Digital twin-driven Delta mechanical arm sorting real-time monitoring system and method

Abstract

The invention provides a real-time monitoring system and method for sorting of a digital twin-driven Delta mechanical arm. According to the system, firstly, a physical experiment platform is built, and a virtual model is built through SolidWorks, 3DMAX and Unity 3D. A communication framework is designed, and wireless communication of the PLC, the sensor, the mechanical arm, the conveying belt and the digital twin system is achieved. Positive and inverse kinematics analysis and Cartesian space trajectory planning are carried out on a Delta robot, and a related algorithm is packaged into an MATLAB function and compiled into a DLL library for C # calling. The system collects joint angle data in real time based on a rotary encoder, transmits the joint angle data to a digital twin system through an OPC UA protocol, and drives a virtual mechanical arm model after resolving, so that high-precision and low-delay virtual-real synchronous pose reconstruction is realized. In addition, the system also develops a collision detection function based on a bounding box algorithm, and performs visual detection by using camera calibration and a YOLO model.

Description

Technical Field

[0001] This invention relates to the technical field of digital twin control and monitoring of Delta robotic arms, and in particular, to a real-time monitoring system and method for sorting Delta robotic arms driven by digital twins. Background Technology

[0002] With the rapid development of industrial automation and intelligent manufacturing, robotics is playing an increasingly important role in modern industrial production. Especially in high-precision, high-speed sorting and handling tasks, parallel robots, particularly Delta robotic arms, are widely used in various production lines due to their unique lightweight design, high rigidity, high speed, and high precision.

[0003] However, traditional Delta robotic arm sorting systems still face many challenges in practical applications. These challenges mainly manifest in the following aspects: Information opacity: In complex production environments, traditional systems often lack comprehensive, real-time awareness of the robotic arm's working status, trajectory, and surrounding environment. This makes it difficult for operators to intuitively understand the robotic arm's real-time position, speed, and potential failure risks, thereby increasing the difficulty of maintenance and management.

[0004] Efficiency and safety issues: In high-speed assembly line operations, improper trajectory planning of the robotic arm or unexpected situations during operation can easily lead to collisions, affecting production efficiency and even causing equipment damage and personnel safety hazards. Traditional offline programming and control methods are ill-suited to dynamically changing production environments and lack real-time adjustment and early warning capabilities.

[0005] Insufficient data utilization: The operational data collected by traditional systems is usually discrete and lacks effective integration and analysis. This makes it impossible to conduct in-depth analysis of the long-term operating status of the robotic arm, hindering predictive maintenance and system optimization, and further restricting the improvement of production efficiency. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a real-time monitoring system and method for sorting using a Delta robotic arm driven by a digital twin, which improves information transparency, efficiency and safety, and optimizes data utilization.

[0007] To solve the above-mentioned technical problems, the technical solution of the present invention is: a real-time monitoring system for sorting using a digital twin-driven Delta robotic arm, comprising: A physical sorting unit, comprising a Delta robotic arm, a controller (PLC) for driving the Delta robotic arm, and a sensor for acquiring angle data of at least one joint of the Delta robotic arm. a digital twin unit running on a computing device and comprising a virtual robot model corresponding to the Delta robot, for visualizing a pose of the virtual robot model in a three-dimensional scene; a data interaction module for establishing a communication connection between the physical sorting unit and the digital twin unit, and transmitting joint angle data collected by the sensors from the physical sorting unit to the digital twin unit in real time; and a kinematics solving module configured in the digital twin unit, for calculating an end pose of the virtual robot model in real time according to the received joint angle data through a forward kinematics algorithm, and driving the virtual robot model to move synchronously with the physical movement of the Delta robot.

[0008] To achieve the above technical solutions, when the system is running, the physical Delta robot moves, and the sensors at the joints of the robot collect angle data streams of the joints in real time. The data interaction module transmits these angle data to the digital twin unit in real time and uninterruptedly. After receiving the data, the kinematics solving module in the digital twin unit immediately calculates the accurate pose of the virtual robot model in the three-dimensional space through a forward kinematics algorithm, and drives the virtual model to move in the visualized scene completely synchronously with the physical robot. This is a one-way, real-time state mapping process from the physical to the virtual. A high-fidelity real-time state monitoring system between the physical robot and its digital twin is established. It solves the problem of non-transparent operation information in the traditional system, so that the operator can intuitively and clearly observe the real-time running trajectory and pose of the physical robot through the three-dimensional visual interface, greatly improving the observability, transparency and accuracy of state perception of the system, and laying a foundation for subsequent optimization control and safety management.

[0009] As a preferred scheme of the present application, a user control interface is further included, and: The kinematics solving module is further configured to calculate target angles of the joints required to drive the Delta robot to reach a target pose through an inverse kinematics algorithm according to target pose coordinates input through the user control interface. The technical scheme is realized as follows: an operator inputs a target task space coordinate through a user control interface of a digital twin unit. A kinematics solving module immediately inversely solves the target coordinate to obtain corresponding joint target angles that a physical Delta robot needs to reach through a kinematics inverse solving algorithm. Then, a data interaction module sends the target angle data to a controller PLC of a physical sorting unit, so that the PLC drives the physical Delta robot to accurately move to a specified target pose. An intuitive and efficient remote operation mode is realized. The operator does not need to write complex robot codes, but only needs to specify a target point in a virtual environment, so that the physical device can complete a corresponding action, thereby greatly reducing the operation and programming threshold of the Delta robot and improving the convenience and work efficiency of human-machine interaction.

[0010] As a preferred scheme of the present application, the kinematics solving module further comprises a trajectory planning function for generating a linear interpolation or circular interpolation path based on the input start and target pose coordinates.

[0011] The technical scheme is realized as follows: when the user inputs the start and target pose coordinates, the kinematics solving module does not directly perform point-to-point motion control, but first generates a series of dense intermediate path points between the two coordinate points through the trajectory planning function, and the path points jointly form a smooth linear or circular interpolation trajectory. Then, the system continuously performs inverse solution calculation and instruction sending along the planned path to drive the physical robot to smoothly move along the preset trajectory. The smoothness, continuity and trajectory accuracy of the robot motion process are significantly improved. Compared with simple point control, the trajectory planning function ensures that the end effector of the robot strictly follows the preset geometric path when performing a task, avoiding impact and vibration caused by starting and stopping and speed change.

[0012] As a preferred scheme of the present application, the technical scheme further comprises: A collision detection module is configured in the digital twin unit and is used for monitoring the distance between the virtual robot model and other virtual objects in real time based on a bounding box algorithm. When the distance is less than a preset safety threshold, the collision detection module generates a braking instruction and sends the braking instruction to the controller of the physical sorting unit through the data interaction module to trigger the Delta robot to brake urgently.

[0013] With the above technical scheme, in the whole process of system operation, the collision detection module works in parallel with the motion monitoring in the digital twin unit. By using the bounding box algorithm, the shortest distance between the virtual robot arm model and the geometric model of other virtual objects in the scene is continuously and real-timely monitored. Once it is detected that the distance is less than the preset safety threshold, it is determined that there is a risk of imminent collision, and the module immediately generates a braking instruction and forcibly sends it to the controller of the physical robot arm through the data interaction module to trigger the physical robot arm to perform emergency braking. By pre-rehearsing the motion risk in the virtual space, the possible collision accident in the physical world is predicted and intervened in real time. The risk can be effectively avoided before the collision actually occurs, greatly enhancing the operation safety and robustness of the system, protecting the equipment from damage and ensuring the stable operation of the production line.

[0014] As a preferred scheme of the present application, the physical sorting unit further comprises a camera and a conveyor belt; the system further comprises: a visual detection module for processing images collected by the camera to identify target objects on the conveyor belt and determine their pixel coordinates; and a coordinate conversion module for converting the pixel coordinates into world coordinates recognizable by the Delta robot arm controller as the target pose for grasping.

[0015] With the above technical scheme, when the target object on the conveyor belt enters the field of view of the camera, the camera takes a picture, and the visual detection module immediately processes the image to identify the target object and calculate its pixel coordinates on the two-dimensional image. Then, the coordinate conversion module calls the preset calibration parameters to accurately convert the pixel coordinates into three-dimensional world coordinates recognizable by the Delta robot arm controller. This world coordinate can then be used as the target pose for the control system to drive the robot arm to grasp. The system is endowed with the ability of environmental perception and target autonomous identification. The original target position acquisition process that needs to be manually specified or pre-programmed is automatically completed by machine vision, thereby realizing automatic positioning of random incoming materials on the conveyor belt. This greatly improves the automation level, intelligent degree and adaptability to dynamic working conditions of the entire sorting system.

[0016] As a preferred scheme of the present application, the data interaction module uses the OPCUA protocol; and the kinematics calculation module contains a dynamic link library DLL compiled after encapsulating the kinematics algorithm.

[0017] The above technical solutions are implemented, in the data interaction process, the data interaction module adopts the standardized industrial communication protocol OPC UA to organize and transmit data. When kinematics is solved, the kinematics solving module does not directly perform calculation in the main program, but completes the calculation by calling a dynamic link library DLL which pre-encapsulates and compiles a complex kinematics algorithm. By adopting specific and efficient technical means, the overall performance and engineering level of the system are improved. The adoption of the OPC UA protocol ensures the real-time performance, reliability and cross-platform compatibility of data interaction. The encapsulation of the algorithm into the DLL realizes the decoupling of the core algorithm and the upper application, facilitates independent maintenance, upgrading and reuse of the algorithm, improves the execution efficiency of the code and the modularization degree of the system, and enhances the stability and maintainability of the system.

[0018] To solve the above technical problems, the application further discloses a digital twin driven Delta mechanical arm sorting real-time monitoring method, comprising the following steps: S1, data acquisition: real-time acquisition of angle data of each joint through a sensor installed on a physical Delta mechanical arm; S2, data transmission: transmitting the angle data from the physical mechanical arm end to the digital twin system end through a data interaction channel; S3, pose reconstruction: in the digital twin system, based on the received angle data, the end pose of the virtual mechanical arm is solved in real time through a kinematics forward solution algorithm; S4, synchronous rendering: driving the virtual mechanical arm model to move in the three-dimensional visualization interface according to the solved end pose, and realizing the action synchronization with the physical Delta mechanical arm.

[0019] The above technical solutions are implemented. First, the real-time joint angle data of the physical mechanical arm is collected through the sensor. Then, the data is transmitted to the digital twin system through the data interaction channel. Then, in the twin system, the angle data is restored to the end pose of the virtual mechanical arm by using the kinematics forward solution algorithm. Finally, according to the pose, the synchronous action of the virtual mechanical arm is rendered in the three-dimensional scene. A standard operation procedure for realizing real-time and high-fidelity mapping of physical equipment to digital twin is provided. Through a series of ordered steps, the information synchronization problem between the physical world and the virtual world is solved, and a visual monitoring platform that can intuitively reflect the state of the real equipment is provided for the operator, thereby effectively improving the transparency and observability of the system.

[0020] As a preferred scheme of the application, the following steps are further included: S5, instruction input and inverse solution: inputting target task coordinates through the user interface of the digital twin system, and calculating corresponding target angles of each joint based on the coordinates through a kinematics inverse solution algorithm; S6, instruction issuing and executing: issuing the joint target angles to the controller of the physical Delta robot to drive the actual movement.

[0021] The above technical solution adds a reverse control process from virtual to physical, which starts from the user inputting target coordinates on the interface of the digital twin system, and then the system performs inverse kinematics to calculate the corresponding joint angles. Finally, these calculated target angles are used as instructions to issue to the controller of the physical robot and drive its actual movement. A complete virtual-real interaction and bidirectional closed-loop control method is constructed. Not only can it monitor, but also can control, turning a passive monitoring system into an active teleoperation system. Users can intuitively control physical entities by operating virtual models, achieving convenient and efficient human-machine interaction and remote control.

[0022] As a preferred embodiment of the present application, it further includes a collision warning step: Real-time monitoring of the relative position of the virtual robot and surrounding objects in the virtual environment; When the distance between the two is less than the preset threshold, an emergency stop instruction is sent to the controller of the physical Delta robot.

[0023] The above technical solution adds a safety warning step, which continuously monitors the relative position of the virtual robot and other object models in the virtual environment. When the distance between the two is less than the safety threshold, indicating that a collision is about to occur, the system immediately triggers the warning mechanism and sends an emergency stop instruction to the controller of the physical robot. By predicting risks in the virtual domain and intervening in the physical domain, this method can brake the physical equipment in time before a potential collision accident occurs, effectively avoiding equipment damage and production interruption, and significantly improving the safety of the entire operation process.

[0024] As a preferred embodiment of the present application, the target task coordinates of the instruction input step are obtained by recognizing the target object on the conveyor belt through a vision system and converting its pixel coordinates to world coordinates.

[0025] To achieve the technical scheme, before the execution of the S5 step, a visual system automatically identifies the target object on the conveying belt and converts its position from pixel coordinates to world coordinates. The world coordinates generated by the visual system will be directly used as the target task coordinates required in the S5 step, thereby starting the subsequent inverse solution, issuing and execution process. The instruction input link that requires human participation is upgraded to a perception link that is automatically completed by machine vision, thereby forming a full-closed-loop automated operation method from "perception" to "decision" to "execution". This enables the system to completely autonomously complete the identification and grabbing of dynamic targets, greatly improving the automation level and intelligence of the system. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a schematic diagram of a physical Delta robot arm; Figure 2 is a schematic diagram of a digital twin Delta robot arm sorting interface; Figure 3 is a schematic diagram of the system communication architecture design; Figure 4 is a schematic diagram of the image processing training model flow; Figure 5 is a schematic diagram of the Delta robot arm sorting pipeline structure flow. DETAILED DESCRIPTION

[0027] The specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings, so that the technical scheme of the present application is easier to understand and master.

[0028] The present application is a digital twin driven Delta robot arm sorting real-time monitoring system and method, which is composed of a physical robot arm, a twin robot arm, a kinematics algorithm, a trajectory planning algorithm, a collision detection algorithm, a visual detection, and a rotary encoder.

[0029] The specific implementation includes the following steps: Step 1: Build a Delta robot arm physical entity experiment platform and perform SolidWorks material-free modeling of the physical experiment platform including Delta robot arm, end effector, conveying belt, and cloth. Then import the model into 3D MAX for model lightweight processing. After lightweight processing, import the model into Unity3D for scene building; Step 1.1, Figure 1 The robot arm in step 1.1 is a Delta robot arm that has been built on an experimental platform, and the material-free modeling in SolidWorks includes Delta robot arm, end effector, conveying belt, and cloth; Step 1.2, import the model into 3DMAX to perform lightweight processing on the model and process redundant surfaces; Step 1.3, export the model file in.fbx format and import it into Unity3D for scene building; Step 1.4, bind and limit the skeleton through parent-child relationship, and combine with script logic to simulate the parallel mechanical arm structure characteristics of Delta mechanical arm. The parent-child relationship in Delta mechanical arm is static platform, active arm, driven arm, moving platform, end effector connection, forming a whole common motion. It also includes conveyor belt and cloth, adding physical properties such as gravity and collision to the cloth and conveyor belt. Then bind the component script to the object, including the mechanical arm script and the script hanging on the cloth. The mechanical arm script uses Unity frame cycle mechanism to execute kinematics calculation in Update(), and receives angle data from OPC UA in real time to make the twin mechanical arm and the physical mechanical arm have the same pose. The script on the cloth makes the cloth move to the specified position through the Rigidbody.MovePostion function.

[0030] Step 1.5, the physical mechanical arm is controlled by PLC, and the connection between PLC and twin system is established through OPC UA communication protocol, with PLC as server and twin system as client.

[0031] Step 2: As shown in Figure 3 , the communication architecture of the system is designed, and PLC, sensor, mechanical arm, conveyor belt and digital twin system are wirelessly communicated; Step 2.1, the perception layer performs data collection and transmission, and obtains mechanical arm joint angle, angular velocity, conveyor belt linear velocity and other working condition parameters through rotary encoder and servo driver; Step 2.2, the connection layer constructs the communication channel of virtual and real systems, realizes cross-platform data transmission by using EtherCAT bus and OPC UA protocol, and provides standardized data interface and communication protocol station; Step 2.3, the data and logic layer undertakes data buffering and analysis function, establishes shared memory buffer to store input data frame, and realizes protocol unpacking and structured conversion through C# analysis module; Step 2.4, the service application layer provides three-dimensional visualization and data service, drives real-time state update of digital twin, realizes global monitoring scene rendering, and synchronously persists data to time series database.

[0032] Step 3: Analysis of forward and inverse kinematics of Delta parallel robot and Cartesian space trajectory planning algorithm.

[0033] Step 3.1, forward kinematics analysis of Delta parallel robot:

[0034] ​ Substitute the above formula, after simplification, the following formula is obtained: Thus the position coordinates of the end effector of the parallel robot in the space are determined.

[0035] Step 3.2, Delta parallel robot inverse kinematics analysis:

[0036] Substitute the above formula, after simplification, the following formula is obtained: And according to the given end effector position coordinates, the rotation angle of each joint can be known.

[0037] Wherein represents the angle between the line connecting each vertex of the static platform and the coordinate origin and the X axis of the coordinate system, , represents the angle between the active arm and the static platform, the length of the active arm is set as , the length of the driven arm is set as , R is the radius of the static platform, and r is the radius of the moving platform.

[0038] Step 3.3, Cartesian space trajectory planning algorithm The position coordinates of each interpolation point in linear interpolation are:

[0039] The position coordinates of each interpolation point in circular interpolation are:

[0040] Step 3.4, In view of the large amount of calculation involved in the forward and inverse kinematics and trajectory planning of the Delta robot, MATLAB has obvious advantages in syntax optimization compared with other languages. Therefore, the above kinematics equations and trajectory planning algorithms are packaged as MATLAB functions and compiled as DLL libraries for C# calling. When integrated in Unity3D, the generated DLL library and MWArray.dll need to be placed in the Assets directory, and the MathWorks.NET array tool library, MATLAB basic tool set and custom DLL program set need to be called in the C# code. Step 3.5, as Figure 2 ​​The Cartesian space straight-line interpolation trajectory planning is taken as a verification case, and a real-time communication link with the Delta robot is established through the EtherCAT industrial bus. In the Unity3D robot motion control interface, the manual input mode is called, and the three-dimensional task space coordinate points are input in turn: (55.41, 0, -371) (55.41, 0, -170) (55.41, 381, -170). The kinematics analysis solves the joint angles according to the given three coordinates of the end effector, the virtual twin generates a continuous path based on the linear interpolation algorithm, and the joint space trajectory is solved in real time through inverse kinematics. The physical Delta robot executes kinematics simulation according to the planned instructions, and realizes closed-loop feedback of the end effector pose data through the OPC UA protocol, and completes the virtual-real synchronous pose reconstruction.

[0041] Step 4, collision detection function development.

[0042] Step 4.1, the bounding box collision detection technology has become the mainstream method of collision detection due to its wide application, mature stability and high computational efficiency. This technology constructs a hierarchical bounding box tree, tightly envelopes the object surface mesh with a convex polyhedron space placeholder, and performs bounding box intersection detection based on the separation axis theorem. In this paper, BoxCollider and Mesh Collider are used in Unity3D to realize the collision detection function of Delta robot. During the simulation process, the real-time collision detection algorithm continuously monitors the relative position of the end effector and the conveyor belt. When the pre-collision state is detected, that is, the bounding box of the end effector and the bounding box of the conveyor belt are below the preset intersection safety threshold, the system sends an emergency stop instruction to the PLC through the OPC UA protocol, triggering the emergency brake of the robot. At the same time, the system interface dynamically displays the collision warning information.

[0043] Step 5, camera calibration and image processing model training.

[0044] Step 5.1, camera calibration. This paper uses the eight-point calibration method to convert pixel coordinates to robot coordinates. The "eye outside hand" configuration is adopted, and during the setting of the experimental device, the camera is located outside the robot hand. The main formula of the eight-point calibration method is as follows.

[0045]

[0046] In this context, represents the robot coordinates, represents the rotation coefficient, refers to the pixel coordinates, represents the displacement coefficient. As can be seen from the transformation formula, there are a total of six unknown parameters. Therefore, six coordinate points are sufficient to completely solve the transformation equation. However, in order to minimize errors, this paper selects 8 points for camera calibration; Step 5,2, to ensure that the camera can identify the object, the image needs to be processed and the model needs to be trained. First, simulate the target center-symmetric morphology, train the model to identify the symmetric structure. Next, let the model learn the features of the left and right flipped target, enhance the generalization ability. After that, process the monitoring of the upside-down target, cover more actual scenes. Finally, test the robustness under noise interference. Finally, randomly place the cloth to simulate the real pipeline detection, use the coordinate conversion formula of the last step to convert the pixel coordinates to robot coordinates, these coordinates are transmitted to the digital twin system and PLC system at the same time.

[0047] Step 6, based on the digital twin Delta mechanical arm sorting pipeline process analysis.

[0048] Step 6.1, target coordinate generation, the vision system detects the object output (x, y, z) coordinates in the world coordinate system, and the coordinate data is directly transmitted to the mechanical arm controller. The actual position is sent to the digital twin system at the same time; Step 6.2, the mechanical arm controller performs local inverse solution calculation to convert (x, y, z) to the pulse number required by the joint motor, and sends pulse instructions to the physical mechanical arm to drive the motor to move. The encoder of the physical mechanical arm feeds back the actual pulse number to the controller in real time. The controller checks the position error and triggers the arrival position signal; Step 6.3, the virtual mechanical arm in the digital twin system drives the virtual mechanical arm to simulate the movement of the physical mechanical arm through kinematics forward solution calculation, which converts the joint angle to the twin system world coordinate to drive the virtual mechanical arm. The virtual conveyor belt relies on delay control to transport the cloth.

[0049] Of course, the above is only a typical example of the present application, in addition to this, the present application can have other various specific embodiments, any technical solution formed by equivalent replacement or equivalent transformation falls within the scope of the present application.

Claims

1. A real-time monitoring system for sorting a Delta robotic arm driven by a digital twin, characterized in that, include: A physical sorting unit, comprising a Delta robotic arm, a controller (PLC) for driving the Delta robotic arm, and a sensor for acquiring angle data of at least one joint of the Delta robotic arm. A digital twin unit, which runs on a computing device and includes a virtual robotic arm model corresponding to the Delta robotic arm, is used to visualize the pose of the virtual robotic arm model in a three-dimensional scene. The data interaction module is used to establish a communication connection between the physical sorting unit and the digital twin unit, and to transmit the joint angle data collected by the sensor from the physical sorting unit to the digital twin unit in real time. as well as The kinematics calculation module, configured in the digital twin unit, is used to calculate the end pose of the virtual robotic arm model in real time based on the received joint angle data using a forward kinematics algorithm, and to drive the virtual robotic arm model to synchronize its physical movement with that of the Delta robotic arm.

2. The real-time monitoring system for Delta robotic arm sorting driven by digital twin according to claim 1, characterized in that, It also includes a user control interface, and: The kinematics calculation module is also used to: calculate the target angles of each joint required to drive the Delta robotic arm to reach the target pose based on the target pose coordinates input through the user control interface and the inverse kinematics algorithm. The data interaction module is also used to transmit the target angles of each joint to the controller of the physical sorting unit to drive the Delta robotic arm to move to the target pose.

3. The real-time monitoring system for Delta robotic arm sorting driven by digital twin according to claim 2, characterized in that, The kinematics calculation module also includes a trajectory planning function, which generates linear interpolation or circular interpolation paths based on the input initial and target pose coordinates.

4. The real-time monitoring system for Delta robotic arm sorting driven by digital twin according to claim 1, characterized in that, Also includes: A collision detection module, configured in the digital twin unit, is used to monitor the distance between the virtual robotic arm model and other virtual objects in real time based on the bounding box algorithm; When the distance is less than a preset safety threshold, the collision detection module generates a braking command and sends it to the controller of the physical sorting unit through the data interaction module to trigger the Delta robotic arm to brake urgently.

5. The real-time monitoring system for Delta robotic arm sorting driven by digital twin according to claim 1, characterized in that, Also includes: The physical sorting unit also includes a camera and a conveyor belt; The system also includes: A visual detection module is used to process images captured by the camera to identify targets on the conveyor belt and determine their pixel coordinates; as well as The coordinate transformation module is used to convert the pixel coordinates into world coordinates that can be recognized by the Delta robotic arm controller, as the target pose for grasping.

6. The real-time monitoring system for Delta robotic arm sorting driven by digital twin according to claim 1, characterized in that, The data interaction module adopts the OPCUA protocol; the kinematics calculation module includes a dynamic link library (DLL) that encapsulates and compiles the kinematics algorithm.

7. A real-time monitoring method for sorting using a Delta robotic arm driven by a digital twin, characterized in that, The Delta robotic arm sorting real-time monitoring system, driven by a digital twin as described in any one of claims 1-6, includes the following steps: S1. Data Acquisition: The angle data of each joint is collected in real time through sensors installed on the physical Delta robotic arm; S2. Data transmission: The angle data is sent from the physical robotic arm to the digital twin system via a data interaction channel; S3. Pose Reconstruction: In the digital twin system, based on the received angle data, the end pose of the virtual robotic arm is calculated in real time using a forward kinematics algorithm. S4. Synchronous Rendering: Based on the calculated end-effector pose, drive the virtual robotic arm model to move in the 3D visualization interface to achieve synchronization with the physical Delta robotic arm's movements.

8. The method for real-time monitoring of Delta robotic arm sorting driven by digital twin according to claim 7, characterized in that, It also includes the following steps: S5. Command Input and Inverse Kinematics Solution: Input the target task coordinates through the user interface of the digital twin system, and calculate the corresponding target angles of each joint based on the coordinates using the inverse kinematics algorithm. S6. Command Issuance and Execution: The target angles of each joint are issued to the controller of the physical Delta robotic arm to drive its actual movement.

9. A real-time monitoring method for sorting using a Delta robotic arm driven by a digital twin according to claim 7, characterized in that, It also includes collision warning steps: Real-time monitoring of the relative position of the virtual robotic arm to surrounding objects in a virtual environment; When the distance between the two is detected to be less than a preset threshold, an emergency stop command is sent to the controller of the physical Delta robotic arm.

10. A real-time monitoring method for sorting using a Delta robotic arm driven by a digital twin, as described in claim 8, is characterized in that... The target task coordinates of the instruction input step are obtained by identifying the target object on the conveyor belt through a vision system and converting its pixel coordinates into world coordinates.

Images