improve classification efficiency and accuracy, and reduce manual intervention; the robot is collaboratively controlled to sort clean panels and protective cover sheets according to the preset sorting action sequence combined with the separation trajectory constraint, which can make the robot sorting action more accurate and orderly, avoid collision or damage to the material during the sorting process, and ensure the safety and reliability of the sorting operation. The spatiotemporal collaborative model is used to optimize the dual-station collaborative path according to the robot workspace and obtain the optimal collaborative path, which can make full use of the robot workspace, reduce dual-station action conflicts, shorten the path planning time, and improve the efficiency and smoothness of the robot's dual-station collaborative operation.

[0032] See also Figure 1 and Figure 2, a terminal device for controlling a double-station robot based on clean panel sorting adopts the above-mentioned double-station robot control method based on clean panel sorting, including a control module 1, an electric rotating base 2, an electric rotating module 3, a first telescopic module 4 and a sorting module 5, the second telescopic module 6 is arranged at the upper end of the electric rotating base 2, and the two sides of the first telescopic module 4 are respectively rotatably connected with the sorting module 5 and the second telescopic module 6, and the electric rotating modules 3 are arranged between the second telescopic module 6 and the electric rotating base 2, between the first telescopic module 4 and the sorting module 5, and between the first telescopic module 4 and the sorting module 5. The control module 1 controls the electric rotating base 2, the electric rotating module 3, the first telescopic module 4, the sorting module 5 and the second telescopic module 6 through the circuit.

[0033] The outside of the control module 1 is rotatably connected to a closed door 11, and the outside of the closed door 11 is respectively provided with a perception control terminal 12, a robot control module 13 and an I / O scheduling module 14, and the inner cavity of the control module 1 is respectively provided with an edge computing terminal 15 and a human-computer interaction terminal 16. Control module 1, serving as the core hub, rapidly processes collected data via edge computing terminal 15. Combining this with a pre-set sorting algorithm, it sends instructions to robot control module 13 and I / O scheduling module 14 to coordinate the operation of various components. Perception control terminal 12 senses the external environment in real time and feeds this data back to control module 1, forming a closed-loop control system. The motorized rotating base 2 rotates under the control of control module 1, driving the entire device to adjust its working position. The motorized rotating module 3 enables flexible rotation between the various connected components, coordinating the telescopic movements of the first and second telescopic modules 4 and 6, allowing the sorting module 5 to precisely reach its designated position.

[0034] The human-machine interaction terminal 16 facilitates operator configuration of equipment parameters, monitoring of equipment operating status, and troubleshooting. Effectively, this structural design enables precise control and flexible scheduling of the equipment. Through efficient data processing by the edge computing terminal 15, command response time is significantly shortened, enabling rapid adaptation to varying sorting task requirements and significantly improving clean panel sorting efficiency. The closed door 11 effectively protects the delicate components within the control module 1, preventing the intrusion of dust and other impurities, ensuring the stability and reliability of the equipment's operation. See also Figure 5The bottom sides of the sorting module 5 are connected to a vacuum suction cup module 51 and a monitoring module 52, respectively. A protective housing surrounds the monitoring module 52. When the sorting module 5 reaches the clean panel, the vacuum suction cup module 51 activates under the control of the control module 1, creating negative pressure by pumping air to firmly hold the clean panel and achieve stable grip. The monitoring module 52 monitors the suction status, position, and posture of the clean panel in real time, and feeds this data back to the control module 1, allowing it to adjust the sorting operation in a timely manner. The protective housing protects the monitoring module 52, preventing damage from collisions or contamination during operation.

[0035] The cooperation between the vacuum suction cup module 51 and the monitoring module 52 ensures stable grasping and accurate sorting of clean panels, reduces the risk of panel damage due to unstable grasping, and improves the sorting success rate and product quality; the protective shell extends the service life of the monitoring module 52 and reduces the maintenance cost of the equipment.

[0036] See also Figure 3 and Figure 4 A load-bearing seat 21 is provided on the top of the electric rotating base 2 , and one end of the load-bearing seat 21 is connected to the electric rotating module 3 . After receiving the rotation command from the control module 1, the electric rotating base 2 drives the load-bearing base 21 to rotate. The load-bearing base 21 provides stable support for the electric rotating module 3 and other components connected to it, and transmits the rotation action to the electric rotating module 3, thereby driving the entire sorting mechanism to adjust the working direction.

[0037] The setting of the load-bearing seat 21 enhances the stability of the connection between the electric rotating base 2 and the upper components, can withstand the large load generated during the sorting process, ensure that the equipment maintains stable operation during the rotation process, avoids sorting errors caused by shaking, and improves the reliability and sorting accuracy of the equipment operation.

[0038] The outer rotation of the electric rotating module 3 is connected to the cylinder 61 , and the output end of the cylinder 61 is connected to the telescopic end of the outer portion of the second telescopic module 6 . The control module 1 drives the second telescopic module 6 to perform telescopic movement by controlling the extension and retraction of the cylinder 61. At the same time, the electric rotation module 3 provides flexible rotation coordination for the extension and retraction of the second telescopic module 6, so that the second telescopic module 6 can achieve precise extension and retraction at different angles and directions.

[0039] The coordinated work of the cylinder 61 and the electric rotation module 3 greatly enhances the movement flexibility and controllability of the second telescopic module 6, can adapt to complex working environments and diverse sorting tasks, improves the working range and operating accuracy of the equipment, and ensures that the sorting module 5 can accurately reach clean panels at different heights and positions for sorting operations.

[0040] A reinforcement base is provided at the bottom of the electric rotating base 2, and fixing plates are evenly distributed on the outside of the reinforcement base. The reinforced base is connected to the ground or other fixed plane via a fixing plate, so as to firmly fix the device in the workplace and prevent the electric rotating base 2 from being displaced or shaken during the rotation process.

[0041] The setting of the reinforced base and fixed plate significantly improves the overall stability of the equipment, enabling it to withstand the large forces and torques generated during the sorting process. Even in the case of high-speed rotation or frequent movements, the equipment can remain stable, ensuring the continuity and accuracy of the sorting work, while also reducing the risk of damage to the equipment due to unstable operation.

[0042] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0043] Although the embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A dual-station robot control method based on clean panel sorting, characterized in that: include: Obtain 3D point clouds and 2D images of different clean panels and protective panels, and analyze them to obtain material characteristics; An intelligent classification model is used to identify and output panel types and panel cover materials based on the material characteristics, wherein the panel types include liquid crystal panels, new energy photovoltaic panels, and integrated circuit PCB boards; According to the preset sorting action sequence combined with the separation trajectory constraints, the robot is collaboratively controlled to sort clean panels and protective plate covers; Training the robot's dual-station collaborative path according to a preset material stacking method and a preset inter-layer pressure threshold; The spatiotemporal collaborative model is used to optimize the dual-station collaborative path according to the robot workspace to obtain the optimal collaborative path.

2. The dual-station robot control method based on clean panel sorting according to claim 1 is characterized in that: The three-dimensional point clouds and two-dimensional images of different clean panels and protective plate covers are obtained and analyzed to obtain material characteristics, including: Acquire 3D point clouds and 2D images through the visual positioning module; Preprocessing of 3D point clouds and 2D images, including denoising, enhancement, and normalization; Extract the stereo shape and size features of 3D point clouds and the color and texture features of 2D images.

3. The dual-station robot control method based on clean panel sorting according to claim 1 is characterized in that: The intelligent classification model is used to identify and output the panel type and the panel sleeve material according to the material characteristics, including: Input material features into the convolutional neural network model based on deep learning; The material features are extracted and classified through a convolutional neural network model, and the panel type and panel material are output.

4. The dual-station robot control method based on clean panel sorting according to claim 1 is characterized in that: The method of collaboratively controlling the robot to sort the clean panels and protective plate covers according to a preset sorting action sequence combined with separation trajectory constraints includes: The preset sorting action sequence is pre-established according to different panel types and panel cover materials; The separation trajectory constraint is established through the robot kinematic model and dynamic model to prevent the robot from colliding with the clean panel and the protective plate cover during the sorting process.

5. The dual-station robot control method based on clean panel sorting according to claim 1 is characterized in that: The method of training the robot's dual-station collaborative path according to a preset material stacking method and a preset inter-layer pressure threshold includes: The preset material stacking method includes layered stacking and vertical stacking, and the interlayer pressure threshold is set according to the material and thickness of the clean panel to ensure the stability of the clean panel stacking.

6. A terminal device for controlling a double-station robot based on clean panel sorting, adopting the double-station robot control method based on clean panel sorting according to any one of claims 1 to 5, comprising a control module (1), an electric rotating base (2), an electric rotating module (3), a first telescopic module (4) and a sorting module (5), wherein the second telescopic module (6) is arranged at the upper end of the electric rotating base (2), the two sides of the first telescopic module (4) are rotatably connected to the sorting module (5) and the second telescopic module (6), respectively, and the second telescopic module (6) and the electric rotating base (2), the first telescopic module (4) and the sorting module (5), and the first telescopic module (4) and the sorting module (5) are all connected to the electric rotating module (3), characterized in that: The exterior of the control module (1) is rotatably connected to a closed door (11), and a sensing control terminal (12), a robot control module (13), and an I / O scheduling module (14) are respectively provided on the exterior of the closed door (11), and an edge computing terminal (15) and a human-computer interaction terminal (16) are respectively provided in the interior of the control module (1).

7. The terminal device controlled by a double-station robot based on clean panel sorting according to claim 6 is characterized in that: The two sides of the bottom of the sorting module (5) are respectively connected to a vacuum suction cup module (51) and a monitoring module (52), and a protective shell is provided on the outside of the monitoring module (52).

8. The terminal device controlled by a double-station robot for clean panel sorting according to claim 6, characterized in that: A bearing seat (21) is provided on the top of the electric rotating base (2), and one end of the bearing seat (21) is connected to the electric rotating module (3).

9. The terminal device controlled by a double-station robot based on clean panel sorting according to claim 6, characterized in that: The external rotation of the electric rotation module (3) is connected to a cylinder (61), and the output end of the cylinder (61) is connected to the external telescopic end of the second telescopic module (6).

10. The terminal device controlled by a double-station robot based on clean panel sorting according to claim 6, characterized in that: A reinforcement base is provided at the bottom of the electric rotating base (2), and fixing plates are evenly distributed on the outside of the reinforcement base.

Images