Visual sorting sheet material guiding high-speed assembly system and method
By using a vision-based sorting system to guide high-speed assembly of sheet materials, the entire assembly process of workpieces is automated, solving the problems of low efficiency and poor precision in traditional manual assembly, and improving assembly quality and production efficiency.
Patent Information
- Application Number
- CN202511899918.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-16
- Publication Date
- 2026-02-24
AI Technical Summary
Traditional dual-workpiece assembly processes are labor-intensive and have a slow operating cycle, making it difficult to match the high-speed output efficiency of the front-end processing line. Furthermore, assembly deviations can easily occur due to personnel fatigue or differences in operational proficiency, affecting the accuracy and service life of the finished product.
A high-speed assembly system guided by vision sorting of sheet materials is adopted, including a conveying mechanism, a sheet feeding mechanism, a multi-axis robot, and a vision inspection mechanism. The entire process of workpiece transfer, sheet material supply, assembly operation, and quality inspection is automated through a central control system. The force sensor of the multi-axis robot and the vision inspection mechanism are used for adaptive adjustment and quality assessment.
It has achieved highly efficient automation of workpiece assembly, improved assembly accuracy and quality, reduced human error, and increased production efficiency and finished product yield.
Smart Images

Figure CN121552030A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of non-standard automation technology, and in particular to a high-speed assembly system and method for visually sorting sheet materials. Background Technology
[0002] In the field of mass production of various precision parts, the assembly process of two workpieces with dissimilar mating structures is the core link that determines the performance and reliability of the finished product. It is widely used in high-end manufacturing scenarios such as automotive precision parts, intelligent instrument mechanisms, and micro hydraulic components.
[0003] Traditional dual-workpiece assembly processes require manual completion of a series of discrete steps, including workpiece posture correction, alignment of concave and convex structures, fitting assembly, and manual re-inspection. This process is not only labor-intensive and slow-paced, making it difficult to match the high-speed output efficiency of the front-end processing line and creating a significant capacity bottleneck, but it is also prone to assembly deviations due to operator fatigue or differences in skill level. For example, problems such as inaccurate alignment of concave and convex structures, insufficient fitting depth, and scratches on the workpiece surface can seriously affect the assembly accuracy and service life of the finished product. Summary of the Invention
[0004] Based on this, the purpose of the present invention is to provide a high-speed assembly system and method for visual sorting of sheet materials.
[0005] The present invention adopts the following technical solution: A high-speed assembly system for guiding sheet materials using vision sorting is used to assemble sheet materials and workpieces. The workpiece has an assembly slot that matches the shape of the sheet material. The high-speed assembly system for guiding sheet materials using vision sorting includes a conveying mechanism, a sheet feeding mechanism, a multi-axis robot, and a vision inspection mechanism. The conveying mechanism includes a workpiece conveyor line for transporting workpieces, and the workpiece conveyor line has an assembly station. The sheet feeding mechanism is located to the side of the assembly station and is used to store and provide sheet materials to be assembled. The multi-axis robot is located to the side of the assembly station and is used to pick up the sheet materials and assemble them into the assembly slot of the workpiece. The end of the multi-axis robot is equipped with a force sensor. The vision inspection mechanism includes a first vision module and a second vision module, both located downstream of the assembly station. The first vision module is used to detect whether the sheet material is assembled onto the workpiece, and the second vision module is used to detect the assembly quality.
[0006] Preferably, the conveying mechanism further includes a support frame and a plurality of clamping modules fixed on the support frame and located on both sides of the workpiece conveying line; the clamping modules are used to clamp the workpiece.
[0007] Preferably, the clamping module includes a clamping cylinder fixed on the bracket and a clamping block connected to the telescopic end of the clamping cylinder. The clamping cylinder drives the clamping block to press down to clamp the workpiece. The bottom of the clamping block is provided with a cushioning pad.
[0008] Preferably, the sheet feeding mechanism includes a vibratory feeder, a discharge track, and a linear distribution table; one end of the discharge track is connected to the discharge port of the vibratory feeder; the linear distribution table is located at the other end of the discharge track, and the linear distribution table is used to receive sheet materials and arrange them at equal intervals.
[0009] Preferably, the linear sorting platform includes a linear drive module and a sorting plate connected by a drive; the linear drive module drives the sorting plate to reciprocate along the conveying direction of the workpiece conveyor line; the sorting plate is provided with a plurality of material receiving grooves arranged at equal intervals along its moving direction; the linear drive module drives the sorting plate to perform intermittent motion, so that each of the material receiving grooves is sequentially connected to the discharge track.
[0010] Preferably, the multi-axis manipulator includes a base, a first horizontal rotary module, a second horizontal rotary module, a third horizontal rotary module, a lifting module, and a picking module connected in sequence; the first horizontal rotary module, the second horizontal rotary module, and the third horizontal rotary module are used to cooperate in driving the picking module to move horizontally; the lifting module is used to drive the picking module to move vertically.
[0011] Preferably, the picking module includes a mounting base plate connected to the output end of the lifting module and a plurality of vacuum nozzles fixed to the mounting base plate, wherein the arrangement spacing of the vacuum nozzles matches the arrangement spacing of the material receiving groove.
[0012] A high-speed assembly control method, executed by a central control system for coordinating the control of a high-speed assembly system for guiding sheet materials as described above using vision sorting, is characterized by comprising the following steps: Task parameter input and material supply steps: Receive the input assembly task parameters, which include at least the workpiece model, sheet material specifications, and workpiece assembly quantity; Based on the assembly task parameters, control the sheet material supply mechanism to start to supply the sheet material to be assembled; Intelligent gripping and assembly steps: control the workpiece conveyor line to transport the workpiece to the assembly station; control the multi-axis robot to pick up the sheet material supplied by the sheet feeding mechanism, transfer it and assemble it into the assembly slot of the workpiece; Visual inspection and quality judgment steps: After the sheet material and the workpiece are assembled to form a finished product, the workpiece conveyor line is controlled to drive the finished product through the first vision module and the second vision module in sequence; First, the first vision module acquires images and determines whether the sheet material has been assembled on the workpiece based on the first preset features; For the assembly determined to be assembled, the second vision module acquires images and evaluates the assembly quality parameters of the finished product based on the second preset features, and outputs the quality judgment result.
[0013] Preferably, in the intelligent gripping and assembly step, when the multi-axis robot performs the assembly action, the central control system is configured as follows: After the sheet material comes into contact with the workpiece, based on the real-time feedback from the force sensor, the multi-axis robot is controlled to adaptively adjust the contact force to complete the assembly within a constant preset contact force range. If the real-time contact force exceeds the preset contact force range, the multi-axis robot pauses and fine-tunes its end position. Once the contact force returns to the preset contact force range, the assembly operation continues. Meanwhile, the central control system records the contact force-displacement curve in real time during each successful assembly process, and, in conjunction with subsequent assembly quality parameters, adaptively adjusts the preset contact force parameters and assembly trajectory of the multi-axis robot through an iterative optimization algorithm.
[0014] Preferably, in the visual inspection and quality judgment step, the evaluation of assembly quality parameters based on the second preset features is specifically performed through a pre-trained convolutional neural network model, including the following sub-steps: Image preprocessing sub-step: Perform grayscale conversion, noise filtering, and contrast enhancement on the original image acquired by the second vision module; Target segmentation and feature extraction sub-steps: The convolutional neural network model is a dedicated model for the fit between thin sheet materials and workpiece assembly slots. Its training samples include assembly state images of workpieces and thin sheet materials of different specifications, as well as various defect images. The convolutional neural network model is used to perform pixel-level segmentation on the preprocessed images to accurately identify the boundary between the thin sheet material area and the workpiece assembly slot area, and extract key geometric features such as the geometric center point, edge feature points, and assembly slot reference feature points of the thin sheet material. Quantitative evaluation sub-step: Based on the segmentation and extraction results, calculate the concentricity deviation, planar fit, and edge gap uniformity between the sheet material and the assembly groove as core assembly quality parameters; compare the assembly quality parameters with preset values, output qualified or unqualified quality judgment results, and synchronize the assembly quality parameters and quality judgment results to the process database of the central control system.
[0015] The beneficial effects of this invention are as follows: The high-speed assembly system for guiding sheet materials using vision sorting involved in this invention includes a conveying mechanism, a sheet feeding mechanism, a multi-axis robot, and a vision inspection mechanism, realizing full-process automation of workpiece transfer, sheet material supply, assembly operations, and quality inspection. Specifically, the sheet feeding mechanism, through the cooperation of a vibratory feeder and a linear sorting table, transforms a single sheet material into a multi-station array arranged at equal intervals, creating conditions for parallel grasping. The multi-axis robot's picking module is equipped with a multi-nozzle structure, with the nozzle spacing corresponding to the material receiving groove spacing of the linear sorting table, achieving a highly efficient operation mode of simultaneously grasping multiple materials in a single stroke. The conveying mechanism also includes a clamping module, which flexibly fixes the workpiece during assembly and conveying, avoiding assembly impact and improving assembly accuracy. The vision inspection mechanism is located downstream of the assembly station, using a first vision module and a second vision module to quickly determine whether assembly has occurred and to finely evaluate assembly quality, forming a streamlined inspection layout that does not interfere with the assembly process.
[0016] The high-speed assembly control method involved in this invention achieves intelligent collaboration of multiple processes through integrated scheduling of a central control system. In the assembly stage, force sensor feedback and compliant control algorithms enable a multi-axis robot to adaptively adjust its motion trajectory and force based on real-time contact force, achieving constant force assembly and improving assembly quality. The visual inspection stage employs a specially trained convolutional neural network model to analyze images acquired by the second vision module and calculate quality parameters such as assembly concentricity and fit. Furthermore, a data feedback mechanism is established to correlate assembly quality parameters with quality judgment results, and iterative algorithms are used to optimize assembly parameters, enabling the system to continuously improve. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the structure of the visual sorting and high-speed assembly system for sheet materials of the present invention; Figure 2 for Figure 1 A partial structural diagram of circle A in the middle; Figure 3 This is a schematic diagram of the structure of the visual sorting sheet material guiding high-speed assembly system of the present invention after hiding some components; Figure 4 for Figure 3 A schematic diagram of the partial structure of circle B in the middle; Figure 5 This is a schematic diagram of the structure of the sheet material involved in this invention; Figure 6 This is a schematic diagram of the structure of the workpiece involved in this invention; Figure 7 This is a flowchart illustrating the high-speed assembly control method of the present invention.
[0018] Numbering on the map: 10-Conveying mechanism; 11-Workpiece conveying line; 12-Support; 13-Clamping module; 131-Clamping cylinder; 132-Clamping block; 133-Buffer pad; 20-Sheet feeding mechanism; 21-Vibrating plate; 22-Discharge track; 23-Linear distribution table; 231-Linear drive module; 232-Distribution plate; 233-Material receiving groove; 30-Multi-axis robot; 31-Base; 32-First horizontal rotary module; 33-Second horizontal rotary module; 34-Third horizontal rotary module; 35-Lifting module; 36-Pickup module; 361-Mounting base plate; 362-Vacuum nozzle; 40 - Visual inspection mechanism; 41 - First visual module; 42 - Second visual module; 50 - Thin sheet material; 60 - Workpiece; 61 - Assembly slot. Detailed Implementation
[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0020] In the description of this invention, it should be noted that the terms "vertical direction," "up," "down," and "horizontal," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0021] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or a connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0022] like Figures 1 to 4As shown, the high-speed assembly system for guiding sheet materials using vision sorting according to the present invention includes a conveying mechanism 10, a sheet feeding mechanism 20, a multi-axis robot 30, and a vision inspection mechanism 40. The above-mentioned high-speed assembly system for guiding sheet materials using vision sorting is used for assembling materials such as… Figure 5 and Figure 6 The sheet material 50 and workpiece 60 are shown, wherein the workpiece 60 is provided with an assembly groove 61 that matches the shape of the sheet material 50. The workpiece 60 is conveyed to the assembly station via the conveying mechanism 10; the sheet material 50 is supplied by the sheet feeding mechanism 20; the multi-axis robot 30 picks up the sheet material 50 from the sheet feeding mechanism 20 and assembles it into the assembly groove 61 of the workpiece 60; after assembly, the finished product continues to move with the conveying mechanism 10 and passes through the vision inspection mechanism 40 to complete the vision inspection.
[0023] Please see Figure 1 The conveying mechanism 10 includes a workpiece conveyor line 11 for transporting workpiece 60, and the workpiece conveyor line 11 has an assembly station. In this embodiment, the workpiece conveyor line 11 uses a motor and a conveyor belt, which can effectively coordinate with the assembly speed to transport the workpiece 60 to the assembly station. The conveying mechanism 10 also includes a support 12 and several clamping modules 13 fixed on the support 12 and located on both sides of the workpiece conveyor line 11. The clamping module 13 includes a clamping cylinder 131 fixed on the support 12 and a clamping block 132 driven by the extension end of the clamping cylinder 131. The clamping cylinder 131 drives the clamping block 132 to press down to clamp the workpiece 60, so that the workpiece 60 is stably clamped during assembly, avoiding the workpiece 60's positional displacement from affecting subsequent assembly accuracy. The bottom of the clamping block 132 is provided with a buffer pad 133, which absorbs the impact during the assembly process and prevents damage to the workpiece 60.
[0024] Please see Figure 1 , Figure 3 and Figure 4The sheet feeding mechanism 20 is located to the side of the assembly station and is used to store and provide sheet materials 50 to be assembled. The sheet feeding mechanism 20 includes a vibratory feeder 21, a discharge track 22, and a linear sorting table 23, which work together to achieve automated supply of sheet materials 50. One end of the discharge track 22 is connected to the discharge port of the vibratory feeder 21, conveying the sorted individual sheet materials 50 to the linear sorting table 23. The linear sorting table 23 is located at the other end of the discharge track 22 and includes a linear drive module 231 and a sorting plate 232. The sorting plate 232 has multiple receiving slots 233 arranged at equal intervals along its moving direction. The linear drive module 231 drives the sorting plate 232 to move intermittently back and forth along the conveying direction of the workpiece conveyor line 11, so that each receiving slot 233 sequentially docks with the end of the discharge track 22 to receive individual sheet materials 50. Through this intermittent motion, the sheet material 50 from the vibratory feeder 21 is fed one by one into different receiving grooves 233, forming a multi-station array with equal spacing. This design transforms single-piece sequential feeding into parallel array feeding, greatly improving feeding efficiency and laying the foundation for subsequent synchronous gripping and assembly.
[0025] Please see Figure 1 and Figure 3 The multi-axis robot 30 is located to the side of the assembly station and is used to pick up the sheet material 50 and assemble it into the assembly slot 61 of the workpiece 60. The multi-axis robot 30 includes a base 31, a first horizontal rotary module 32, a second horizontal rotary module 33, a third horizontal rotary module 34, a lifting module 35, and a picking module 36 connected in sequence. Each module works together to achieve multi-degree-of-freedom spatial movement. In particular, the end of the multi-axis robot 30 is equipped with a force sensor for real-time monitoring of contact forces during the assembly process. The picking module 36 includes a mounting base plate 361 connected to the output end of the lifting module 35 and multiple vacuum nozzles 362 fixed on the mounting base plate 361. The spacing between the vacuum nozzles 362 is precisely matched with the spacing between the material receiving slots 233. The matching design of the vacuum nozzle 362 and the material receiving groove 233 enables the picking module 36 to pick up multiple sheet materials 50 arranged in the material receiving groove 233 simultaneously in one descent action, thereby improving the assembly speed.
[0026] Please see Figure 1The visual inspection mechanism 40 includes a first visual module 41 and a second visual module 42, both located downstream of the assembly station. The first visual module 41 and the second visual module 42 are both positioned downstream of the assembly station to avoid interference with the movement range of the multi-axis robot 30. The first visual module 41 is located upstream and is used to quickly determine whether the sheet material 50 has been assembled onto the workpiece 60, achieving an initial screening for assembly presence. The second visual module 42 is located downstream and is used to perform more detailed assembly quality inspection on the assemblies that have passed the initial screening. This two-stage progressive inspection scheme separates rapid initial inspection from detailed re-inspection, ensuring both inspection efficiency and inspection quality.
[0027] The high-speed assembly system for visual sorting of sheet materials involved in this invention includes a conveying mechanism 10, a sheet feeding mechanism 20, a multi-axis robot 30, and a visual inspection mechanism 40, realizing the full-process automation of workpiece 60 transmission, sheet material 50 supply, assembly operation, and quality inspection. The sheet feeding mechanism 20, through the cooperation of the vibratory feeder 21 and the linear sorting table 23, transforms a single sheet material 50 into a multi-station array with equal spacing, creating conditions for parallel grasping; the picking module 36 of the multi-axis robot 30 is equipped with a multi-nozzle structure, and the spacing of the nozzles corresponds to the spacing of the material receiving groove 233 of the linear sorting table 23, realizing an efficient operation mode of simultaneously grasping multiple materials in a single stroke; the conveying mechanism 10 is also equipped with a clamping module 13, which flexibly fixes the workpiece 60 during the assembly and conveying process, avoiding assembly impact and improving assembly accuracy; the vision inspection mechanism 40 is set downstream of the assembly station, and uses the first vision module 41 and the second vision module 42 to perform rapid judgment of whether assembly has occurred and fine evaluation of assembly quality, forming a streamlined inspection layout that does not interfere with the assembly process.
[0028] like Figure 7 As shown, the present invention also relates to a high-speed assembly method, executed by a central control system for coordinating the control of a vision-based sheet material guiding high-speed assembly system as described above, comprising the following steps: S1. Task Parameter Input and Material Feeding Steps: The central control system receives the input assembly task parameters, which include at least the workpiece 60 model, the sheet material 50 specifications, and the assembly quantity per workpiece 60. Based on these parameters, the system automatically calls the pre-stored process template to complete the parameter configuration of each actuator. Subsequently, the control system starts the sheet material feeding mechanism 20: the vibratory feeder 21 sorts the sheet material 50 and feeds it into the discharge track 22; at the same time, the linear drive module 231 drives the material distribution plate 232 to perform intermittent motion, so that the sheet material 50 enters each receiving groove 233 in sequence, forming an array of equally spaced materials.
[0029] S2. Intelligent Gripping and Assembly Steps: The central control system first controls the workpiece conveyor line 11 to transfer the workpiece 60 to the assembly station and drives the clamping module 13 to clamp and fix the workpiece 60, creating a stable environment for assembly. Subsequently, it controls the multi-axis robot arm 30 to move into the sheet feeding mechanism 20, driving the picking module 36 to simultaneously pick up multiple arranged sheet materials 50. After picking up, the robot arm moves the materials above the assembly station to perform the assembly action.
[0030] In this step, the system utilizes a force sensor at the end effector of the multi-axis robot 30 to achieve compliant assembly control. When the sheet material 50 contacts the workpiece 60, the central control system, based on real-time feedback from the force sensor, adaptively adjusts the contact force of the multi-axis robot 30 to complete the assembly within a constant preset contact force range. If the real-time contact force exceeds the preset range, the system controls the robot to pause and automatically fine-tune its end effector position, resuming assembly once the contact force returns to normal. These control actions effectively compensate for mechanical positioning errors, workpiece 60 tolerances, and micro-deformations, effectively avoiding hard collisions and improving assembly success rate and product yield.
[0031] Meanwhile, the central control system records the contact force-displacement curve of each assembly process in real time, and combines it with the visual quality judgment results obtained in subsequent steps. Through iterative optimization algorithms, it continuously and adaptively adjusts the preset contact force parameters and assembly trajectory of the multi-axis robot 30. By learning from historical data and optimizing, it gradually adapts to current production conditions, continuously optimizes the assembly process, and further improves assembly quality.
[0032] S3. Visual Inspection and Quality Judgment Steps: After assembly, the central control system drives the workpiece conveyor line 11 to sequentially pass through two downstream visual inspection stations. First, images are acquired by the first vision module 41. The central control system quickly determines whether assembly has occurred based on first preset features (such as whether a specific structure of the sheet material 50 appears in the field of view), achieving efficient initial screening. For workpieces 60 determined to be assembled, images are then acquired by the second vision module 42. Based on the second preset features, the assembly quality parameters of the finished product are evaluated, and the quality judgment result is output.
[0033] The visual inspection and quality assessment steps are executed through a pre-trained dedicated convolutional neural network model, specifically including: an image preprocessing sub-step, which performs grayscale conversion, noise reduction, and contrast enhancement on the image to improve feature visibility; a target segmentation and feature extraction sub-step, which uses the CNN model to perform pixel-level segmentation of the image, accurately identifying the boundary between the sheet material 50 and the assembly groove 61, and extracting key geometric features such as geometric center and edge feature points; and a quantitative evaluation sub-step, which calculates assembly quality parameters such as concentricity deviation, planar fit, and edge gap uniformity between the sheet material 50 and the assembly groove 61 based on the extracted features, compares them with preset standards, and outputs a qualified or unqualified quality assessment result. Finally, the assembly quality parameters and quality assessment results are synchronized to the process database of the central control system.
[0034] The high-speed assembly control method involved in this invention achieves intelligent collaboration of multiple processes through integrated scheduling of a central control system. In the assembly stage, force sensor feedback and compliant control algorithms enable the multi-axis robot 30 to adaptively adjust its motion trajectory and force based on real-time contact force, achieving constant force assembly and improving assembly quality. In the visual inspection stage, a specially trained convolutional neural network model is used to analyze images acquired by the second vision module 42, calculating quality parameters such as assembly concentricity and fit. Furthermore, a data feedback mechanism is established to correlate assembly quality parameters with quality judgment results, and iterative algorithms are used to optimize assembly parameters, enabling the system to continuously improve.
[0035] The above description merely illustrates preferred technical solutions of the present invention, and while the description is relatively specific and detailed, it should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and the present invention also intends to include these modifications and variations.
Claims
1. A high-speed assembly system for guiding sheet materials through vision sorting, used for assembling sheet materials and workpieces, wherein the workpiece is provided with an assembly groove that matches the shape of the sheet material; characterized in that, The vision-based high-speed assembly system for guiding sheet materials includes a conveying mechanism, a sheet feeding mechanism, a multi-axis robot, and a vision inspection mechanism. The conveying mechanism includes a workpiece conveyor line for transporting workpieces, with an assembly station on the conveyor line. The sheet feeding mechanism is located to the side of the assembly station and is used to store and provide sheet materials to be assembled. The multi-axis robot is located to the side of the assembly station and is used to pick up the sheet materials and assemble them into the assembly slots of the workpieces. A force sensor is located at the end of the multi-axis robot. The vision inspection mechanism includes a first vision module and a second vision module, both located downstream of the assembly station. The first vision module detects whether the sheet materials are assembled onto the workpiece, and the second vision module detects the assembly quality.
2. The high-speed assembly system for guiding sheet materials in vision sorting according to claim 1, characterized in that, The conveying mechanism further includes a support frame and several clamping modules fixed on the support frame and located on both sides of the workpiece conveying line; the clamping modules are used to clamp the workpiece.
3. The high-speed assembly system for guiding sheet materials in vision sorting according to claim 2, characterized in that, The clamping module includes a clamping cylinder fixed on the bracket and a clamping block connected to the telescopic end of the clamping cylinder. The clamping cylinder drives the clamping block to press down to clamp the workpiece. The bottom of the clamping block is provided with a cushioning pad.
4. The high-speed assembly system for guiding sheet materials in vision sorting according to claim 1, characterized in that, The sheet feeding mechanism includes a vibratory feeder, a discharge track, and a linear distribution platform; one end of the discharge track is connected to the discharge port of the vibratory feeder; the linear distribution platform is located at the other end of the discharge track and is used to receive sheet materials and arrange them at equal intervals.
5. The high-speed assembly system for guiding sheet materials in vision sorting according to claim 4, characterized in that, The linear material distribution table includes a linear drive module and a material distribution plate connected by a drive; the linear drive module drives the material distribution plate to reciprocate along the conveying direction of the workpiece conveyor line; the material distribution plate is provided with a plurality of material receiving grooves arranged at equal intervals along its moving direction; the linear drive module drives the material distribution plate to perform intermittent motion, so that each of the material receiving grooves is sequentially connected to the discharge track.
6. The high-speed assembly system for guiding sheet materials in vision sorting according to claim 5, characterized in that, The multi-axis manipulator includes a base, a first horizontal rotary module, a second horizontal rotary module, a third horizontal rotary module, a lifting module, and a picking module connected in sequence; the first horizontal rotary module, the second horizontal rotary module, and the third horizontal rotary module are used to work together to drive the picking module to move horizontally; the lifting module is used to drive the picking module to move vertically up and down.
7. The high-speed assembly system for guiding sheet materials in vision sorting according to claim 6, characterized in that, The picking module includes a mounting base plate connected to the output end of the lifting module and multiple vacuum nozzles fixed to the mounting base plate. The arrangement spacing of the vacuum nozzles matches the arrangement spacing of the material receiving groove.
8. A high-speed assembly control method, executed by a central control system, for coordinating control of a vision-based sheet material guiding high-speed assembly system as described in any one of claims 1 to 7, characterized in that, Includes the following steps: Task parameter input and material supply steps: Receive the input assembly task parameters, which include at least the workpiece model, sheet material specifications, and workpiece assembly quantity; Based on the assembly task parameters, control the sheet material supply mechanism to start to supply the sheet material to be assembled; Intelligent gripping and assembly steps: control the workpiece conveyor line to transport the workpiece to the assembly station; control the multi-axis robot to pick up the sheet material supplied by the sheet feeding mechanism, transfer it and assemble it into the assembly slot of the workpiece; Visual inspection and quality judgment steps: After the sheet material and the workpiece are assembled to form a finished product, the workpiece conveyor line is controlled to drive the finished product through the first vision module and the second vision module in sequence; First, the image is acquired through the first vision module, and the sheet material is judged to be assembled on the workpiece based on the first preset feature. For assemblies that are determined to be assembled, images are acquired through the second vision module, and the assembly quality parameters of the finished product are evaluated based on the second preset features, and the quality judgment result is output.
9. The high-speed assembly control method according to claim 8, characterized in that, In the intelligent gripping and assembly step, when the multi-axis robot performs the assembly action, the central control system is configured as follows: After the sheet material comes into contact with the workpiece, based on the real-time feedback from the force sensor, the multi-axis robot is controlled to adaptively adjust the contact force to complete the assembly within a constant preset contact force range. If the real-time contact force exceeds the preset contact force range, the multi-axis robot pauses and fine-tunes its end position. Once the contact force returns to the preset contact force range, the assembly operation continues. Meanwhile, the central control system records the contact force-displacement curve in real time during each successful assembly process, and, in conjunction with subsequent assembly quality parameters, adaptively adjusts the preset contact force parameters and assembly trajectory of the multi-axis robot through an iterative optimization algorithm.
10. The high-speed assembly control method according to claim 8, characterized in that, In the visual inspection and quality assessment step, the evaluation of assembly quality parameters based on the second preset features is specifically performed through a pre-trained convolutional neural network model, including the following sub-steps: Image preprocessing sub-step: Perform grayscale conversion, noise filtering, and contrast enhancement on the original image acquired by the second vision module; Target segmentation and feature extraction sub-steps: The convolutional neural network model is a dedicated model for the fit between thin sheet materials and workpiece assembly slots. Its training samples include assembly state images of workpieces and thin sheet materials of different specifications, as well as various defect images. The convolutional neural network model is used to perform pixel-level segmentation on the preprocessed images to accurately identify the boundary between the thin sheet material area and the workpiece assembly slot area, and extract key geometric features such as the geometric center point, edge feature points, and assembly slot reference feature points of the thin sheet material. Quantitative evaluation sub-step: Based on the segmentation and extraction results, calculate the concentricity deviation, planar fit, and edge gap uniformity between the sheet material and the assembly groove, as the core assembly quality parameters; The assembly quality parameters are compared with preset values, and a qualified or unqualified quality judgment result is output. The assembly quality parameters and the quality judgment result are then synchronized to the process database of the central control system.