A mover identification method and device based on magnetic suspension flexible assembly line

Abstract

The application relates to a mover identification method and device based on a magnetic suspension flexible flow line, wherein the method comprises the following steps: activating a pre-generated topological model; periodically scanning the data of a mover encoder of a station driver to obtain real-time position and speed data of the mover and mapping the data to corresponding coordinates; performing noise filtering and smoothing processing on the data; selecting a corresponding identification algorithm according to the size of a line body, the bus communication frequency, the preset task sequence of the mover and the processed speed data; performing correlation analysis on the processed data to generate mover trajectory data; checking the trajectory data, and giving a warning if the trajectory data is abnormal; updating the data that passes the check to a database, generating a control instruction set based on the topological model coordinates, the motion state of the mover and external instruction parameters, and outputting the control instruction set to a path planning system to realize multi-mover high-precision real-time identification and dynamic adaptive control and provide support for asynchronous scheduling.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to a method and device for identifying moving parts based on a magnetically levitated flexible production line. Background Technology

[0002] With the increasing market demand for product customization and personalization, small-batch, diversified production models are gradually becoming mainstream. To cope with this trend, companies need to frequently switch product types during production. However, in traditional conveyor line systems, production line design typically requires customized development based on specific processing techniques. Each time product types change, the production line needs to be redesigned and adjusted, which not only increases the cost of production line upgrades but also extends production preparation time. In contrast, flexible conveyor systems employ a modular and structured design approach, offering extremely high reusability. Through flexible conveyor systems, companies can retain existing intelligent equipment and only need to change the operating logic, thereby significantly reducing the cost and time of production line adjustments.

[0003] In magnetic levitation flexible transport systems, the asynchronous movement of multiple movers is one of its core characteristics. The realization of asynchronous movement relies on the accurate identification and control of multiple movers. Therefore, mover identification technology has become one of the key technologies for magnetic levitation flexible transport systems. However, existing mover identification methods often struggle to balance identification accuracy and real-time performance when faced with complex line shapes, high-speed moving movers, and diverse application scenarios, leading to numerous challenges in practical applications. Summary of the Invention

[0004] The main objective of this invention is to provide a method and apparatus for identifying movers in a magnetically levitated flexible production line. By solving the problem of high-precision identification and dynamic adaptation control in the asynchronous operation scenario of multiple movers in a magnetically levitated flexible production line, this invention achieves high-precision real-time identification and dynamic adaptation control of multiple movers in complex production environments through intelligent algorithm selection, spatiotemporal correlation analysis, and multi-level verification mechanisms. This provides highly flexible and reliable technical support for asynchronous scheduling in complex production scenarios.

[0005] To achieve the above objectives, this invention provides a method for identifying moving parts based on a magnetically levitated flexible pipeline, comprising the following steps:

[0006] Activate the pre-generated topology model, which is virtualized based on the configuration file of the pipeline controller. The configuration file includes line size data, bus communication frequency data, and preset task sequence data for each mover.

[0007] The encoder data of each slave driver is periodically scanned to obtain the real-time position and velocity data of each mover, and the position data of each mover is mapped to the corresponding coordinates of the topology model.

[0008] The real-time position and velocity data of the mover are subjected to noise filtering and smoothing.

[0009] Based on the line size data, bus communication frequency data, preset task sequence data of each mover, and processed mover speed data, a corresponding recognition algorithm is selected.

[0010] Based on the selected recognition algorithm, the processed real-time position and velocity data of the mover are correlated and analyzed to generate the mover trajectory data;

[0011] The trajectory data of the moving part is verified, and an early warning is triggered if the verification is abnormal.

[0012] The verified trajectory data of the mover is updated to the real-time status database of the mover. Based on the coordinates of the topology model and the motion state of the mover, and combined with external command parameters, a set of control commands is generated and output to the path planning system.

[0013] Furthermore, prior to the step of activating the pre-generated topology model, the method further includes:

[0014] Parse the line size data in the configuration file of the production line controller;

[0015] Select the corresponding topology template based on the line size data. The topology template includes a ring template, a rectangular template, and a custom polygon template.

[0016] Based on the selected topology template, the motion constraint data of each mover is loaded to generate a digital line body topology model including normalized virtual coordinates.

[0017] Furthermore, the step of periodically scanning the mover encoder data of each slave driver includes:

[0018] Periodic data polling is initiated on each slave driver to obtain the real-time position and speed data of each mover encoder;

[0019] Add a timestamp to the data corresponding to each actuator, and synchronize the data of different actuators in time.

[0020] Based on the synchronized mover velocity data, the real-time positions of each mover are mapped onto the normalized virtual coordinates of the topological model.

[0021] Furthermore, the step of performing noise filtering and smoothing processing on the real-time position and velocity data of the mover includes:

[0022] The window length for data processing is dynamically adjusted based on the real-time velocity data of each mover. The higher the mover speed, the smaller the window, in order to meet the needs of capturing trajectory details of high-speed movers.

[0023] The position and velocity data within the window are processed by weighted averaging, with the weight decreasing as the data becomes more timely, to suppress noise interference and enhance trajectory continuity.

[0024] Furthermore, the step of selecting the corresponding recognition algorithm includes:

[0025] When the system computing power reaches the preset sufficient computing power threshold after evaluation, and the current topology is not the same as the ring topology, the improved Hungarian algorithm is selected. The improved Hungarian algorithm can be adapted to any topology. Once the current topology is detected to be different from the ring topology, the improved Hungarian algorithm must be selected.

[0026] When the bus communication frequency is monitored to be lower than the preset frequency threshold, or the hardware stability is lower than the preset stability threshold and multiple verifications are required based on system operation requirements, the frame matching algorithm is selected.

[0027] When the system's computing power is lower than the preset computing power standard after evaluation, and the corresponding topology is determined to be a simple structure homeomorphic to a ring, the sequence tracking algorithm is selected.

[0028] Furthermore, the step of performing correlation analysis on the processed real-time position and velocity data of the mover according to the selected recognition algorithm includes:

[0029] If the selected recognition algorithm is the improved Hungarian algorithm, the matching priority between the mover and the topological coordinate nodes is optimized based on the real-time position change of the mover, so as to resolve the allocation conflict when multiple mover trajectories intersect.

[0030] If the frame matching algorithm is selected, the displacement features of the movers in adjacent data frames are extracted to enhance the accuracy of continuous matching of mover positions across frames;

[0031] If a sequence tracking algorithm is selected, the logical order of the mover's operation is determined based on the mover's current position, velocity, and preset task sequence.

[0032] Furthermore, the step of verifying the trajectory data of the moving part includes:

[0033] Verify the integrity of the data frame; if the checksum deviation exceeds the allowable value, mark it as an abnormal frame.

[0034] Verify the continuity of the moving part's trajectory; if the acceleration exceeds the allowable range, it is determined to be a trajectory jump.

[0035] The system checks whether the arrangement of moving parts conforms to preset rules. If the minimum safety distance is violated, an alert is triggered.

[0036] Furthermore, after the step of verifying the anomaly, the method further includes:

[0037] For abnormal frames, a retransmission request is made. If the retransmission fails, the corresponding active element is marked as offline.

[0038] Backtracking analysis of the motioner initiation trajectory for trajectory jumps, and collision verification with the trajectories of neighboring motioners;

[0039] Based on the degree of deviation, tiered warnings are triggered, including recording anomalies, sending audible and visual alarms, or freezing the device.

[0040] Furthermore, the step of generating the control instruction set includes:

[0041] Each mover is assigned an independent data version number, and an optimistic locking mechanism is used to resolve concurrent write conflicts.

[0042] Based on the motion state of the mover, topological coordinates, and external command parameters, the predicted path for the next N cycles is calculated.

[0043] A dynamic control instruction set is generated based on the predicted path. The instruction set includes braking instructions, speed adjustment instructions, and path replanning instructions.

[0044] The present invention also provides a moving part identification device based on a magnetically levitated flexible assembly line, comprising:

[0045] Topology building blocks are used to activate pre-generated topology models;

[0046] The data mapping module is used to periodically scan the encoder data of each slave driver, obtain the real-time position and velocity data of each mover, and map the position data of each mover to the corresponding coordinates of the topology model.

[0047] The data processing module is used to perform noise filtering and smoothing on the real-time position and velocity data of the mover;

[0048] The algorithm matching module is used to select the corresponding recognition algorithm based on the line size data, bus communication frequency data, preset task sequence data of each mover, and processed mover velocity data. ；

[0049] The data processing module is used to perform correlation analysis on the processed real-time position and velocity data of the mover according to the selected recognition algorithm, and generate the mover trajectory data;

[0050] The data verification module is used to verify the trajectory data of the moving part; if the verification is abnormal, an early warning will be triggered.

[0051] The data application module is used to update the verified trajectory data of the mover to the real-time status database of the mover, and generate a set of control instructions based on the coordinates of the topology model and the motion state of the mover, combined with external instruction parameters, and output them to the path planning system.

[0052] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described moving part recognition method based on a magnetically levitated flexible pipeline.

[0053] The present invention provides a mover identification method and device based on a magnetically levitated flexible production line, which has the following beneficial effects: By constructing a topological model, the present invention can accurately map the mover position, providing precise coordinate references for identification, significantly improving the accuracy of mover positioning in complex production line environments and reducing assembly error rates. In terms of data acquisition, the mover encoder data is periodically scanned and synchronized, while data processing is dynamically optimized based on the mover's real-time speed, improving data quality and processing efficiency, saving time, and contributing to the efficient operation of the production line. Furthermore, the algorithm selection can be flexibly adapted to various data features; compared to a single algorithm, the present invention can achieve efficient mover identification and trajectory generation under different operating conditions. Subsequently, for abnormal situations, multi-dimensional verification can promptly detect and handle issues such as trajectory jumps and excessively small spacing, ensuring the safety and stability of the production line and reducing downtime losses. In the system control stage, the verified trajectory data is updated to the database, and various mechanisms are used to generate precise control commands, improving production smoothness and coordination, ultimately improving production efficiency and product quality. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating a moving part identification method based on a magnetically levitated flexible pipeline in one embodiment of the present invention.

[0055] Figure 2 This is a structural block diagram of a moving part identification device based on a magnetically levitated flexible production line in one embodiment of the present invention.

[0056] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0058] Reference Figure 1 The diagram below illustrates a process for identifying moving parts based on a magnetically levitated flexible pipeline, as proposed in this invention. The process includes the following steps:

[0059] S1, activate the pre-generated topology model;

[0060] S2, periodically scan the encoder data of each slave driver to obtain the real-time position and velocity data of each mover, and map the position data of each mover to the corresponding coordinates of the topology model;

[0061] S3, perform noise filtering and smoothing on the real-time position and velocity data of the moving part. ；

[0062] S4. Based on the line size data, bus communication frequency data, preset task sequence data of each mover, and processed mover speed data, select the corresponding recognition algorithm.

[0063] S5, perform correlation analysis on the processed real-time position and velocity data of the mover according to the selected recognition algorithm to generate the mover trajectory data;

[0064] S6, verify the trajectory data of the moving part; if the verification is abnormal, trigger an early warning.

[0065] S7 updates the verified trajectory data of the mover to the real-time status database of the mover, and generates a set of control instructions based on the coordinates of the topology model and the motion state of the mover, combined with external command parameters, and outputs it to the path planning system.

[0066] As described in step S1 above, a topology model needs to be constructed before implementing the mover identification method. First, the line size data in the pipeline controller configuration file is parsed. Based on this data, a suitable topology template is selected from the ring template, rectangular template, and custom polygon template. Then, the motion constraint data of each mover is loaded to generate a digital line topology model containing normalized virtual coordinates. This model is the basic framework of the entire mover identification process. Activating the line topology model provides a unified coordinate system and structural reference for subsequent steps, facilitating the association of the actual operating data of the movers with it.

[0067] As described in step S2 above, by periodically polling each slave driver, the real-time position and velocity data of each mover encoder can be obtained. To ensure data accuracy and traceability, a timestamp is added to the data corresponding to each mover, and the data from different drivers are synchronized in time. After synchronization, based on the synchronized mover velocity data, the real-time position of each mover is mapped to the normalized virtual coordinates of the topology model activated in step S1. This step realizes the integration of the actual operating state data of the mover with the model coordinate system, enabling subsequent processing and analysis of the mover data to be based on a unified coordinate reference, laying the foundation for accurate identification of the mover trajectory.

[0068] As described in step S3 above, the window length for data processing is dynamically adjusted based on the real-time velocity data of each mover; the higher the velocity, the smaller the window, thus adapting to the need for high-speed movers to capture trajectory details. Next, a weighted average is applied to the position and velocity data within the window, with the weight decreasing as the data becomes more timely, thereby suppressing noise interference and enhancing trajectory continuity. This processing removes noise interference from the collected data, making the mover's position and velocity data smoother and more accurate, providing a reliable data foundation for the accurate application of subsequent recognition algorithms.

[0069] As described in step S4 above, when the system computing power reaches a preset sufficient computing power threshold after evaluation, and the current topology is not a homoeuphemism of the ring topology, the improved Hungarian algorithm is selected. The improved Hungarian algorithm can adapt to any topology. The homoeuphemism of the ring topology means that the current topology is structurally different from the ring topology and cannot be transformed into a ring topology through continuous deformation. Once the detection of a homoeuphemism between the current topology and the ring topology is met, the improved Hungarian algorithm must be selected to achieve efficient computation under the conditions of system computing power and topology. When the bus communication frequency is monitored and falls below a preset frequency threshold, or the hardware stability is lower than a preset stability threshold and multiple verifications are required based on system operation needs, the frame matching algorithm is selected to ensure stable system operation under communication or hardware anomalies. When the system computing power is evaluated and falls below a preset computing power standard, and the corresponding topology is determined to be a simple structure homoeuphemism of the ring (i.e., meeting conditions such as few structural components and simple connection relationships), the sequence tracking algorithm is selected to effectively process simple topologies under limited computing power.

[0070] As described in step S5 above, if the selected recognition algorithm is the improved Hungarian algorithm, the matching priority between the mover and the topological coordinate nodes is optimized based on the real-time position changes of the mover, resolving the allocation conflict when multiple mover trajectories intersect, thereby accurately associating the mover position data to generate the trajectory; if the frame matching algorithm is selected, the displacement features of the mover in adjacent data frames are extracted to enhance the accuracy of continuous matching of mover positions across frames, and the mover trajectory is obtained by processing multiple frames of data; if the sequence tracking algorithm is selected, the logical order of the mover's operation is determined based on the mover's current position, velocity, and preset task sequence, thereby generating coherent mover trajectory data. Through step S5, the processed mover data is subjected to in-depth analysis according to the selected algorithm to generate trajectory data reflecting the mover's motion path.

[0071] As described in step S6 above, the integrity of the mover trajectory data is first checked. If the checksum deviation exceeds the allowable value, it is marked as an abnormal frame. Then, the continuity of the mover trajectory is checked. If the acceleration exceeds the allowable range, it is determined to be a trajectory jump. Finally, the arrangement order of the movers is checked to see if it conforms to the preset rules. If the minimum safety distance is violated, an early warning is triggered. Through multi-dimensional data verification, errors or anomalies that may exist in the mover trajectory data can be detected in a timely manner, ensuring the reliability of mover operation status monitoring.

[0072] As described in step S7 above, when updating the database with verified data, an independent data version number is assigned to each mover, and an optimistic locking mechanism is used to resolve concurrent write conflicts. Based on the mover's motion state, topological coordinates, and external command parameters, a predicted path for the next N cycles is calculated; a dynamic control command set is generated based on the predicted path, including braking commands, speed adjustment commands, and path replanning commands. If a verification error occurs, a retransmission request is executed for the abnormal frame; if the retransmission fails, the corresponding mover is marked as offline; trajectory backtracking analysis is initiated for movers with trajectory jumps, and collision verification is performed with the trajectories of neighboring movers; graded early warnings are triggered based on the degree of deviation, including recording the abnormality, sending audible and visual alarms, or freezing the mover. This step enables the accurate use of mover trajectory data to update the database, providing real-time status information to the system, and simultaneously generating a control command set based on the mover state, outputting it to the path planning system to achieve precise control of the magnetic levitation flexible assembly line.

[0073] In a specific embodiment, in an electronics manufacturing workshop, 15 movers are distributed on a 60-meter-circumference circular magnetic levitation assembly line for transporting various electronic components. The configuration file of the assembly line controller shows a circular layout with a radius of 9.55m, a bus communication frequency of 1kHz, and a preset task sequence for the movers using a cyclic delivery mode. By monitoring the CPU utilization (92%) and memory utilization (85%) of the control server, the current system computing power is determined to be 48%, significantly lower than the preset sufficient computing power threshold (80%). A morphological comparison is performed between the circular layout and a standard circular topology, confirming a continuous bidirectional mapping relationship (i.e., homeomorphism). A circular template is selected from the topology template library, and mover motion constraints (maximum speed 1.5m / s, centripetal acceleration limit 0.6m / s²) are applied to generate a digital topology model with the 3 o'clock position (9.55, 0) as the origin.

[0074] After completing the model construction, the topology model is activated. The encoder data of each slave driver is scanned at 1-millisecond intervals to obtain the real-time position and velocity data of each mover. For example, the driver containing mover C3 feeds back its encoder data, showing a real-time position of (4.775000, 8.270000) and a velocity of 1.2 m / s; the position of mover D7 is (-9.549000, 0.001000), and the velocity is 0.8 m / s, etc. Each mover data includes micrometer-level precision coordinates; for example, the position data of mover E2 is accurate to (6.752000, 6.752000). Precise IEEE 1588 timestamps are added to all data, and the timing deviation between different drivers is controlled within ±50 microseconds using the PTP protocol. Based on the synchronized data, the positions of each mover are accurately mapped to the normalized virtual coordinates of the topology model.

[0075] Noise is filtered and smoothed from the real-time position and velocity data of the mover. Taking mover F5 as an example, with a velocity of 1.0 m / s, the system automatically selects a 7-point sliding window for processing. The position data sequence within the current window is (8.270100, 4.774900), (8.270095, 4.774910), (8.270088, 4.774915), (8.270082, 4.774920), (8.270078, 4.774925), (8.270073, 4.774930), (8.270070, 4.774935). A weighted average is calculated using exponential decay weights, with the latest data having a weight of 0.35. The final smoothed position is (8.270085, 4.774920).

[0076] Given the system's insufficient computing power (48%) and topological homeomorphism, a sequence tracking algorithm was automatically selected as the primary identification method. During operation, the algorithm only predicts the trajectory for the next 200ms (originally designed for 500ms), narrowing the detection range to three adjacent movers (originally global detection). When the distance between mover G4 and its adjacent mover H8 is less than the safe value of 0.5 meters for five consecutive cycles, the system automatically switches to a frame matching algorithm for cross-frame continuity verification. SIFT feature point matching is used to ensure trajectory consistency, achieving a matching accuracy of ±10 micrometers.

[0077] During trajectory verification, if the position difference between two consecutive frames of mover J3 reaches 0.000125 meters (125 micrometers), exceeding the allowable threshold of 95 micrometers, it is immediately marked as a trajectory jump anomaly. If the same mover triggers three jump warnings within 1 second, an automatic isolation procedure is initiated, marking the mover as offline and sending a maintenance request.

[0078] When the mover K9 approaches the annular connection point (coordinates (9.55,0) ± 0.1 meters), the graded protection mechanism is triggered sequentially: first, the speed is limited to below 0.3 meters per second; if it continues to move towards the boundary, an emergency stop command (EtherCAT command 0x6040 = 0x000B) is sent within 10 ms; when the final position exceeds (9.55 ± 0.005) meters, the drive enable is immediately cut off and the boundary fault code 0x8F01 is reported.

[0079] The improved Hungarian algorithm is only used for offline trajectory analysis when engineers manually activate the diagnostic mode. During online operation, the system strictly adopts a working mode that primarily uses sequence tracking algorithms and secondarily uses frame matching algorithms. All mover control commands are version-managed through an optimistic locking mechanism to ensure stable system operation even under 45% computing power conditions.

[0080] Reference Figure 2 Here is a structural block diagram of a moving part identification device based on a magnetically levitated flexible assembly line according to an embodiment of the present invention, comprising:

[0081] Topology building blocks are used to activate pre-generated topology models;

[0082] The data mapping module is used to periodically scan the encoder data of each slave driver, obtain the real-time position and velocity data of each mover, and map the position data of each mover to the corresponding coordinates of the topology model.

[0083] The data processing module is used to perform noise filtering and smoothing on the real-time position and velocity data of the mover;

[0084] The algorithm matching module is used to select the corresponding recognition algorithm based on the line size data, bus communication frequency data, preset task sequence data of each mover, and processed mover velocity data. ；

[0085] The data processing module is used to perform correlation analysis on the processed real-time position and velocity data of the mover according to the selected recognition algorithm, and generate the mover trajectory data;

[0086] The data verification module is used to verify the trajectory data of the moving part; if the verification is abnormal, an early warning will be triggered.

[0087] The data application module is used to update the verified trajectory data of the mover to the real-time status database of the mover, and generate a set of control instructions based on the coordinates of the topology model and the motion state of the mover, combined with external instruction parameters, and output them to the path planning system.

[0088] For the specific implementation of each module in the above device example, please refer to the above method embodiments, which will not be repeated here.

[0089] In summary, this invention activates a pre-generated topology model, which is virtualized based on the configuration file of the pipeline controller. This configuration file includes pipeline size data, bus communication frequency data, and preset task sequence data for each mover. It periodically scans the encoder data of each slave driver to obtain the real-time position and velocity data of each mover, and maps the position data of each mover to the corresponding coordinates in the topology model. Noise filtering and smoothing are performed on the real-time position and velocity data of the movers. Based on the pipeline size data, bus communication frequency data, preset task sequence data of each mover, and the processed data, the invention... The system uses subvessel velocity data and selects the corresponding recognition algorithm. Based on the selected algorithm, it performs correlation analysis on the processed real-time position and velocity data of the moving parts to generate moving part trajectory data. The moving part trajectory data is then verified; if an anomaly is detected, an early warning is triggered. The verified moving part trajectory data is updated to the moving part real-time status database. Based on the coordinates of the topology model and the moving part's motion state, and combined with external command parameters, a control command set is generated and output to the path planning system. This achieves high-precision real-time identification and dynamic adaptive control of multiple moving parts in complex production environments, providing highly flexible and reliable technical support for asynchronous scheduling in complex production scenarios.

[0090] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0091] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0092] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for identifying moving parts based on a magnetically levitated flexible assembly line, characterized in that, Includes the following steps: Activate the pre-generated topology model, which is virtualized based on the configuration file of the pipeline controller. The configuration file includes line size data, bus communication frequency data, and preset task sequence data for each mover. Periodically scan the mover encoder data of each slave driver, including periodically polling each slave driver to obtain the real-time position and speed data of each mover encoder; Add a timestamp to the data corresponding to each actuator, and synchronize the data of different actuators in time. Based on the synchronized mover velocity data, the real-time position of each mover is mapped to the normalized virtual coordinates of the topology model; the real-time position and velocity data of each mover are obtained, and the position data of each mover is mapped to the corresponding coordinates of the topology model; The real-time position and velocity data of the mover are subjected to noise filtering and smoothing. Based on the line size data, bus communication frequency data, preset task sequence data of each mover, and processed mover velocity data, a corresponding recognition algorithm is selected. This includes selecting the improved Hungarian algorithm when the system computing power reaches a preset sufficient computing power threshold after evaluation, and the current topology is not homomorphic to the ring topology (meaning the current topology cannot be transformed into a ring topology through continuous deformation). The improved Hungarian algorithm can adapt to any topology. When the bus communication frequency is monitored to be lower than a preset frequency threshold, or the hardware stability is lower than a preset stability threshold and multiple verifications are required based on system operation needs, a frame matching algorithm is selected. When the system computing power is evaluated to be lower than a preset computing power standard, and the corresponding topology is analyzed and determined to be a simple structure homomorphic to the ring, a sequence tracking algorithm is selected. Based on the selected recognition algorithm, the processed real-time position and velocity data of the mover are correlated and analyzed to generate the mover trajectory data; The trajectory data of the moving part is verified, and an early warning is triggered if the verification is abnormal. The verified trajectory data of the mover is updated to the real-time status database of the mover. Based on the coordinates of the topology model and the motion state of the mover, and combined with external command parameters, a control command set is generated and output to the path planning system.

2. The method for identifying moving parts based on a magnetically levitated flexible assembly line according to claim 1, characterized in that, Before the step of activating the pre-generated topology model, the following are also included: Parse the line size data in the configuration file of the production line controller; Select the corresponding topology template based on the line size data. The topology template includes a ring template, a rectangular template, and a custom polygon template. Based on the selected topology template, the motion constraint data of each mover is loaded to generate a digital line body topology model including normalized virtual coordinates.

3. The method for identifying moving parts based on a magnetically levitated flexible assembly line according to claim 1, characterized in that, The step of performing noise filtering and smoothing on the real-time position and velocity data of the mover includes: The window length for data processing is dynamically adjusted based on the real-time velocity data of each mover. The higher the mover speed, the smaller the window, in order to meet the needs of capturing trajectory details of high-speed movers. The position and velocity data within the window are processed by weighted averaging, with the weight decreasing as the data becomes more timely, to suppress noise interference and enhance trajectory continuity.

4. The method for identifying moving parts based on a magnetically levitated flexible assembly line according to claim 1, characterized in that, The step of performing correlation analysis on the processed real-time position and velocity data of the mover according to the selected recognition algorithm includes: If the selected recognition algorithm is the improved Hungarian algorithm, the matching priority between the mover and the topological coordinate nodes is optimized based on the real-time position change of the mover, so as to resolve the allocation conflict when multiple mover trajectories intersect. If the frame matching algorithm is selected, the displacement features of the movers in adjacent data frames are extracted to enhance the accuracy of continuous matching of mover positions across frames; If a sequence tracking algorithm is selected, the logical order of the mover's operation is determined based on the mover's current position, velocity, and preset task sequence.

5. The method for identifying moving parts based on a magnetically levitated flexible assembly line according to claim 1, characterized in that, The step of verifying the trajectory data of the moving part includes: Verify the integrity of the data frame; if the checksum deviation exceeds the allowable value, mark it as an abnormal frame. Verify the continuity of the moving part's trajectory; if the acceleration exceeds the allowable range, it is determined to be a trajectory jump. The system checks whether the arrangement of moving parts conforms to preset rules. If the minimum safety distance is violated, an alert is triggered.

6. The method for identifying moving parts based on a magnetically levitated flexible assembly line according to claim 1, characterized in that, Following the step of verifying the anomaly, the following steps are also included: For abnormal frames, a retransmission request is made. If the retransmission fails, the corresponding active element is marked as offline. Backtracking analysis of the motioner initiation trajectory for trajectory jumps, and collision verification with the trajectories of neighboring motioners; Based on the degree of deviation, tiered warnings are triggered, including recording anomalies, sending audible and visual alarms, or freezing the device.

7. The method for identifying moving parts based on a magnetically levitated flexible assembly line according to claim 1, characterized in that, The step of generating the control instruction set includes: Each mover is assigned an independent data version number, and an optimistic locking mechanism is used to resolve concurrent write conflicts. Based on the motion state of the mover, topological coordinates, and external command parameters, the predicted path for the next N cycles is calculated. A dynamic control instruction set is generated based on the predicted path. The instruction set includes braking instructions, speed adjustment instructions, and path replanning instructions.

8. A mover identification device based on a magnetically levitated flexible assembly line, used to execute the mover identification method based on a magnetically levitated flexible assembly line as described in any one of claims 1 to 7, characterized in that, include: Topology building blocks are used to activate pre-generated topology models; The data mapping module is used to periodically scan the mover encoder data of each slave driver, including initiating periodic data polling of each slave driver to obtain the real-time position and speed data of each mover encoder; Add a timestamp to the data corresponding to each actuator, and synchronize the data of different actuators in time. Based on the synchronized mover velocity data, the real-time position of each mover is mapped to the normalized virtual coordinates of the topology model; the real-time position and velocity data of each mover are obtained, and the position data of each mover is mapped to the corresponding coordinates of the topology model; The data processing module is used to perform noise filtering and smoothing on the real-time position and velocity data of the mover; The algorithm matching module is used to select the corresponding recognition algorithm based on the line size data, bus communication frequency data, preset task sequence data of each mover, and processed mover velocity data. This includes selecting the improved Hungarian algorithm when the system computing power reaches a preset sufficient computing power threshold after evaluation, and the current topology is not homomorphic to the ring topology (meaning the current topology cannot be transformed into a ring topology through continuous deformation). The improved Hungarian algorithm can adapt to any topology. When the bus communication frequency is monitored to be lower than a preset frequency threshold, or the hardware stability is lower than a preset stability threshold and multiple verifications are required based on system operation needs, a frame matching algorithm is selected. When the system computing power is evaluated to be lower than a preset computing power standard, and the corresponding topology is analyzed and determined to be a simple structure homomorphic to the ring, a sequence tracking algorithm is selected. The trajectory generation module is used to perform correlation analysis on the processed real-time position and velocity data of the mover according to the selected recognition algorithm to generate the mover trajectory data; The data verification module is used to verify the trajectory data of the moving part; if the verification is abnormal, an early warning will be triggered. The data application module is used to update the verified trajectory data of the mover to the real-time status database of the mover. Based on the coordinates of the topology model and the motion state of the mover, and combined with external command parameters, it generates a set of control commands and outputs them to the path planning system.

Images