Cutting waste grabbing and transferring control system and method

By combining inductive sensors and visual detection with high-definition cameras to identify waste information, and controlling the robotic arm to grasp and transport it, the problems of high labor costs and safety hazards in waste cutting and processing are solved, realizing the automated and intelligent processing of waste.

CN121017831APending Publication Date: 2025-11-28CHONGQING SHANLIAN SCI & TECH DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511150073.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-08-18
Publication Date
2025-11-28

AI Technical Summary

Technical Problem

Existing technologies have difficulty adapting to the handling and storage of cutting waste, resulting in the need for a large amount of manual operation, increasing labor costs, and posing safety hazards such as waste slippage and collisions.

Method used

An array of inductive sensors is used to detect the edge position of the waste material. Combined with phase difference analysis, the laser cutting power is dynamically matched. A vision detection module generates an avoidance cutting path. A high-definition camera identifies the waste material information and controls the robotic arm to plan the optimal gripping path and force, thereby realizing the automated gripping and transfer of waste material.

Benefits of technology

It has enabled automated waste processing, reduced labor costs, avoided safety hazards such as slippage and collision, and improved production efficiency and safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121017831A_ABST
    Figure CN121017831A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of industrial production, in particular to a grabbing and transferring control system and method for cutting waste materials, and after a cutting device completes material cutting, a conveying mechanism conveys the waste materials to a secondary cutting station; the inductive sensor array detects the edge of the waste, analyzes the type of the metal material, and dynamically matches the laser cutting power; the visual detection module scans surface defects of the waste materials to generate an avoiding cutting path, so that the laser cutting machine divides the waste materials into fragments with set sizes; the high-definition camera collects images, waste information is recognized through processing and transmitted to the control center, and the control center plans the optimal grabbing path, grabbing point and force by combining a mechanical arm model and an algorithm; the mechanical arm moves according to the plan to grab the waste materials and then conveys the waste materials to a waste material frame according to a transfer path; by means of the mode, the problems that when traditional equipment treats large-size cutting waste, adaptation is difficult, the labor cost is high, and potential safety hazards exist in carrying are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of industrial production technology, and in particular to a control system and method for grabbing and transferring cutting waste. Background Technology

[0002] In modern industrial production, especially in the metal processing industry, cutting is a crucial production process. With continuous technological advancements, laser cutting technology has gained widespread application in metal cutting due to its advantages such as high precision, high speed, and excellent cutting quality. During the cutting process, materials are processed by the cutting device, resulting in finished products and waste. While there are generally mature and well-established processes for handling finished products, the handling of waste is often overlooked.

[0003] In the existing technology, when dealing with cutting waste, the waste is large in volume, so traditional handling and storage equipment is difficult to adapt. This not only requires a lot of manpower for auxiliary operation, increasing labor costs, but also poses safety hazards such as waste slipping and collision during the handling process. Summary of the Invention

[0004] The purpose of this invention is to provide a control system and method for grabbing and transferring cutting waste, which aims to solve the technical problems in the prior art where, due to the large volume of cutting waste, traditional handling and storage equipment is difficult to adapt, which not only requires a lot of manpower for auxiliary operation, increasing labor costs, but also easily leads to safety hazards such as waste slipping and collision during the handling process.

[0005] To achieve the above objectives, the present invention provides a method for controlling the grabbing and transfer of cutting waste, comprising the following steps:

[0006] After the cutting device finishes cutting the material to obtain the finished product, the conveying mechanism transports the waste material to the secondary cutting station;

[0007] The edge position of the waste material is detected by an array of inductive sensors, and the laser cutting power is dynamically matched based on the phase difference analysis of the metal material type.

[0008] The visual inspection module scans the defective areas on the surface of the waste material, generates an avoidance cutting path, and controls the laser cutting machine to cut the waste material into fragments of a set size.

[0009] High-definition cameras are used to capture images of the cutting area, and image processing algorithms are used to identify the location, shape, and size of the waste material, which is then transmitted to the control center.

[0010] Based on the waste information transmitted from the visual recognition and positioning steps, the control center combines the kinematic model of the robotic arm and the grasping force control algorithm to plan and generate the optimal grasping path, optimal grasping point, and grasping force of the robotic arm.

[0011] The robotic arm moves to the waste location according to the optimal gripping path generated by the control center, grips the waste according to the determined gripping point and gripping force, and then transfers the waste to the waste box according to the planned transfer path.

[0012] Among these methods, when using an inductive sensor array to detect the edge position of the waste material and dynamically matching the laser cutting power based on phase difference analysis of the metal material type:

[0013] Apply multi-frequency eddy current signals ranging from 0.1 to 10 MHz and collect the phase responses of different metals;

[0014] When the phase difference is between 30-45°, the material is identified as steel, and a high-power cutting mode is selected.

[0015] When the phase difference is between 60-75°, it is determined to be aluminum material, and a low-power cutting mode is matched.

[0016] In this process, when dividing waste into fragments of a set size, the cutting process is modeled as a Markov decision process, and the cutting parameters are dynamically adjusted through deep reinforcement learning to maximize cutting efficiency and minimize material waste.

[0017] The algorithm continuously optimizes the control strategy based on real-time feedback to adapt to different materials and cutting conditions.

[0018] In the step "scanning the defect area on the surface of the waste material using a visual inspection module, generating an avoidance cutting path, and controlling the laser cutting machine to cut the waste material into fragments of a set size":

[0019] An improved YOLOv5-seg model incorporating an attention mechanism was used to segment the pore and crack regions on the waste surface;

[0020] Construct a 10mm radius buffer zone centered on the defect area;

[0021] The cutting path is replanned based on the topology optimization algorithm to ensure that the fragment size error is ≤2mm.

[0022] When the robotic arm grabs waste fragments, it uses an ultrasonic scanner to monitor the volume status of the two waste bins in real time. If the current bin volume is ≥85%, it switches to the backup bin and triggers an alarm.

[0023] Specifically, piezoelectric ceramic weighing modules are deployed at the bottom of the two waste bins, and ultrasonic scanners are installed at the top;

[0024] The real-time volume is calculated by reconstructing the 3D point cloud. When the volume is ≥85% after 3 consecutive scans, the material frame is switched and the audible and visual alarm is activated.

[0025] During the switching process, the coordinates of the full material box are automatically locked to prevent the waste material gripper from feeding.

[0026] When the robotic arm grabs waste fragments, sensors monitor the movement of the robotic arm and the grabbing of waste in real time. If any abnormality or deviation occurs, the control center will make adjustments immediately.

[0027] When the robotic arm grabs waste fragments...

[0028] If an obstacle is detected, a collision-free path is generated using the fast expanding random tree algorithm;

[0029] Furthermore, based on the real-time posture of the waste fragments, the gradient descent method is used to optimize the gripping point position to ensure gripping stability.

[0030] During the entire process, heterogeneous data collected from different components are synchronized using timestamps and spatial coordinate transformations for spatiotemporal alignment, eliminating sensor delays and installation position deviations.

[0031] Specifically, a dynamic time warping algorithm based on Kalman filtering is used to achieve time synchronization of multi-source data.

[0032] In the step "the laser cutting machine divides the waste into fragments of a set size":

[0033] A coaxial infrared thermal imager was used to monitor the temperature gradient in the cutting area;

[0034] When the local temperature exceeds 90% of the material's melting point, the laser power is automatically reduced by 20%.

[0035] After cutting, the consistency of fragment composition was verified by plasma spectroscopy analysis.

[0036] The present invention also provides a control system for gripping and transferring cutting waste, including a control subsystem for performing the gripping and transferring control method for cutting waste as described above, and a power supply module for supplying power to the control subsystem.

[0037] This invention discloses a control system and method for grabbing and transferring cutting waste. In practical use, after the cutting device completes the cutting of the material to obtain the finished product, the conveying mechanism transports the waste to the secondary cutting station. An inductive sensor array is used to detect the edge position of the waste, and the laser cutting power is dynamically matched based on phase difference analysis of the metal material type. A vision detection module scans the defect area on the waste surface to generate an avoidance cutting path, controlling the laser cutting machine to divide the waste into fragments of a set size. A high-definition camera captures images of the cutting area, and an image processing algorithm identifies the position, shape, and size information of the waste, transmitting this information to the control center. The control center then uses the visual recognition positioning steps... The transmitted waste information, combined with the kinematic model of the robotic arm and the grasping force control algorithm, is used to plan and generate the optimal grasping path, optimal grasping point, and grasping force of the robotic arm. The robotic arm moves to the waste location according to the optimal grasping path generated by the control center, and grasps the waste according to the determined grasping point and grasping force. Then, the waste is transferred to the waste box according to the planned transfer path. This method solves the technical problems in the existing technology when handling cutting waste. Due to the large volume of waste, traditional handling and storage equipment is difficult to adapt. This not only requires a lot of manpower for auxiliary operation, increasing labor costs, but also poses safety hazards such as waste slippage and collision during the handling process. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart of the cutting waste grabbing and transfer control method of the present invention.

[0040] Figure 2 This is a schematic diagram of the cutting waste grabbing and transfer control system of the present invention.

[0041] 1-Control subsystem, 2-Power supply module. Detailed Implementation

[0042] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, but should not be construed as limiting the present invention.

[0043] This invention provides a method for controlling the grabbing and transfer of cutting waste, comprising the following steps:

[0044] S1. After the cutting device finishes cutting the material to obtain the finished product, the conveying mechanism will transport the waste material to the secondary cutting station.

[0045] In this specific implementation, the conveying mechanism relies on an advanced path planning algorithm to accurately calculate the optimal conveying path for the waste material from its initial position to the secondary cutting station. This algorithm comprehensively considers factors such as the layout of the conveying channel, the size and shape of the waste material, and the operating efficiency of the equipment, ensuring that the waste material can be conveyed to the secondary cutting station in the shortest time and with the most stable posture.

[0046] S2. Utilize an array of inductive sensors to detect the edge position of the waste material and dynamically match the laser cutting power based on phase difference analysis of the metal material type.

[0047] In this specific implementation, an inductive sensor array is used to detect the edge position of the waste material, and the laser cutting power is dynamically matched based on phase difference analysis of the metal material type.

[0048] Apply multi-frequency eddy current signals ranging from 0.1 to 10 MHz and collect the phase responses of different metals;

[0049] When the phase difference is between 30-45°, the material is identified as steel, and a high-power cutting mode is selected.

[0050] When the phase difference is between 60-75°, it is determined to be aluminum material, and a low-power cutting mode is matched.

[0051] S3. The visual inspection module scans the defect area on the surface of the waste material, generates an avoidance cutting path, and controls the laser cutting machine to cut the waste material into fragments of a set size.

[0052] In this specific implementation, when dividing waste into fragments of a set size, the cutting process is modeled as a Markov decision process, and the cutting parameters are dynamically adjusted through deep reinforcement learning to maximize cutting efficiency and minimize material waste.

[0053] The algorithm continuously optimizes the control strategy based on real-time feedback to adapt to different materials and cutting conditions.

[0054] Meanwhile, an improved YOLOv5-seg model incorporating an attention mechanism was used to segment the pore and crack areas on the waste surface;

[0055] Construct a 10mm radius buffer zone centered on the defect area;

[0056] The cutting path is replanned based on the topology optimization algorithm to ensure that the fragment size error is ≤2mm.

[0057] Meanwhile, a coaxial infrared thermal imager was used to monitor the temperature gradient in the cutting area;

[0058] When the local temperature exceeds 90% of the material's melting point, the laser power is automatically reduced by 20%.

[0059] After cutting, the consistency of fragment composition was verified by plasma spectroscopy analysis.

[0060] S4. Use a high-definition camera to capture images of the cutting area, use image processing algorithms to identify the location, shape and size of the waste, and transmit this information to the control center;

[0061] In this specific embodiment, a high-definition camera serves as the visual front end, capturing images of the cutting area at a specific frame rate. Its high resolution allows it to capture subtle features of the waste material. The captured image data is transmitted in real-time to the image processing unit via a high-speed interface.

[0062] Within the unit, noise is first eliminated using a deep learning denoising algorithm, and then the improved YOLO algorithm trained on a large number of waste images is used to locate the waste and provide bounding boxes.

[0063] Next, the DeepLab series of semantic segmentation algorithms are used to finely segment the waste region and obtain accurate contours;

[0064] Then, a contour-based shape description algorithm extracts feature parameters and matches them with a preset template to determine the shape;

[0065] By calibrating camera parameters, a mapping between pixels and actual size is established, and the size of waste material is calculated by combining the pixel area.

[0066] Finally, the location, shape, and size information are encrypted and transmitted to the control center via industrial Ethernet for production decision-making.

[0067] S5. Based on the waste information transmitted in the visual recognition and positioning steps, the control center combines the kinematic model of the robotic arm and the grasping force control algorithm to plan and generate the optimal grasping path, optimal grasping point and grasping force of the robotic arm.

[0068] In this specific implementation, after receiving the waste information from visual recognition, the control center calls the kinematic model of the robotic arm and calculates all feasible motion trajectories using forward and inverse kinematics algorithms.

[0069] Then, by using path optimization algorithms, the best crawling path is selected by taking into account efficiency, energy consumption, and collision risk.

[0070] Subsequently, by combining the shape and size of the waste material with the grasping force control algorithm, the optimal grasping point and force are precisely planned, and instructions are generated to direct the robotic arm to grasp it.

[0071] S6. The robotic arm moves to the waste location according to the optimal gripping path generated by the control center, grips the waste according to the determined gripping point and gripping force, and then transfers the waste to the waste box according to the planned transfer path.

[0072] In this specific implementation, when the robotic arm grabs waste fragments, an ultrasonic scanner is used to monitor the volume status of the two waste bins in real time. If the current bin volume is ≥85%, the system switches to the backup bin and triggers an alarm.

[0073] Specifically, piezoelectric ceramic weighing modules are deployed at the bottom of the two waste bins, and ultrasonic scanners are installed at the top;

[0074] The real-time volume is calculated by reconstructing the 3D point cloud. When the volume is ≥85% after 3 consecutive scans, the material frame is switched and the audible and visual alarm is activated.

[0075] During the switching process, the coordinates of the full material box are automatically locked to prevent the waste material gripper from feeding.

[0076] When the robotic arm grabs waste fragments, sensors monitor the movement of the robotic arm and the grabbing of waste in real time. If any abnormality or deviation occurs, the control center will make adjustments immediately.

[0077] If an obstacle is detected, a collision-free path is generated using the fast expanding random tree algorithm;

[0078] Furthermore, based on the real-time posture of the waste fragments, the gradient descent method is used to optimize the gripping point position to ensure gripping stability.

[0079] During the entire process, heterogeneous data collected from different components are synchronized using timestamps and spatial coordinate transformations for spatiotemporal alignment, eliminating sensor delays and installation position deviations.

[0080] Specifically, a dynamic time warping algorithm based on Kalman filtering is used to achieve time synchronization of multi-source data.

[0081] When cutting waste into fragments, an integrated dust concentration sensor and environmental protection module are used.

[0082] When laser cutting generates dust, the negative pressure air volume controller automatically increases the air volume;

[0083] When laser cutting generates explosive dust, the inert gas protection module is triggered.

[0084] In the specific application of the waste material grabbing and transfer control method of this embodiment, after the cutting device completes the cutting of the material to obtain the finished product, the transfer mechanism transports the waste material to the secondary cutting station; the edge position of the waste material is detected by an inductive sensor array, and the laser cutting power is dynamically matched based on the phase difference analysis of the metal material type; the defect area on the surface of the waste material is scanned by a vision detection module to generate an avoidance cutting path, and the laser cutting machine is controlled to cut the waste material into fragments of a set size; images of the cutting area are captured by a high-definition camera, and the position, shape, and size information of the waste material are identified by an image processing algorithm, and this information is transmitted to the control center; the control center performs the visual recognition and positioning steps... The transmitted waste information, combined with the kinematic model of the robotic arm and the grasping force control algorithm, is used to plan and generate the optimal grasping path, optimal grasping point, and grasping force of the robotic arm. The robotic arm moves to the waste location according to the optimal grasping path generated by the control center, and grasps the waste according to the determined grasping point and grasping force. Then, the waste is transferred to the waste box according to the planned transfer path. This method solves the technical problems in the existing technology when handling cutting waste. Due to the large volume of waste, traditional handling and storage equipment is difficult to adapt. This not only requires a lot of manpower for auxiliary operation, increasing labor costs, but also poses safety hazards such as waste slippage and collision during the handling process.

[0085] Beneficial effects:

[0086] In this application, the conveying mechanism adopts an advanced path planning algorithm, which comprehensively considers multiple factors to plan the optimal conveying path for the waste, ensuring that the waste can be conveyed to the secondary cutting station in the shortest time and in the most stable manner, effectively improving the efficiency of the waste conveying process and reducing production waiting time.

[0087] Simultaneously, an array of inductive sensors detects the edge position of the waste material, and phase difference analysis accurately determines the type of metal material, thereby dynamically matching the laser cutting power. This intelligent identification and precise matching method can optimize cutting based on the characteristics of different materials, improving cutting quality and reducing energy waste during the cutting process.

[0088] Furthermore, the cutting process is modeled as a Markov decision process, and cutting parameters are dynamically adjusted by combining deep reinforcement learning. This enables the algorithm to continuously optimize the control strategy based on real-time feedback, thereby maximizing cutting efficiency, minimizing material waste, adapting to different materials and cutting conditions, and improving the intelligence level and economic benefits of the cutting process.

[0089] Secondly, this application employs an improved YOLOv5-seg model with an attention mechanism to segment the surface pores and cracks of the waste material, enabling more accurate detection of surface defects. An avoidance buffer zone is constructed centered on the defect area, and the cutting path is replanned based on a topology optimization algorithm, ensuring that the size error of the segmented fragments is ≤2mm, effectively guaranteeing cutting quality and improving the product yield.

[0090] A high-definition camera captures images of the cutting area. Through a series of advanced image processing algorithms, including noise reduction, localization, segmentation, shape description, and size calculation, the system can comprehensively and accurately identify the location, shape, and size of the waste material. This information is then encrypted and transmitted to the control center, providing precise data support for subsequent robotic arm grasping. Based on the waste material information transmitted from the visual recognition and localization steps, and combined with the robotic arm's kinematic model and grasping force control algorithm, the control center scientifically plans and generates the optimal grasping path, optimal grasping point, and grasping force for the robotic arm. This planning method fully considers the robotic arm's motion characteristics, the shape and size of the waste material, and the stability requirements of grasping, ensuring that the robotic arm can accurately and stably grasp the waste material.

[0091] For heterogeneous data collected from different components, a dynamic time warping algorithm based on Kalman filtering is used to achieve time synchronization. Spatiotemporal alignment is then achieved through timestamp synchronization and spatial coordinate transformation, eliminating sensor delay and installation position deviation. This data alignment method ensures more accurate collaborative work between components, improving the overall system's operational stability and reliability.

[0092] This method automates and intelligently controls the entire process of waste material handling, from conveying, cutting, identifying, grabbing to transfer, eliminating the need for extensive manual labor and effectively reducing labor costs. Simultaneously, it avoids safety hazards such as waste material slippage and collisions that may occur during manual handling, thus improving the safety of the production process.

[0093] Please see Figure 2 , Figure 2 This is a schematic diagram of the cutting waste grabbing and transfer control system of the present invention.

[0094] The present invention also provides a control system for gripping and transferring cutting waste, including a control subsystem 1 for executing the gripping and transferring control method for cutting waste as described above, and a power supply module 2 for supplying power to the control subsystem 1.

[0095] Using the waste material grabbing and transfer control system and method of the present invention, in specific use, after the cutting device completes the cutting of the material to obtain the finished product, the transfer mechanism transports the waste material to the secondary cutting station; the edge position of the waste material is detected by an inductive sensor array, and the laser cutting power is dynamically matched based on the phase difference analysis of the metal material type; the defect area on the surface of the waste material is scanned by a vision detection module to generate an avoidance cutting path, and the laser cutting machine is controlled to divide the waste material into fragments of a set size; images of the cutting area are captured by a high-definition camera, and the position, shape, and size information of the waste material are identified by an image processing algorithm, and this information is transmitted to the control center; the control center positions the waste material according to the visual recognition steps. The waste information transmitted in a short time, combined with the kinematic model of the robotic arm and the grasping force control algorithm, plans and generates the optimal grasping path, optimal grasping point and grasping force of the robotic arm. The robotic arm moves to the waste location according to the optimal grasping path generated by the control center, grasps the waste according to the determined grasping point and grasping force, and then transfers the waste to the waste box according to the planned transfer path. This method solves the technical problems in the existing technology when handling cutting waste. Due to the large volume of waste, traditional handling and storage equipment is difficult to adapt. This not only requires a lot of manpower for auxiliary operation, increasing labor costs, but also easily leads to safety hazards such as waste slippage and collision during the handling process.

[0096] The above description discloses only one preferred embodiment of the present invention, and should not be construed as limiting the scope of the present invention. Those skilled in the art will understand that all or part of the processes of the above embodiments can be implemented, and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for controlling the grabbing and transfer of cutting waste, characterized in that, Includes the following steps: After the cutting device finishes cutting the material to obtain the finished product, the conveying mechanism transports the waste material to the secondary cutting station; The edge position of the waste material is detected by an array of inductive sensors, and the laser cutting power is dynamically matched based on the phase difference analysis of the metal material type. The visual inspection module scans the defective areas on the surface of the waste material, generates an avoidance cutting path, and controls the laser cutting machine to cut the waste material into fragments of a set size. High-definition cameras are used to capture images of the cutting area, and image processing algorithms are used to identify the location, shape, and size of the waste material, which is then transmitted to the control center. Based on the waste information transmitted from the visual recognition and positioning steps, the control center combines the kinematic model of the robotic arm and the grasping force control algorithm to plan and generate the optimal grasping path, optimal grasping point, and grasping force of the robotic arm. The robotic arm moves to the waste location according to the optimal gripping path generated by the control center, grips the waste according to the determined gripping point and gripping force, and then transfers the waste to the waste box according to the planned transfer path.

2. The method for gripping and transferring cutting waste as described in claim 1, characterized in that, When using an array of inductive sensors to detect the edge position of waste material and dynamically matching the laser cutting power based on phase difference analysis of the metal material type: Apply multi-frequency eddy current signals ranging from 0.1 to 10 MHz and collect the phase responses of different metals; When the phase difference is between 30-45°, the material is identified as steel, and a high-power cutting mode is selected. When the phase difference is between 60-75°, it is determined to be aluminum material, and a low-power cutting mode is matched.

3. The method for gripping and transferring cutting waste as described in claim 2, characterized in that, When dividing waste into fragments of a set size, the cutting process is modeled as a Markov decision process, and the cutting parameters are dynamically adjusted through deep reinforcement learning to maximize cutting efficiency and minimize material waste. The algorithm continuously optimizes the control strategy based on real-time feedback to adapt to different materials and cutting conditions.

4. The method for gripping and transferring cutting waste as described in claim 3, characterized in that, In the step "scanning the defect area on the surface of the waste material through the vision inspection module, generating an avoidance cutting path, and controlling the laser cutting machine to cut the waste material into fragments of a set size": An improved YOLOv5-seg model incorporating an attention mechanism was used to segment the pore and crack regions on the waste surface; Construct a 10mm radius buffer zone centered on the defect area; The cutting path is replanned based on the topology optimization algorithm to ensure that the fragment size error is ≤2mm.

5. The method for gripping and transferring cutting waste as described in claim 4, characterized in that, When the robotic arm grabs waste fragments, it uses an ultrasonic scanner to monitor the volume status of the two waste bins in real time. If the current bin volume is ≥85%, it switches to the backup bin and triggers an alarm. Specifically, piezoelectric ceramic weighing modules are deployed at the bottom of the two waste bins, and ultrasonic scanners are installed at the top; The real-time volume is calculated by reconstructing the 3D point cloud. When the volume is ≥85% after 3 consecutive scans, the material frame is switched and the audible and visual alarm is activated. During the switching process, the coordinates of the full material box are automatically locked to prevent the waste material gripper from feeding.

6. The method for gripping and transferring cutting waste as described in claim 5, characterized in that, When the robotic arm grabs waste fragments, sensors monitor the movement of the robotic arm and the grabbing of waste in real time. If any abnormality or deviation occurs, the control center will make adjustments immediately.

7. The method for gripping and transferring cutting waste as described in claim 6, characterized in that, When the robotic arm grabs scrap materials, If an obstacle is detected, a collision-free path is generated using the fast expanding random tree algorithm; Furthermore, based on the real-time posture of the waste fragments, the gradient descent method is used to optimize the gripping point position to ensure gripping stability.

8. The method for gripping and transferring cutting waste as described in claim 7, characterized in that, Throughout the process, heterogeneous data collected from different components are synchronized using timestamps and spatial coordinate transformations for spatiotemporal alignment, eliminating sensor delays and installation position deviations. Specifically, a dynamic time warping algorithm based on Kalman filtering is used to achieve time synchronization of multi-source data.

9. The control system and method for gripping and transferring cutting waste as described in claim 8, characterized in that, In the step "The laser cutting machine divides the waste into fragments of a set size": A coaxial infrared thermal imager was used to monitor the temperature gradient in the cutting area; When the local temperature exceeds 90% of the material's melting point, the laser power is automatically reduced by 20%. After cutting, the consistency of fragment composition was verified by plasma spectroscopy analysis.

10. A control system for gripping and transferring cutting waste, characterized in that, It includes a control subsystem for performing the gripping and transfer control method for cutting waste as described in claim 9, and a power supply module for supplying power to the control subsystem.