System and method for conveying and sorting electronic products

Abstract

The invention discloses a system and method for conveying and sorting electronic products, and the system and method are characterized in that the system comprises a multi-stage conveying device, a visual detection module, a sorting mechanical arm and a central control module, the multi-stage conveying device comprises a first-stage conveying line and a second-stage conveying line, and the first-stage conveying line is provided with a dynamic weighing sensor array; the visual detection module comprises a high-frame-rate industrial camera and an infrared contour scanner, and the visual detection module is mounted above the second-stage conveying line; the sorting mechanical arm comprises an electromagnetic-vacuum dual-mode grabber and an end effector integrated pressure feedback sensor; and the central control module is connected with all the modules through an industrial Ethernet, a built-in path optimization algorithm is used for carrying out dynamic analysis on the sorting mechanical arm, an obtained analysis result is fed back to the visual detection module and the central control module, and the central control module is connected with a product database.

Description

Technical Field

[0001] The present invention relates to the technical field of sorting, and in particular to a system and method for conveying and sorting electronic products. Background Art

[0002] Electronic products refer to various devices and apparatuses that use electronic technology for information processing, data storage, signal transmission, and energy conversion. They are widely used in multiple fields such as communication, entertainment, education, medical treatment, military, and industrial control, constituting an important infrastructure of the modern information society. During the processing of electronic products, it is necessary to sort different electronic products.

[0003] When the existing electronic product conveying and sorting device conveys and sorts electronic products, it is not convenient to automatically sort the electronic products to be sorted, so that it is necessary for personnel to continuously place the electronic products to be sorted, which increases the labor intensity of the staff while reducing the sorting efficiency of electronic products.

[0004] Therefore, it is necessary to provide a system and method for conveying and sorting electronic products to solve the above-mentioned existing technical problems. Summary of the Invention

[0005] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title, but such simplifications or omissions shall not be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed.

[0007] To solve the above technical problems, the present invention provides the following technical solution: A system for conveying and sorting electronic products, characterized by comprising: a multi-stage conveying device, a vision detection module, a sorting robotic arm, and a central control module, including a first-stage conveyor line and a second-stage conveyor line, and a dynamic weighing sensor array is provided on the first-stage conveyor line; The vision detection module includes a high-frame-rate industrial camera and an infrared contour scanner, and the vision detection module is installed above the second-stage conveyor line; The sorting robotic arm includes an electromagnetic-vacuum dual-mode gripper and an end effector integrated with a pressure feedback sensor; The central control module is connected to each module through an industrial Ethernet, dynamically analyzes the sorting robotic arm using a built-in path optimization algorithm, and feeds the obtained analysis results back to the vision detection module and the central control module. The central control module is connected to a product database.

[0008] As a preferred solution of the electronic product conveying and sorting system described in the present invention, the dynamic weighing sensor array adopts piezoelectric ceramic sensors to collect weight data in real time with a sampling period of 5 ms, and establishes a data verification mechanism with the vision detection module.

[0009] As a preferred solution of the electronic product conveying and sorting system described in the present invention, the sorting robotic arm is of a six-axis collaborative robotic arm configuration. The grasping surface of the end effector is provided with an adaptive deformation silicone layer, and the contact pressure threshold is dynamically adjusted through a machine learning model.

[0010] As a preferred solution of the electronic product conveying and sorting system described in the present invention, the path optimization algorithm is implemented based on an improved genetic algorithm. The real-time conveyor line load parameter is introduced as a constraint condition for the fitness function, and the response time of the sorting path planning is ≤ 50 ms.

[0011] As a preferred solution of the electronic product conveying and sorting system described in the present invention, it further includes an exception handling module, which includes an audible and visual alarm unit and an emergency braking unit. When the product parameters are detected to deviate from the database threshold by ±15%, a three-level emergency response mechanism is triggered.

[0012] As a preferred solution of the electronic product conveying and sorting system described in the present invention, the product database stores multi-dimensional characteristic parameters, including dimensional tolerances, weight ranges, and surface finish indexes, and realizes two-way data synchronization with the MES system through the OPC UA protocol.

[0013] As a preferred solution of the electronic product conveying and sorting system described in the present invention, the vision detection module adopts multi-spectral imaging technology and integrates a composite light source array in the visible light band (400 - 700 nm) and the short-wave infrared band (900 - 1700 nm).

[0014] As a preferred solution of the electronic product conveying and sorting method described in the present invention, it includes: S1. Obtain the multi-dimensional characteristics of the product through the dynamic weighing sensor and the vision detection module; S2. Perform feature fusion and classification based on the fuzzy logic algorithm; S3. Dynamically adjust the sorting priority according to the real-time conveyor line state; S4. Adopt a predictive maintenance model to evaluate the health of key components.

[0015] As a preferred solution of the electronic product conveying and sorting method described in the present invention, the fuzzy logic algorithm is used to fuse the data collected by the weighing sensor and the vision detection module in S1.

[0016] As a preferred solution of the method for transporting and sorting electronic products according to the present invention, in S2, a fuzzy logic rule base is obtained by using a fuzzy logic algorithm to improve the quality of cluster head selection and achieve balanced cluster energy at the same time. The fuzzy control model is combined with an optimization algorithm. In S3, based on S2, the sorting priority is adjusted by classifying electronic products.

[0017] Advantages of the present invention: Parallel processing is achieved through a multi-level conveying structure. The path optimization algorithm is used to reduce the idle stroke of the robotic arm. Multi-sensor data fusion (weight + vision + pressure feedback) enhances surface defect recognition in multi-spectral detection. At the same time, the sorting robotic arm is designed as a dual-mode gripper, and the gripping surface is set as an adaptive deformation silicone layer; The multi-dimensional feature database can be used to predictively maintain the model to evaluate the health of the equipment. According to the state of the real-time conveyor line, a three-level emergency response mechanism is divided to achieve the priority of intelligent sorting. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them: Figure 1 It is a schematic structural diagram of the system for transporting and sorting electronic products according to an embodiment provided by the present invention; Figure 2 It is a schematic diagram of the method steps of the system and method for transporting and sorting electronic products according to an embodiment provided by the present invention; Figure 3 It is a schematic diagram of a pure fuzzy logic system of the system and method for transporting and sorting electronic products according to an embodiment provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them.

[0020] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0021] In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Persons skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0022] Embodiment 1 Referring to Figure 1 , according to an embodiment of the present invention, a conveying and sorting system for electronic products, characterized in that it includes: a multi-stage conveying device 100, a vision detection module 200, a sorting robotic arm 300, and a central control module 400. The multi-stage conveying device 100 includes a first-stage conveying line 101 and a second-stage conveying line 102. A dynamic weighing sensor array 103 is provided on the first-stage conveying line 101; the weighing sensor array 103 is precisely installed inside the roller support bearing seat of the first-stage conveying line 101. Each end support point of each roller (with a diameter of 50 mm) integrates a sensor unit. Along the running direction of the conveyor belt (X-axis), the sensors are arranged in a matrix at equal intervals of 200 mm, covering the entire effective weighing area (usually in the range of 1.5 - 2 m in the middle section of the conveying line). In the width direction of the conveyor belt (Y-axis), a staggered layout is adopted to ensure that products of different sizes cover at least 3 sensor units; direct force transmission path: product gravity → conveyor belt → roller → sensor → base, forming the shortest force conduction chain and reducing energy loss in intermediate links.

[0023] The vision detection module 200 includes a high-frame-rate industrial camera and an infrared profile scanner, and the vision detection module is installed above the second-stage conveying line 102; The sorting robotic arm 300 includes an electromagnetic-vacuum dual-mode gripper and an end effector integrated with a pressure feedback sensor 302; The central control module 400 is connected to each module through an industrial Ethernet, dynamically analyzes the sorting robotic arm using a built-in path optimization algorithm, and feeds the obtained analysis results back to the vision detection module and the central control module. The central control module is connected to a product database.

[0024] Specifically, the multi-stage conveying device is a conveyor belt, which is divided into a first-stage conveying line and a second-stage conveying line. The difference between the first-stage conveying line and the second-stage conveying line is that a dynamic weighing sensor array is provided on the surface of the first-stage conveying line. The dynamic weighing sensor array 103 uses piezoelectric ceramic sensors to collect weight data in real time with a sampling period of 5 ms and establishes a data verification mechanism with the vision detection module 200. The sensor array 103 is composed of multiple piezoelectric ceramic weighing units and is evenly distributed at equal intervals along the conveying direction of the first-stage conveying line 101. Each weighing unit is embedded in the roller support structure of the conveying line and is in direct contact with the bottom surface of the conveyor belt.

[0025] Layout parameters: 3 - 5 sensor units are arranged per meter of the conveyor line; the distance between adjacent sensors ≤ 200 mm; the effective measurement range of the sensors: 0.1 - 10 kg (covering the weight of typical electronic products). Among them, anti-interference design is carried out for the sensors. The sensor unit adopts a double-layer vibration isolation design. The upper layer is an elastic silica gel buffer layer to isolate the vibration noise of the conveyor belt; the lower layer is a rigid aluminum alloy base bolted to the conveyor line frame. Each sensor is connected in parallel through the RS-485 bus and communicates with the central control module 400 using the Modbus-RTU protocol. The dynamic weighing trigger mechanism is triggered by a photoelectric trigger signal. A transmissive photoelectric sensor is set at the entrance of the conveyor line. When a product enters the weighing area, sampling is triggered, including outputting a trigger pulse signal to the sensor array and synchronously starting a timer (for calculating the moving speed of the product).

[0026] A vision detection module is installed on the multi-stage conveying device. The vision detection module is connected to the sorting robotic arm and the central control module. The sorting robotic arms are installed at intervals on both sides of the multi-stage conveying device. The sorting robotic arm 300 has a six-axis collaborative robotic arm configuration. Among them, the grasping surface of the end effector 301 is provided with an adaptive deformation silica gel layer, and the contact pressure threshold is dynamically adjusted through a machine learning model. The sorting robotic arm 300 straddles the first-stage conveyor line 101 and the second-stage conveyor line 102, and the working radius of the end effector 301 covers the two conveyor lines and the sorting buffer area. The position data of the conveyor belt is obtained in real time through the conveyor line encoder, and the robotic arm motion coordinate system is dynamically bound to the conveyor line speed.

[0027] The event-driven mechanism is shown in Table 1: Table 1

[0028] Among them, the action path of the sorting robotic arm is realized based on an improved genetic algorithm through the built-in chip. The real-time conveyor line load parameter is introduced as a constraint condition of the fitness function, and the response time of the sorting path planning ≤ 50 ms. Among them, the path planning mathematical model: Given a starting point A, an ending point B, and a series of target points T1, T2,..., Tn, where each target point must be covered. The grasping speed of the sorting robotic arm is z, and it can sort and grasp at multiple angles. This application is to design a path such that the sorting robotic arm starts from the starting point, sorts and grasps at multiple angles, passes through all target points in sequence, and finally reaches the ending point, while satisfying the following constraint conditions and objective functions. The grasping of electronic products by the sorting robotic arm is already a very mature existing technology and will not be elaborated here.

[0029] Starting and ending constraints: The sorting robotic arm must start from the starting point A and end at the ending point B.

[0030]

[0031] Coverage Constraint: All target points must be covered.

[0032]

[0033] 3) Sorting Area Constraint: The sorting and grasping path of the sorting robotic arm must always be kept within area D.

[0034]

[0035] 4) Collision Avoidance: The path of the robotic arm cannot intersect with any known obstacles intersect.

[0036] In summary, the basic model establishes a mathematical model for the path planning of the sorting robotic arm.

[0037] Through the improved genetic algorithm: Encode the chromosomes of the robotic arm, and encode the joint angles of the robotic arm, the speed of the conveyor line, and the positions of the bins as gene sequences. The fitness function expression is: Fitness = 1 / (t_total + 0.5×E_energy + 10×Collision_risk) The evolution strategy is: Generate 200 groups of solutions in each batch and converge to the optimal path within 5 generations. The control chip of the sorting robotic arm is updated, which can achieve zero-waiting sorting and grasping. Before the vision detection is completed, a preliminary path is generated based on the historical data of the product database. When the actual detection data arrives, the path correction time ≤ 15 ms.

[0038] It also includes an exception handling module 500, which includes an audible and visual alarm unit and an emergency braking unit, and triggers a three-level emergency response mechanism when it detects that the product parameters deviate from the database threshold by ±15%. The product database stores multi-dimensional characteristic parameters, including dimensional tolerances, weight ranges, and surface finish indicators, and realizes two-way data synchronization with the MES system through the OPC UA protocol.

[0039] The priority matrix of the multi-objective collaborative scheduling tasks is shown in Table 2: Table 2

[0040] The robotic arm is technically upgraded to have a collision prevention system with three-dimensional dynamic protection, including static protection, dynamic protection and emergency strategies. Static protection: Preset the safety boundary (spacing ≥ 50 mm) between the robotic arm and the conveyor line. Dynamic protection: Real-time monitor moving objects within a distance of 10 - 500 mm through ToF sensors. The sorting robotic arm deeply interacts with the multi-stage conveying device, including blockage warning, beat synchronization and process quality traceability. The blockage warning monitors the load of the conveyor line through a pressure sensor array. When the load per unit length > 5 kg / m, the upstream feeding speed is reduced. The beat synchronization increases the conveyor line speed to 1.5 times the nominal value during the sorting gap of the robotic arm (t > 200 ms). When the grasping action is initiated, the conveyor line speed is reduced to 0.8 times the nominal value. The process quality traceability records the grasping parameters (force / angle / time consumption) of each product. When defects are found in subsequent processes, the sorting parameters are optimized in reverse.

[0041] The product database stores multi-dimensional characteristic parameters, including dimensional tolerances, weight ranges, surface finish indicators, and realizes two-way data synchronization with the MES system through the OPC UA protocol. The vision detection module 200 adopts multi-spectral imaging technology and integrates a composite light source array in the visible light band (400 - 700 nm) and the short-wave infrared band (900 - 1700 nm). The vision detection module sends the detected electronic products to be sorted to the central control module through Ethernet. The central control module sends sorting instructions to the sorting robotic arm controller, and the sorting and grasping are executed through the end effector. Through the clock synchronization of the vision detection module and the sorting robotic arm controller, using the IEEE 1588 Precision Time Protocol (PTP), the clock deviation between the vision module and the robotic arm is < 1 μs, and the whole process from image acquisition to robotic arm response is ≤ 20 ms.

[0042] In the preprocessing part of the photos by the visual detection module, white balance and grayscale preprocessing are performed through the WhiteBalance and ColorToGreyscaleFilter functions to remove irrelevant information. In the part of extracting and sorting the coordinates of the high-frame-rate industrial camera, the camera coordinates of 6 electronic products are obtained by using the ExtractBlobs function to extract blobs. Then, the SortBlobs function is used to sort the 6 electronic products according to the X coordinate. In the 9-point calibration part, the CalibrateAdvanced function is used to establish the coordinate conversion relationship between the two by giving the pixel coordinates and the robotic arm coordinates of 9 points. In the part of extracting the color of the electronic product, the TrainExtractColor function is first used to train the color library, which includes four colors: black, yellow, red, and blue. Then, the ExtractColor function is used to obtain the color number of the electronic product: 1 represents black, 2 represents yellow, 3 represents red, and 4 represents blue. In the coordinate conversion stage, the conversion matrix obtained by 9-point calibration is used to convert the sorted camera coordinates into the coordinate values of the robotic arm. In the TCP communication part, a socket connection is established with the robotic arm controller by using the TCPDevice function. After receiving the photo-taking command sent by the robotic arm controller, the system will send the color information and position information to the robotic arm controller. In order to enable the robotic arm controller, the intelligent camera, and the PLC to work in coordination, system debugging is also required. The main work of system debugging is the parameter setting of the conveyor chain tracking and the communication debugging between the intelligent camera and the robotic arm controller. In order to ensure that when the conveyor chain runs 1 meter, the number of pulses output by the encoder is between 1250 and 2500, the robotic arm controller simultaneously collects the number of rising edges and falling edges of phase A and phase B. 4 signals are collected in one cycle, and the calculated signals collected by the controller are between 5000 and 10000. Less than 5000 will affect the tracking accuracy of the robotic arm, and more than 10000 will not improve the tracking accuracy of the robotic arm. The minimum speed of the conveyor chain is 4 mm / s, and the maximum speed is 2000 mm / s. The conveyor chain tracking parameter settings are shown in Table 3: Table 3

[0043] The IPs of the intelligent camera and the robotic arm controller should be in the same network segment. The intelligent camera is the server side for Socket communication, and the robotic arm is the client side. When both the intelligent camera and the industrial robotic arm controller are started, first, the intelligent camera establishes a socket communication server through the TCPDevice function. The robotic arm controller creates a socket client through the SocketCreate function and establishes a connection with the intelligent camera server through the SocketConnect function. When the robotic arm controller detects that the workpiece is connected to the conveyor chain through the WaitWObi function, it immediately sends a photographing instruction to the intelligent camera through the SocketSend function. After receiving the instruction through the Read function, the intelligent camera compares it with the string "OK". After successful comparison and confirmation of the photographing instruction, it sends the color coordinate information to the robotic arm in the form of a string through the Write function. The robotic arm controller stores the received string in String2 through the SocketReceive function, and then conducts data analysis to parse the string into robotic arm coordinate values and color information.

[0044] In summary, by using a multi-stage conveying device, through multi-modal perception fusion and cross-domain collaborative control, the visual detection module and the sorting robotic arm have achieved full-chain optimization from "seeing" to "grabbing accurately", solving the efficiency bottleneck problem caused by the separation of vision and execution in traditional sorting systems.

[0045] Embodiment 2 Refer to Figures 2 - 3 , what is different from the previous embodiment in this embodiment is that this embodiment provides a sorting method, including: S1, obtaining multi-dimensional features of products through dynamic weighing sensors and visual detection modules; S2, performing feature fusion and classification on products based on fuzzy logic algorithms; S3, dynamically adjusting the sorting priority according to the real-time state of the multi-stage conveyor line; S4, using a predictive maintenance model to evaluate the health of key components.

[0046] Specifically, in step 1, the multi-dimensional features of electronic products are analyzed through a multi-stage conveying device and a visual detection module. In step 2, based on step 1, the multi-dimensional features of the products in step 1 are fused and classified through a fuzzy logic algorithm. In step 3, the sorting priority is adjusted according to the product classification in step 2. In step 4, the relevant operating components of the above steps are monitored.

[0047] The fuzzy inference system can be divided into two types: the pure fuzzy logic system, the Sugeno type and the Mamadani type. Among them, the pure fuzzy logic system is the core part of other types of fuzzy logic systems and is a general model that utilizes this type of linguistic information under the principles of general fuzzy logic. The pure fuzzy logic system can be regarded as a mapping relationship. The input fuzzy set A obtains the output fuzzy set B through the fuzzy rule base and fuzzy inference in the fuzzy logic system, as Figure 3 shown. A node will participate in the election of the cluster head only when the remaining energy it has is at least equal to the average remaining energy of its neighbor nodes. Since the selection of the CH is based on the principle of high score first, the emergence of one CH often affects the emergence of other CHs. This will lead to the formation of a CH vacancy area, making it difficult for the nodes in this area to establish communication connections. To address this problem, this application adopts a CH coexistence mechanism. In this mechanism, after one CH is generated, other nodes still have the opportunity to become CHs. Using the density of neighbor nodes and artificial experience, competing nodes can calculate the number of CHs allowed to coexist among the current neighbors:

[0048] where: N is the number of CHs allowed to coexist, is the number of neighbor nodes, N is the total number of nodes, is the optimal ratio of CHs in the ideal case. Once the score ranking of the competing node is less than , this node will automatically become a CH.

[0049] In the process of optimizing the trajectory planning of industrial robotic arms, it is crucial to select a suitable optimization algorithm. The genetic algorithm is an optimization algorithm that simulates the principles of natural selection and genetics, and has advantages such as strong global search ability and fast convergence speed. It is suitable for solving complex, non-linear, multi-modal optimization problems.

[0050] The path optimization algorithm is an optimization algorithm for path behavior, and has advantages such as good parallelism and easy implementation. It is suitable for solving problems such as path planning and scheduling optimization.

[0051] The stochastic gradient descent method is an optimization algorithm based on the principle of gradient descent and is suitable for solving convex optimization problems. In the trajectory planning of robotic arms, it can be used to adjust fuzzy control parameters.

[0052] The simulated annealing algorithm is an optimization algorithm based on the physical annealing process and is suitable for solving local optimum problems. In the trajectory planning of a robotic arm, it can be used to jump out of local optimum solutions. Combining the fuzzy control model with the optimization algorithm can give full play to the advantages of both and improve the accuracy and stability of the robotic arm trajectory planning. The optimization algorithm is used to optimize the parameters in the fuzzy control model to improve the control effect. For example, the genetic algorithm is used to optimize the membership function parameters in the fuzzy controller. The optimization algorithm is used to optimize the fuzzy control rules to improve the control accuracy. For example, the ant colony algorithm is used to optimize the rule strength in the fuzzy control rules. The optimization algorithm is used to optimize the fuzzy control structure to improve the control performance. For example, the simulated annealing algorithm is used to optimize the structure of the fuzzy controller, such as increasing or decreasing the fuzzy control rules.

[0053] According to the requirements of the robotic arm trajectory planning, a fuzzy control model is established, including inputs, outputs, membership functions, fuzzy control rules, etc. Select a suitable optimization algorithm according to the actual situation and design the corresponding optimization algorithm parameters. The optimization algorithm is used to optimize the parameters in the fuzzy control model to improve the control effect.

[0054] The optimization algorithm is used to optimize the fuzzy control rules to improve the control accuracy. The optimization algorithm is used to optimize the fuzzy control structure to improve the control performance. Through simulation experiments or actual applications, verify the optimized trajectory planning effect to ensure the smoothness and accuracy of the robotic arm motion trajectory. According to the verification results, adjust the optimization strategy to further improve the optimization effect of the robotic arm trajectory planning.

[0055] In summary, WT2F-CRA establishes a fitness function based on FND, uses the fuzzy logic algorithm to obtain the optimal fuzzy logic rule base, improves the quality of cluster head selection and realizes balanced clustering energy. On this basis, the communication radius of the cluster head is inferred by T2FL, the communication range of the cluster head is restricted, and the cluster entry mechanism with CH coexistence is adopted to further extend the network lifetime.

[0056] Furthermore, the method can be implemented in any type of computing platform operably connected, including but not limited to personal computers, minicomputers, mainframes, workstations, network or distributed computing environments, separate or integrated computer platforms, or communicating with charged particle tools or other imaging devices.

[0057] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An electronic product conveying and sorting system, characterized in that Including: A multi-stage conveying device (100), a vision detection module (200), a sorting robotic arm (300) and a central control module (400). The multi-stage conveying device (100) includes a first-stage conveying line (101) and a second-stage conveying line (102), and a dynamic weighing sensor array (103) is provided on the first-stage conveying line (101); The vision detection module (200) includes a high-frame-rate industrial camera and an infrared contour scanner, and the vision detection module (200) is installed above the second-stage conveying line (102); The sorting robotic arm (300) includes an electromagnetic-vacuum dual-mode gripper and an end effector integrated pressure feedback sensor (302); The central control module (400) is connected to each module through an industrial Ethernet, dynamically analyzes the sorting robotic arm by using a built-in path optimization algorithm, and feeds back the obtained analysis results to the vision detection module and the central control module. The central control module is connected to a product database.

2. The electronic product conveying and sorting system according to claim 1, wherein The dynamic weighing sensor array (103) adopts piezoelectric ceramic sensors to collect weight data in real time with a sampling period of 5 ms, and establishes a data verification mechanism with the vision detection module (200).

3. The electronic product conveying and sorting system according to claim 1, wherein, The sorting robotic arm (300) is of a six-axis collaborative robotic arm configuration. An adaptive deformation silica gel layer is provided on the grasping surface of the end effector (301), and the contact pressure threshold is dynamically adjusted through a machine learning model.

4. The electronic product conveying and sorting system according to claim 1, characterized in that, The path optimization algorithm is implemented based on an improved genetic algorithm, and the real-time conveyor line load parameter is introduced as a constraint condition of the fitness function. The sorting path planning response time ≤ 50 ms.

5. The electronic product conveying and sorting system according to claim 1, wherein It further includes an exception handling module (500), which includes an audible and visual alarm unit and an emergency braking unit, and triggers a three-level emergency response mechanism when it detects that the product parameters deviate from the database threshold by ±15%.

6. The electronic product conveying and sorting system according to claim 1, wherein The product database stores multi-dimensional characteristic parameters, including dimensional tolerances, weight ranges, and surface finish indexes, and realizes two-way data synchronization with the MES system through the OPC UA protocol.

7. The electronic product conveying and sorting system according to claim 1, wherein The vision detection module (200) adopts multi-spectral imaging technology and integrates a composite light source array in the visible light band and the short-wave infrared band.

8. A sorting method, characterized in that: A conveying and sorting system for electronic products according to any one of claims 1-7, including the following steps: S1. Obtain the multi-dimensional characteristics of the product through the dynamic weighing sensor and the vision detection module; S2. Perform feature fusion and classification based on the fuzzy logic algorithm; S3. Dynamically adjust the sorting priority according to the real-time conveyor line state; S4. Adopt a predictive maintenance model to evaluate the health of key components.

9. The sorting method according to claim 8, wherein, Use the fuzzy logic algorithm to fuse the data collected by the weighing sensor and the vision detection module in S1.

10. The sorting method according to claim 8, characterized in that, In S2, the fuzzy logic algorithm is used to obtain a fuzzy logic rule base, improve the quality of cluster head selection and achieve balanced cluster energy at the same time. The fuzzy control model is combined with the optimization algorithm. In S3, the sorting priority is adjusted based on the classification of electronic products on the basis of S2.

Images