Robot-based hard alloy sorting system and method

Abstract

The invention discloses a hard alloy sorting system and method based on a robot. The hard alloy sorting system comprises a multi-joint robot body; the conveying mechanism is used for conveying hard alloy workpieces to be sorted; the eye multi-mode sensing device is used for carrying out synchronous detection and component analysis on the hard alloy workpiece; the eye multi-mode sensing device comprises a laser-induced breakdown spectroscopy unit for acquiring element composition information and a hyperspectral imaging unit for acquiring surface chemical composition distribution information; the control unit receives and fuses the element composition information and the surface chemical composition distribution information, identifies the chemical material type of the hard alloy workpiece and generates a corresponding sorting control instruction; and the end effector grabs and sorts the hard alloy workpieces of the corresponding workpiece types according to the sorting control instruction. By integrating the LIBS unit, the high chemical component sensing capacity is achieved, and automatic and high-precision sorting of hard alloy based on inherent chemical components is achieved.

Description

Technical Field

[0001] This invention relates to the field of waste resource utilization technology, and in particular to a robot-based cemented carbide sorting system and method. Background Technology

[0002] Tungsten carbide (WC) is a key material with tungsten carbide (WC) as the hard phase and cobalt (Co) or nickel (Ni) as the binder phase. It is widely used in cutting tools (drills, end mills), molds, and wear-resistant parts. During recycling, the subsequent processing of cemented carbide with different binder phases differs significantly. Mixed recycling can drastically reduce metal recovery efficiency and purity, and increase processing costs. Therefore, precise sorting of WC-Co, WC-Ni, and other alloys at the upstream of recycling is crucial.

[0003] Currently, the sorting of cemented carbide scrap faces two major technical bottlenecks: First, it relies on manual sorting, which suffers from low efficiency, high cost, poor consistency, and difficulty in scaling up. Second, existing automated sorting equipment (such as sorters based on color, size, weight, or RGB vision) can only distinguish objects based on their external physical characteristics, completely failing to access and identify the internal chemical composition that determines the material's essence, such as the types and contents of cobalt and nickel. This makes the automated sorting of alloys like WC-Co and WC-Ni, which have similar appearances but different compositions, a long-standing problem that the industry has been unable to solve.

[0004] Therefore, it is necessary to provide a cemented carbide sorting system that has the capability to perform in-situ, non-destructive chemical composition analysis on cemented carbide workpieces, and realizes automated, high-precision sorting based on chemical composition, thereby improving recycling efficiency and metal purity. Summary of the Invention

[0005] In view of this, it is necessary to provide a robot-based cemented carbide sorting system and method to solve the technical problem that existing cemented carbide sorting systems lack the ability to perform in-situ, non-destructive chemical composition analysis on cemented carbide workpieces, thus failing to achieve automated, high-precision sorting based on intrinsic chemical composition.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides a robot-based cemented carbide sorting system, comprising: Multi-jointed robot body; A conveying mechanism is located within the working area of ​​the multi-joint robot body and is used to convey cemented carbide workpieces to be sorted. An eye-based multimodal sensing device is integrated into the head of the multi-joint robot body for synchronous detection and composition analysis of the cemented carbide workpiece. The eye-based multimodal sensing device includes at least a laser-induced breakdown spectroscopy unit for acquiring elemental composition information and a hyperspectral imaging unit for acquiring surface chemical composition distribution information. The control unit is communicatively connected to the eye multimodal sensing device and the multi-joint robot body, and is used to receive and fuse the elemental composition information and surface chemical composition distribution information, identify the chemical material category of the cemented carbide workpiece, and generate corresponding sorting control instructions. An end effector, installed at the operating end of the multi-joint robot body, is used to grasp and sort cemented carbide workpieces of the corresponding workpiece category according to the sorting control command.

[0007] In one possible implementation, the laser-induced breakdown spectral unit includes a pulsed laser, a spectrometer, and a signal processing unit; The pulsed laser is used to generate high-energy pulsed laser and focus it on the test point on the surface of the cemented carbide workpiece to excite and generate plasma. The spectrometer is used to collect the characteristic spectral signals emitted by the plasma; The signal processing unit is used to analyze the characteristic spectral lines and their intensities of preset metal elements in the characteristic spectral signal to determine the elemental composition information of the cemented carbide workpiece.

[0008] In one possible implementation, the hyperspectral imaging unit includes a hyperspectral camera, the response wavelength of which includes the visible light band and the near-infrared band, for acquiring the characteristic spectral reflectance curve of the coating on the surface of the cemented carbide workpiece, forming a surface chemical image, and performing chemical composition identification on the surface chemical image to obtain surface chemical composition distribution information.

[0009] In one possible implementation, the eye multimodal sensing device further includes a three-dimensional contour measurement unit for acquiring the three-dimensional geometric information and spatial position information of the cemented carbide workpiece; the control unit is also used to fuse the three-dimensional geometric information and assist in identifying the material type based on the three-dimensional geometric information, and / or assist in generating sorting control commands based on the spatial position information.

[0010] In one possible implementation, the control unit includes a chemical material classification module and a motion control module; The chemical material classification module is used to fuse the elemental composition information and the surface chemical composition distribution information to identify the chemical material category of the cemented carbide workpiece and obtain the identification result. A motion control module is used to generate the sorting control command based on the recognition result.

[0011] In one possible implementation, the chemical material classification module includes a pre-trained deep learning model or pattern recognition algorithm.

[0012] In one possible implementation, the chemical material category includes at least WC-Co alloys and WC-Ni alloys.

[0013] In a second aspect, the present invention also provides a cemented carbide sorting method, applied to the cemented carbide sorting method described in any one of the first aspects, comprising: The cemented carbide workpieces to be sorted are continuously conveyed by a conveying mechanism; When the cemented carbide workpiece enters the inspection station, the surface spectral information, three-dimensional geometric information and surface electrical conductivity information of the cemented carbide workpiece are simultaneously acquired through the eye multimodal sensor. By fusing elemental composition information and surface chemical composition distribution information through the control unit, the chemical material category of the cemented carbide workpiece is identified, and sorting control instructions are generated. The sorting control command is executed to grasp and sort cemented carbide workpieces of the corresponding chemical material category.

[0014] In one possible implementation, the chemical material category includes at least WC-Co alloys and WC-Ni alloys; the step of identifying the chemical material category of the cemented carbide workpiece by fusing elemental composition information and surface chemical composition distribution information through a control unit includes: Based on the elemental composition information, the characteristic spectral line intensities of cobalt and nickel are extracted. Based on the characteristic spectral line intensities of cobalt and nickel, the intensity ratio characterizing the relative content of cobalt and nickel is calculated. Based on the strength ratio and the preset ratio range, the base material of the cemented carbide workpiece is preliminarily classified to obtain a preliminary classification result; Based on the surface chemical composition distribution information, the preliminary classification results are verified or corrected to determine the chemical material category.

[0015] The beneficial effects of this invention are: The present invention provides a robot-based cemented carbide sorting system, comprising a multi-jointed robot body, enabling efficient processing of cemented carbide workpieces of various shapes and improving the flexibility of sorting execution; a conveying mechanism, disposed within the working area of ​​the multi-jointed robot body, for conveying the cemented carbide workpieces to be sorted; and an eye-based multimodal sensing device, integrated into the head of the multi-jointed robot body, for simultaneous detection and compositional analysis of the cemented carbide workpieces. The eye-based multimodal sensing device includes at least a laser-induced breakdown spectroscopy unit for acquiring elemental composition information and a hyperspectral imaging unit for acquiring surface chemical composition distribution information, achieving complete chemical property analysis of the cemented carbide workpiece from the inside out. It can penetrate the interference of surface physical states, directly identify the material nature of the workpiece, and obtain accurate chemical properties. The compositional analysis results enhance the system's robustness and accuracy in identifying real industrial waste such as worn, contaminated, and coated materials. The control unit, communicating with the eye-based multimodal sensing device and the multi-joint robot body, receives and fuses elemental composition information and surface chemical composition distribution information to identify the chemical material category of cemented carbide workpieces and generate corresponding sorting control commands. This enables synchronous and collaborative analysis of the workpiece's internal chemical composition and surface chemical state, thereby improving the reliability of identification based on intrinsic material properties and enhancing the accuracy and system stability of automated sorting of mixed cemented carbide in real industrial scenarios. The end effector, installed at the operating end of the multi-joint robot body, is used to grasp and sort cemented carbide workpieces of the corresponding category according to the sorting control commands. This invention, by integrating a LIBS unit, possesses a high chemical composition sensing capability, significantly improving the identification accuracy of complex mixed cemented carbide chemical materials and the sorting efficiency of cemented carbide workpieces, achieving automated and high-precision sorting of cemented carbide based on its intrinsic chemical composition. Attached Figure Description

[0016] Figure 1 A schematic diagram of an embodiment of the robot-based cemented carbide sorting system provided by the present invention; Figure 2 A schematic diagram of another embodiment of the robot-based cemented carbide sorting system provided by the present invention; Figure 3 This is a schematic flowchart of an embodiment of the cemented carbide sorting method provided by the present invention. Detailed Implementation

[0017] 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 a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0018] In the description of the embodiments of the present invention, unless otherwise stated, "a plurality of" means two or more.

[0019] The terms "first," "second," etc., used in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.

[0020] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] This invention provides a robot-based cemented carbide sorting system and method, which will be described below.

[0022] Figure 1 This is a schematic diagram of an embodiment of the robot-based cemented carbide sorting system provided by the present invention, as shown below. Figure 1 As shown, the robot-based cemented carbide sorting system includes: Multi-joint robot body 10; The conveying mechanism 20 is located within the working area of ​​the multi-joint robot body and is used to convey the cemented carbide workpieces to be sorted. An eye-based multimodal sensing device 30 is integrated into the head of the multi-joint robot body and is used to perform synchronous detection and composition analysis on the cemented carbide workpiece. The eye-based multimodal sensing device 30 includes at least a laser-induced breakdown spectroscopy unit 31 for acquiring elemental composition information and a hyperspectral imaging unit 32 for acquiring surface chemical composition distribution information. The control unit 40 is communicatively connected to the eye multimodal sensing device 30 and the multi-joint robot body 10, and is used to receive and fuse the elemental composition information and surface chemical composition distribution information, identify the chemical material category of the cemented carbide workpiece, and generate corresponding sorting control instructions. An end effector 50 is installed at the operating end of the multi-joint robot body 10 and is used to grasp and sort cemented carbide workpieces of the corresponding workpiece category according to the sorting control command.

[0023] The cemented carbide workpieces to be sorted are solid wastes of cemented carbide, including drill bits and CNC cutting tools. Drill bits and CNC cutting tools can be distinguished according to their geometry. The drill bits are divided into coated drill bits and uncoated drill bits. Since the binder phases (Co, Ni) of cemented carbide workpieces are different, the subsequent recycling processes will be different. Mixed recycling will reduce the efficiency and purity of metal recycling and increase costs. Therefore, in this embodiment, it is necessary to classify the chemical cost of drill bits and CNC cutting tools, such as distinguishing whether the drill bit is WC-Co or WC-Ni, and then sort them into different collection containers.

[0024] The multi-jointed robot body 10 can be a multi-degree-of-freedom humanoid robot.

[0025] The inventors discovered that currently, the mechanisms for sorting cemented carbide workpieces are usually sorting machines based on fixed robotic arms with simple grippers or pneumatic suction cups, and their operation modes are often relatively simple. Therefore, this embodiment adopts a robot-based sorting system, which can efficiently process cemented carbide workpieces of various shapes and improve the flexibility of sorting execution.

[0026] The conveying mechanism 20 can be a conveyor belt running at a constant speed, located in the working area directly in front of the multi-joint robot body 10. The conveying mechanism 20 is used to transport the cemented carbide workpieces to be sorted, i.e., mixed cemented carbide workpieces (not shown in the figure), to the inspection station in front of the multi-joint robot body 10. The speed of the conveyor belt can be adjusted according to the sorting speed.

[0027] The eye-based multimodal sensing device 30 is integrated into the head of the robot body 10 using a biomimetic design. Simulating the layout of a human eye, it integrates multiple sensors to form the robot's eye, enabling multi-dimensional perception of microscopic components and real-time sorting of solid waste from cemented carbide workpieces.

[0028] The Laser-Induced Breakdown Spectroscopy (LIBS) unit 31 is used to perform compositional analysis on cemented carbide workpieces and obtain elemental composition information, such as quantitatively determining the content of metallic elements such as W, Co and Ni in cemented carbide workpieces, so that the system has the ability to perform in-situ, non-destructive chemical composition analysis.

[0029] The hyperspectral imaging unit 32 and the LIBS unit 31 work together. The hyperspectral imaging unit 32 and the LIBS unit 31 are arranged in a human-eye-like layout on the head of the multi-joint robot body 10. While the LIBS unit 31 performs component analysis, it scans the surface of the cemented carbide workpiece to obtain the continuous spectral curve of each pixel, thereby generating a hyperspectral image that reflects the distribution information of the surface chemical composition. Hyperspectral imaging is more sensitive to the surface condition and can effectively identify the characteristic spectra of wear-resistant coatings such as TiN, TiCN, and Al2O3. It can also distinguish the exposed cemented carbide substrate, oxide layer, or oil stains.

[0030] Understandably, in this embodiment, through the collaborative work of the LIBS unit and the hyperspectral imaging unit, two types of deep chemical characteristic data are simultaneously acquired: compositional information based on internal elemental composition and chemical imaging information based on surface compound distribution. This enables joint identification of cemented carbide workpieces from two key dimensions: "internal composition" and "surface state." This multi-dimensional chemical sensing method achieves complete chemical property analysis of cemented carbide workpieces from the inside out, effectively overcoming the limitations of traditional vision or single-sensor technologies in complex conditions such as coating obscuring, similar composition, surface contamination, or reflection. It enhances the ability to determine the intrinsic material nature of the workpiece. Therefore, it can reliably distinguish workpieces with identical external geometry but different binder phase compositions, such as "WC-Co alloy" and "WC-Ni alloy." It can penetrate the interference of surface physical states, directly identify the material nature of the workpiece, and obtain accurate chemical composition analysis results. This enhances the system's robustness and identification accuracy against real industrial waste such as wear, contamination, and coatings.

[0031] In this embodiment, the control unit 40 can be an industrial computer or a high-performance embedded controller, which is connected to the eye multimodal sensing device 30 and the multi-joint robot body 10 via a cable. The control unit 40 runs specialized data fusion and classification software. This software receives and synchronizes time-stamped elemental composition information and surface chemical composition distribution information in real time, and then performs fusion analysis on the elemental composition information and surface chemical composition distribution information to determine the chemical material category. For example, in one specific embodiment, the fusion analysis logic is as follows: based on the elemental composition information, it determines that the substrate material is "WC-Co alloy" or "WC-Ni alloy". Combined with the surface chemical composition distribution information, it determines whether there is a specific coating on the surface of the cemented carbide workpiece and the coverage of the coating. Finally, it outputs a comprehensive "chemical material category", such as "uncoated WC-Co drill bit", "WC-Co milling cutter with TiN coating" or "WC-Ni CNC insert".

[0032] In another specific implementation, the fusion analysis logic includes: positive verification: when the elemental composition information determines it to be "WC-Co", and the surface chemical composition distribution information identifies the "TiN coating" feature, the two support each other, and the system can output the final category as "WC-Co alloy drill bit with TiN coating"; conflict arbitration and error correction: when the results of the two contradict each other (for example, the LIBS unit signal is extremely weak due to the insulation of the thick coating, making it difficult to determine the composition, but the hyperspectral image clearly identifies the coating), or the hyperspectral image fails due to severe oil contamination, but the LIBS unit penetrates the contamination and provides a clear elemental spectrum, the system will initiate arbitration logic, such as marking low confidence, triggering re-detection, or assigning different weights to the signal quality for weighted decision-making. For example, if the LIBS signal quality is extremely high while the hyperspectral signal is affected by strong reflective interference, the decision depends on the LIBS result.

[0033] Then, the control unit 40 plans the robot's grasping path based on the chemical material category and generates sorting control instructions containing target position, posture, and clamping force parameters. Understandably, by fusing elemental composition information from the laser-induced breakdown spectroscopy unit and surface chemical distribution information from the hyperspectral imaging unit, the control unit 40 possesses a high level of visual chemical composition analysis capability. This improves the recognition accuracy and robustness for complex, coated, or contaminated workpieces, enabling simultaneous and collaborative analysis of the workpiece's internal chemical composition and surface chemical state. This enhances the reliability of material-based identification and improves the accuracy and system stability of automated sorting of mixed cemented carbides in real industrial scenarios.

[0034] The end effector 50 can be an adaptive bionic gripper with force-sensing feedback, mounted on the last joint (operating end) of the multi-joint robot body 10 via a flange. This gripper has multiple phalanges that can move independently or collaboratively, with pressure sensors integrated into the fingertips. Upon receiving instructions from the control unit 40, it automatically adjusts the finger angle, envelope posture, and gripping force according to the chemical material type of the workpiece to be grasped. This achieves precise and efficient identification and sorting of cemented carbide workpieces, enabling the system to simulate human operational flexibility while possessing high chemical composition sensing capabilities. This significantly improves the identification accuracy of complex mixed cemented carbide chemical materials and the sorting efficiency of cemented carbide workpieces, realizing automated and high-precision sorting of cemented carbide based on its intrinsic chemical composition.

[0035] In summary, the robot-based cemented carbide sorting system provided in this embodiment of the invention includes a multi-joint robot body, which can efficiently handle cemented carbide workpieces of various shapes and improve the flexibility of sorting execution; a conveying mechanism, set in the working area of ​​the multi-joint robot body, is used to transport the cemented carbide workpieces to be sorted; an eye-like multimodal sensing device, integrated into the head of the multi-joint robot body, is used for simultaneous detection and composition analysis of the cemented carbide workpieces; the eye-like multimodal sensing device includes at least a laser-induced breakdown spectroscopy unit for acquiring elemental composition information and a hyperspectral imaging unit for acquiring surface chemical composition distribution information, realizing complete chemical property analysis of cemented carbide workpieces from the inside out, able to penetrate the interference of surface physical state, directly identify the material nature of the workpiece, and obtain accurate... Accurate chemical composition analysis results enhance the system's robustness and recognition accuracy against real industrial waste such as wear, contamination, and coatings. The control unit, communicating with the eye-based multimodal sensing device and the multi-joint robot body, receives and fuses elemental composition information and surface chemical composition distribution information to identify the chemical material category of cemented carbide workpieces and generate corresponding sorting control commands. This enables synchronous and collaborative analysis of the workpiece's internal chemical composition and surface chemical state, thereby improving the reliability of material-based identification and enhancing the accuracy and system stability of automated sorting of mixed cemented carbide in real industrial scenarios. The end effector, installed at the operating end of the multi-joint robot body, performs the gripping and sorting of cemented carbide workpieces of the corresponding category according to the sorting control commands. This invention, by integrating a LIBS unit, possesses high chemical composition sensing capabilities, significantly improving the recognition accuracy of complex mixed cemented carbide chemical materials and the sorting efficiency of cemented carbide workpieces, achieving automated and high-precision sorting of cemented carbide based on its internal chemical composition.

[0036] In some embodiments of the present invention, the laser-induced breakdown spectroscopy unit includes a pulsed laser, a spectrometer, and a signal processing unit; the pulsed laser is used to generate high-energy pulsed laser light and focus it on the test point on the surface of the cemented carbide workpiece to excite and generate plasma; the spectrometer is used to collect the characteristic spectral signals emitted by the plasma; the signal processing unit is used to analyze the characteristic spectral lines and their intensities of preset metal elements in the characteristic spectral signals to determine the elemental composition information of the cemented carbide workpiece.

[0037] The preset metal element can be tungsten, cobalt, or nickel.

[0038] Specifically, a pulsed laser generates high-energy pulsed laser light, which is focused onto the surface of the cemented carbide workpiece under test, generating plasma. A spectrometer is used to collect the spectral signals emitted by the plasma. The signal processing unit analyzes the characteristic spectral lines and intensities of metallic elements such as W, Co, and Ni to qualitatively and quantitatively determine the content of each metal in the sample. Understandably, in this embodiment, by integrating laser-induced breakdown spectroscopy technology into the robot head, it is possible to directly excite and analyze the target point during workpiece transport without destructive sampling or contact. The pulsed laser causes only micron-level minimal damage, completely unaffected by the overall recycling value of the workpiece, achieving in-situ, minimally destructive chemical composition analysis of the cemented carbide workpiece.

[0039] In one specific embodiment, the pulsed laser can be a nanosecond-level Q-switched solid-state laser with an operating wavelength of 1064 nm, a pulse energy adjustable in the range of 10-100 mJ, and a repetition frequency of 1-10 Hz. The high-energy pulsed laser emitted by this laser is precisely focused onto the test point on a cemented carbide workpiece (such as the cutting edge or surface of a discarded drill bit) through a focal-adjustable focusing lens group. The laser energy acts on an extremely small area (spot diameter of approximately 50-200 μm) for an extremely short time, causing the material at that point to instantaneously vaporize and be excited to form a high-temperature, high-energy plasma, thereby achieving micron-level ablation and excitation of the sample surface.

[0040] The spectrometer can be a high-resolution, wide-band echelle grating spectrometer with a spectral response range covering 200-900 nm. This spectrometer uses a collection lens group to be aligned with the plasma emission region, efficiently collecting characteristic spectral signals rich in elemental information emitted during plasma cooling. The spectrometer is synchronized with the pulsed laser via a precise timing controller, ensuring data acquisition occurs within the time window of highest plasma emission intensity to obtain the spectrum with the optimal signal-to-noise ratio.

[0041] The signal processing unit can be an integrated high-speed data acquisition card and embedded processor. Its workflow is as follows: First, it receives the raw spectral data stream from the spectrometer and performs preprocessing such as dark current subtraction, spectral correction, and filtering to obtain a clear spectral curve. Then, based on the built-in elemental characteristic spectral line database, it automatically identifies and locates the characteristic emission lines of key metal elements. The preset metal elements include at least tungsten (W) characterizing the cemented carbide matrix (e.g., using 405.88 nm or 407.44 nm spectral lines), cobalt (Co) characterizing the main binder phase (e.g., using 340.51 nm or 345.35 nm spectral lines), and nickel (Ni) characterizing the main binder phase (e.g., using 341.48 nm or 352.45 nm spectral lines). The algorithm performs peak area integration or peak height calculation on the identified spectral lines to quantify their intensity I_W, I_Co, and I_Ni. Finally, using a pre-defined analytical model (e.g., semi-quantitative analysis using calibration curves based on standard samples, or directly calculating the intensity ratio of I_Co / (I_Co+I_Ni) as a criterion for the binder phase type), the extracted spectral line intensities are converted into elemental composition information characterizing the material's intrinsic composition. This information can be a classification label indicating "Co-dominant" or "Ni-dominant," or a more precise estimate of the relative content of Co and Ni.

[0042] In some embodiments of the present invention, the hyperspectral imaging unit includes a hyperspectral camera, the response wavelength of which includes the visible light band and the near-infrared band, for acquiring the characteristic spectral reflectance curve of the coating on the surface of the cemented carbide workpiece, forming a surface chemical image, and performing chemical composition identification on the surface chemical image to obtain surface chemical composition distribution information.

[0043] The hyperspectral camera operates in a band that covers the visible light (e.g., 400-780nm) to near-infrared (e.g., 780-1700nm) range, and can capture the surface spectral characteristics of cemented carbide workpieces in the visible and near-infrared bands, i.e., surface spectral information, which is used to identify the coating state of cemented carbide workpieces.

[0044] Specifically, a hyperspectral camera acquires continuous spectral data for each pixel of a cemented carbide workpiece, forming surface spectral information. This surface spectral information can effectively distinguish the unique spectral characteristics of coating materials such as TiN and Al2O3 from those of the WC-Co substrate. Because coating materials (such as TiN and Al2O3) and the exposed cemented carbide substrate have unique spectral reflectance curves at different wavelengths, accurate identification of the coating condition can be achieved, overcoming the problem of low recognition rate of traditional RGB cameras under coating wear or reflective conditions.

[0045] In one specific implementation, when a cemented carbide workpiece enters the imaging field of view via a conveyor, a hyperspectral camera performs line scanning or area array imaging of the moving workpiece under specific lighting conditions (such as a uniform LED light source). A complete spectral curve is recorded for each pixel in the image, instead of the traditional RGB three-channel values, thus generating a three-dimensional data cube containing spatial and spectral dimensions, i.e., a surface chemical image. The system's built-in image processing algorithm (based on a pre-established spectral library) performs real-time analysis on the acquired surface chemical image, such as spectral curve comparison: matching the measured spectral reflectance curve of each pixel with the characteristic spectral reflectance curves pre-stored in the database. These characteristic curves correspond to different surface states; chemical composition identification and classification: through spectral matching, characteristic wavelength extraction, or chemometric methods (such as spectral angle mapping, linear spectral unmixing, etc.), the algorithm identifies the material of each pixel or region, determining its chemical composition or coating type. Finally, the pixel-level identification results are integrated to obtain comprehensive surface chemical composition distribution information, including the presence or absence of a coating on the workpiece surface, coating type, coating distribution uniformity, and exposed substrate area. This information can be a pseudo-color classification image or a structured data report. The hyperspectral imaging unit works in tandem with the LIBS unit. LIBS provides point-like, deep-layer quantitative information on elements (such as Co and Ni content), while hyperspectral imaging provides surface-level, qualitative distribution information on compounds. The two units operate synchronously or in close coordination in time and space, and their analytical results are transmitted to the control unit for cross-verification. For example, if LIBS identifies a point as a WC-Co alloy, hyperspectral imaging may confirm the presence of a TiN coating on the surface of that area. The combination of these two methods makes the classification results more accurate and reliable (e.g., identifying it as a "WC-Co drill bit with a TiN coating").

[0046] In some embodiments of the present invention, the eye multimodal sensing device further includes a three-dimensional contour measurement unit for acquiring three-dimensional geometric information and spatial position information of the cemented carbide workpiece; the control unit is further used to fuse the three-dimensional geometric information and assist in identifying the material type based on the three-dimensional geometric information, and / or assist in generating sorting control commands based on the spatial position information.

[0047] The three-dimensional contour measurement unit can be a 3D structured light profiler or a dToF lidar, integrated into the head of the multi-joint robot body. It is used to acquire the three-dimensional geometric information and spatial position information of the cemented carbide workpiece to guide the laser alignment of the LIBS unit and the gripping and positioning of the sorting actuator.

[0048] Specifically, to further improve sorting accuracy and reliability, the eye-based multimodal sensing device also includes a 3D contour measurement unit to acquire 3D point cloud data of cemented carbide workpieces, accurately measuring the workpiece's contour dimensions and geometry. By performing point cloud registration and feature extraction on the 3D point cloud data, it can distinguish the helical groove structure of drill bits, the tip angle, and specific planar geometric features of CNC inserts. The 3D geometric information provides crucial context for material classification. For example, the control unit fuses the "drill bit" shape features identified by the 3D contour with the "WC-Co" composition results obtained from LIBS unit analysis and the "TiN coating" surface state characterized by hyperspectral imaging. This allows it to output a comprehensive category—"WC-Co drill bit with TiN coating"—that includes geometric morphology and is more instructive for process control.

[0049] Based on three-dimensional geometric information (such as the shape, size, and center of gravity of the workpiece) and spatial position information, the control unit plans a safe and efficient grasping path and posture for the end effector. For example, for a slender drill bit, the plan is to grasp it along its axis; for a flat cutting tool, the plan is to grip it from the side. Combining the position of the target collection container, the control unit translates the above path planning into a precise sorting control command sequence for each joint of the robot, ensuring that the robot can accurately and smoothly transfer the workpiece to the designated position.

[0050] It is worth noting that the three-dimensional geometric information provides the workpiece's six degrees of freedom (position and orientation) in space, ensuring the success rate of grasping and operational safety.

[0051] In some embodiments of the present invention, such as Figure 2 The diagram shown is a structural schematic of another embodiment of a robot-based cemented carbide sorting system; the control unit 40 includes a chemical material classification module 41 and a motion control module 42. The chemical material classification module 41 is used to fuse the elemental composition information and the surface chemical composition distribution information to identify the chemical material category of the cemented carbide workpiece and obtain the identification result. The motion control module 42 is used to generate the sorting control command based on the recognition result.

[0052] The chemical material classification module supports two algorithm modes: a rule-based judgment mode, which uses IF-THEN logic based on spectral line intensity ratios for fast response and is suitable for online real-time processing; and a machine learning mode, which uses a pre-trained linear discriminant analysis model to map multi-dimensional feature vectors to the classification space, resulting in higher classification accuracy.

[0053] Specifically, the chemical material classification module 41 receives elemental composition information, including characteristic spectral lines and intensity data of metal elements such as tungsten, cobalt, and nickel, as well as surface chemical composition distribution information, including spectral feature images of the workpiece in the visible and near-infrared bands. It integrates multi-source sensor data, performs chemical material identification on the cemented carbide workpiece through a preset algorithm model, and outputs identification results including specific material categories (such as WC-Co alloy, WC-Ni alloy, etc.).

[0054] The motion control module 42 receives the identification results output by the chemical material classification module, combines the spatial position of the cemented carbide workpiece, plans the optimal gripping path and posture of the multi-joint robot body 10, generates sorting control instructions for the end effector, including gripping position, clamping force parameters, etc., controls the start, stop and speed of the conveying mechanism, and coordinates the rhythm of the entire sorting process.

[0055] It is worth noting that the chemical material classification module 41 is also used to compare and verify the elemental composition information with the surface chemical composition distribution information to determine the chemical material category of the cemented carbide workpiece. The specific verification method is as follows: The surface chemical composition distribution information is used as the first criterion for surface condition to identify whether a preset coating characteristic spectrum exists on the workpiece surface; the elemental composition information is used as the second criterion for the material matrix, and the material type of the cemented carbide matrix (such as WC-Co or WC-Ni) is determined by analyzing the characteristic spectral intensity and ratio of binder phase elements such as cobalt and nickel; the first and second criteria are cross-verified: if the first criterion identifies coating characteristics and the second criterion determines that the matrix is ​​a specific cemented carbide material, then the workpiece is determined to be a cemented carbide of the corresponding material with that type of coating; if the first criterion does not identify coating characteristics, but the second criterion clearly determines the matrix material, then the workpiece is determined to be a cemented carbide of the corresponding material without coating; if the judgment results of the first and second criteria contradict each other, for example, coating spectral characteristics are detected but the matrix element signal is abnormal, or the matrix element signal is clear but the surface spectrum is severely interfered with by contamination, then the module will activate at least one of the following countermeasures: a) Mark the classification confidence level of the workpiece as low and output an instruction to suggest manual review or secondary inspection; b) Use the geometric features provided by the three-dimensional contour measurement unit to assist in decision-making, for example, inferring that the surface of the workpiece should have a coating based on the regular blade shape of the workpiece; c) Control the eye multimodal sensor to re-detect different parts of the same workpiece, acquire new sensor data for re-fusion and judgment.

[0056] By using the above-mentioned method of cross-verification of multi-source information, it is possible to effectively distinguish various complex working conditions such as intact surface coating, partially worn coating, surface contaminants, and exposed substrate, thereby improving the ability to accurately and robustly identify the chemical material type and surface condition of cemented carbide workpieces under complex industrial conditions.

[0057] In some embodiments of the present invention, the chemical material classification module includes a pre-trained deep learning model or a pattern recognition algorithm.

[0058] The pattern recognition algorithm can be a logical judgment algorithm or a linear discrimination algorithm based on the spectral line intensity ratio. For example, a logical judgment algorithm based on the spectral line intensity ratio can achieve rapid classification by calculating the ratio R = I_Co / (I_Co + I_Ni) and setting a threshold (e.g., R > 0.8 is classified as WC-Co, R < 0.2 is classified as WC-Ni).

[0059] Specifically, pre-trained deep learning models or pattern recognition algorithms are built and trained offline through the following steps: Data Acquisition: A large number of cemented carbide standard samples of known categories and compositions were collected. Each sample was scanned using an eye-based multimodal sensor, simultaneously acquiring its LIBS characteristic spectral signals and hyperspectral image data. At the same time, the three-dimensional contour information of the samples was recorded.

[0060] Feature engineering: Preprocessing and feature extraction of the acquired raw data. Denoising and baseline correction are performed on the LIBS spectral signal, and the characteristic spectral line intensities of key elements are extracted, such as the spectral line intensities of cobalt (Co) at specific wavelengths (e.g., ~340.51 nm, ~345.35 nm) (I_Co), the spectral line intensities of nickel (Ni) at specific wavelengths (e.g., ~341.48 nm, ~352.45 nm) (I_Ni), and the reference spectral line intensity of tungsten (W) (I_W).

[0061] Dimensionality reduction and feature extraction are performed on hyperspectral image data to obtain spectral indices or feature vectors that reflect the surface chemical composition.

[0062] Geometric features, such as shape factor, aspect ratio, and curvature, are extracted from 3D contour data to distinguish different tool types, such as drill bits, milling cutters, and inserts.

[0063] Model training: The extracted multimodal features (LIBS elemental intensity features, hyperspectral chemical features, and three-dimensional geometric features) and the sample's true labels (such as "WC-Co drill bit - uncoated" and "WC-Ni cutting tool - with TiN coating") are used as training data pairs.

[0064] If a deep learning model (such as a multilayer perceptron MLP or a convolutional neural network CNN) is used, multimodal features are input into the network, and the network weights are adjusted through the backpropagation algorithm, enabling the model to learn the nonlinear mapping relationship from complex features to material categories.

[0065] If pattern recognition algorithms (such as Support Vector Machine (SVM), Random Forest, or Linear Discriminant Analysis (LDA) are used, these algorithms are used to find the optimal classification boundary or construct decision rules in the feature space.

[0066] Model validation and solidification: The model performance is evaluated using an independent validation dataset, the parameters are optimized, and finally the trained model parameters and structure are solidified and deployed in the chemical material classification module of the control unit.

[0067] Understandably, this embodiment integrates the quantitative information on internal elements provided by LIBS and the surface compound distribution information provided by spectral imaging, overcoming the limitations of traditional technologies that rely solely on vision or a single sensor. Based on intelligent judgment through multimodal data fusion, this embodiment achieves automated and high-precision differentiation of the intrinsic chemical composition differences between WC-Co and WC-Ni, solving the technical challenges of inaccurate internal composition determination by manual sorting and the limitation of traditional automated equipment that can only sort by physical morphology.

[0068] In some embodiments of the present invention, the chemical material category includes at least WC-Co alloy and WC-Ni alloy.

[0069] Specifically, by integrating LIBS technology, the robot is equipped with "material composition vision," enabling it to directly and non-destructively read information on key elements (Co, Ni) that determine alloy categories, thus achieving high-precision sorting of solid waste materials from cemented carbide workpieces based on chemical cost.

[0070] In some embodiments of the present invention, the end effector is a force-controlled bionic gripper.

[0071] The end effector is an adaptive bionic gripper with force feedback functionality. For example, a bionic hand often has two or three "fingers" that can move independently or in coupling. Each finger contains multiple bionic phalanges, driven by built-in micromotors or pneumatic artificial muscles, capable of simulating the bending, opening, and lateral swinging movements of human fingers, thereby achieving various complex grasping envelopes. Furthermore, the bionic gripper integrates a multi-dimensional force / torque sensor and a tactile sensor array. The force / torque sensors are mounted on the wrist of the gripper or the base of each finger to measure and provide real-time feedback on the magnitude of the clamping force applied to the workpiece during grasping, as well as any lateral forces or torques that may be present. The tactile sensors can be distributed on the gripping surfaces inside the fingers to sense the pressure distribution and sliding tendency at the contact points with the workpiece.

[0072] Specifically, the force-controlled bionic gripper performs the grasping and sorting of cemented carbide workpieces of the corresponding workpiece categories. By utilizing force feedback and control, it ensures that the clamping force is always limited within the safe threshold, fundamentally eliminating the risk of chipping, micro-cracks, or coating peeling of the precision cutting edge of cemented carbide due to excessive clamping. This protects the cutting edge of the cemented carbide workpiece and improves sorting safety and efficiency.

[0073] In one specific implementation, the motion control module plans the optimal grasping path for the robot based on the recognition results and the real-time position of the workpiece, and controls the end effector (such as an adaptive bionic gripper) to place different types of workpieces into designated collection containers.

[0074] like Figure 3 As shown, the present invention also provides a sorting method based on cemented carbide. Figure 3 This is a schematic flowchart of an embodiment of the cemented carbide-based sorting method provided by the present invention. The method includes: S301. The cemented carbide workpieces to be sorted are continuously conveyed through the conveying mechanism; S302. When the cemented carbide workpiece enters the inspection station, the surface spectral information, three-dimensional geometric information and surface electrical conductivity information of the cemented carbide workpiece are simultaneously acquired through the eye multimodal sensing device. S303. By fusing elemental composition information and surface chemical composition distribution information through the control unit, the chemical material category of the cemented carbide workpiece is identified, and sorting control instructions are generated. S304. Execute the sorting control command to grasp and sort the cemented carbide workpieces of the corresponding chemical material category.

[0075] It should be noted that in the embodiments of the cemented carbide sorting method described above, please refer to the corresponding description in the robot-based cemented carbide sorting system above. The beneficial effects that can be achieved can also be referred to the beneficial effects described above, and will not be repeated here.

[0076] For embodiments based on the cemented carbide sorting method, please refer to the steps performed by the corresponding processing unit in the above text. For other embodiments based on the cemented carbide sorting method that are not described, please refer to the corresponding content mentioned above, which will not be repeated here.

[0077] In some embodiments of the present invention, the chemical material category includes at least WC-Co alloy and WC-Ni alloy; the step of identifying the chemical material category of the cemented carbide workpiece by fusing elemental composition information and surface chemical composition distribution information through the control unit includes: extracting the characteristic spectral line intensities of cobalt and nickel based on the elemental composition information; calculating the intensity ratio characterizing the relative content of cobalt and nickel based on the characteristic spectral line intensities of cobalt and nickel; performing a preliminary classification of the matrix material of the cemented carbide workpiece based on the intensity ratio and a preset ratio range to obtain a preliminary classification result; and verifying or correcting the preliminary classification result in conjunction with the surface chemical composition distribution information to determine the chemical material category.

[0078] Specifically, the control unit fuses, verifies, and corrects the preliminary classification results with the surface chemical composition distribution information, including the following three scenarios: Verification and refinement: If the initial classification is "WC-Co alloy," and hyperspectral image analysis shows obvious TiN coating characteristic spectra in the laser application point and surrounding area, then the control unit ultimately determines the chemical material category as "WC-Co alloy drill bit with TiN coating." This achieves accurate identification of the "coating-substrate" composite state.

[0079] Correction and error handling scenarios: If the initial classification is "WC-Co alloy," but the hyperspectral image shows that the spectrum of this area highly matches the characteristics of severe oil contamination or iron oxide, the control unit may determine that surface contamination interferes with the accuracy of the LIBS signal. Two strategies can be adopted: 1) Mark the classification result as having low confidence and control the robot to sort it to the "re-inspection" bin; 2) Combine the hyperspectral image to find a cleaner exposed substrate area and trigger LIBS to perform a second inspection, thereby improving the robustness of the system in complex industrial environments.

[0080] Decision support scenario: If the initial classification is "WC-Ni alloy", but no typical coating features are detected in the hyperspectral image, the final category can be determined as "uncoated WC-Ni alloy workpiece".

[0081] Based on the final determined chemical material category, the control unit generates sorting control instructions that include the target placement location. These instructions drive the multi-joint robot body, guiding the end effector to grasp the drill bit with an appropriate gripping force and accurately place it into the designated hopper.

[0082] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0083] The above provides a detailed description of the robot-based cemented carbide sorting system and method provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A robot-based cemented carbide sorting system, characterized in that, include: Multi-jointed robot body; A conveying mechanism is located within the working area of ​​the multi-joint robot body and is used to convey cemented carbide workpieces to be sorted. An eye-based multimodal sensing device is integrated into the head of the multi-joint robot body for synchronous detection and composition analysis of the cemented carbide workpiece. The eye-based multimodal sensing device includes at least a laser-induced breakdown spectroscopy unit for acquiring elemental composition information and a hyperspectral imaging unit for acquiring surface chemical composition distribution information. The control unit is communicatively connected to the eye multimodal sensing device and the multi-joint robot body, and is used to receive and fuse the elemental composition information and surface chemical composition distribution information, identify the chemical material category of the cemented carbide workpiece, and generate corresponding sorting control instructions. An end effector, installed at the operating end of the multi-joint robot body, is used to grasp and sort cemented carbide workpieces of the corresponding workpiece category according to the sorting control command.

2. The robot-based cemented carbide sorting system according to claim 1, characterized in that, The laser-induced breakdown spectroscopy unit includes a pulsed laser, a spectrometer, and a signal processing unit. The pulsed laser is used to generate high-energy pulsed laser and focus it on the test point on the surface of the cemented carbide workpiece to excite and generate plasma. The spectrometer is used to collect the characteristic spectral signals emitted by the plasma; The signal processing unit is used to analyze the characteristic spectral lines and their intensities of preset metal elements in the characteristic spectral signal to determine the elemental composition information of the cemented carbide workpiece.

3. The robot-based cemented carbide sorting system according to claim 1, characterized in that, The hyperspectral imaging unit includes a hyperspectral camera, whose response wavelengths include the visible light band and the near-infrared band. It is used to acquire the characteristic spectral reflectance curve of the surface coating of the cemented carbide workpiece, form a surface chemical image, identify the chemical composition of the surface chemical image, and obtain the surface chemical composition distribution information.

4. The robot-based cemented carbide sorting system according to claim 1, characterized in that, The eye multimodal sensing device also includes a three-dimensional contour measurement unit for acquiring the three-dimensional geometric information and spatial position information of the cemented carbide workpiece; the control unit is also used to fuse the three-dimensional geometric information and assist in the identification of the material type based on the three-dimensional geometric information, and / or assist in the generation of sorting control commands based on the spatial position information.

5. The robot-based cemented carbide sorting system according to claim 1, characterized in that, The control unit includes a chemical material classification module and a motion control module; The chemical material classification module is used to fuse the elemental composition information and the surface chemical composition distribution information to identify the chemical material category of the cemented carbide workpiece and obtain the identification result. A motion control module is used to generate the sorting control command based on the recognition result.

6. The robot-based cemented carbide sorting system according to claim 5, characterized in that, The chemical material classification module includes a pre-trained deep learning model or pattern recognition algorithm.

7. The robot-based cemented carbide sorting system according to claim 1, characterized in that, The chemical material categories include at least WC-Co alloys and WC-Ni alloys.

8. The robot-based cemented carbide sorting system according to claim 1, characterized in that, The end effector is a force-controlled bionic gripper.

9. A method for sorting cemented carbide using a robot-based cemented carbide sorting system as described in any one of claims 1-8, characterized in that, include: The cemented carbide workpieces to be sorted are continuously conveyed by a conveying mechanism; When the cemented carbide workpiece enters the inspection station, the surface spectral information, three-dimensional geometric information and surface electrical conductivity information of the cemented carbide workpiece are simultaneously acquired through the eye multimodal sensor. By fusing elemental composition information and surface chemical composition distribution information through the control unit, the chemical material category of the cemented carbide workpiece is identified, and sorting control instructions are generated. The sorting control command is executed to grasp and sort cemented carbide workpieces of the corresponding chemical material category.

10. The cemented carbide sorting method according to claim 9, characterized in that, The chemical material categories include at least WC-Co alloys and WC-Ni alloys; the identification of the chemical material category of the cemented carbide workpiece by fusing elemental composition information and surface chemical composition distribution information through the control unit includes: Based on the elemental composition information, the characteristic spectral line intensities of cobalt and nickel are extracted. Based on the characteristic spectral line intensities of cobalt and nickel, the intensity ratio characterizing the relative content of cobalt and nickel is calculated. Based on the strength ratio and the preset ratio range, the base material of the cemented carbide workpiece is preliminarily classified to obtain a preliminary classification result; Based on the surface chemical composition distribution information, the preliminary classification results are verified or corrected to determine the chemical material category.

Images