Method and device for updating mover identification information, magnetic drive conveying system, readable storage medium and program product

Abstract

The present application relates to the field of magnetic drive conveying technology, specifically to a method, device, magnetic drive conveying system, readable storage medium and program product for updating the identification information of a mover. The method comprises: obtaining multiple frames of images of the mover in the current time period and the actual position information of the mover; inputting the multiple frames of images into the recognition model respectively, and outputting the bounding box information and general identification information of the mover in each frame of the multiple frames; inputting the bounding box information and the general identification information into the tracking model to determine the running trajectory and unique identification information of the mover; determining the image position information of the mover, and performing coordinate transformation on the image position information of the mover to obtain the transformed first position information; when the transformed first position information matches the actual position information, updating the identification information corresponding to the actual position information in the application scenario to the unique identification information corresponding to the transformed first position information. The use of this method can ensure that the identification information corresponding to the mover remains unchanged during operation.

Description

Technical Field

[0001] The present application relates to the field of magnetic drive conveying technology, and in particular to a method and device for updating mover identification information, a magnetic drive conveying system, a readable storage medium and a program product. Background Art

[0002] In the magnetic drive conveying system, the interaction between the magnetic field of the stator module and the magnetic field of the mover generates a torque, which drives the mover to operate, enabling flexible and accurate conveying of materials or products. Therefore, the magnetic drive conveying system has been increasingly widely used in the fields of processing production, logistics transmission, etc.

[0003] When the mover is running, the magnetic drive conveying system needs to track the position information of the mover and control the mover in real time. The position information of the mover corresponds to the mover number, and the control of the mover is performed based on the mover number. However, the correspondence between the mover and the mover number is easy to be confused (for example, when there is a fault or power failure, the correct number is difficult to save), resulting in the inability to accurately control the corresponding mover.

[0004] In the related technology, RFID (Radio Frequency Identification) tags, QR code scanning, adding wireless communication modules, etc. are mainly used to ensure that the corresponding number of the mover remains unchanged when it is in operation, which is inefficient. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device, magnetic drive conveying system, computer-readable storage medium and computer program product for updating the mover identification information, which can ensure that the corresponding identification information of the mover remains unchanged during operation, in order to address the above technical problems.

[0006] In a first aspect, the present application provides a method for updating mover identification information, comprising:

[0007] Acquire multiple frames of images of the mover in the current time period and actual position information of the mover, wherein the multiple frames of images are obtained by continuously shooting the mover in the application scene through a camera;

[0008] Inputting the multiple frames of images into the recognition model respectively, and outputting the bounding box information and general identification information of the mover in each frame of the multiple frames of images;

[0009] Inputting the bounding box information and the universal identification information of the mover into the tracking model to determine the running trajectory of the mover and the unique identification information of the mover;

[0010] Based on the running track of the mover, determine the image position information of the mover, and perform coordinate transformation on the image position information of the mover to obtain transformed first position information;

[0011] In a case where the converted first location information matches the actual location information, the identification information corresponding to the actual location information in the application scenario is updated to the unique identification information corresponding to the converted first location information.

[0012] In one of the embodiments, when the converted first location information matches the actual location information, updating the identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information includes:

[0013] Converting the actual position information to a target coordinate system to obtain converted second position information; wherein the target coordinate system includes the same coordinate system as the image position information conversion;

[0014] In a case where the distance between the converted first location information and the converted second location information is within a preset range, the identification information corresponding to the actual location information in the application scenario is updated to the unique identification information corresponding to the converted first location information.

[0015] In one embodiment, the target coordinate system includes a world coordinate system, the actual position information includes relative position information of the mover relative to the stator module, and the actual position information is converted to the target coordinate system to obtain the converted second position information, including:

[0016] Based on the position of the stator module in the world coordinate system, the relative position information is converted to the world coordinate system to obtain converted second position information.

[0017] In one embodiment, coordinate transformation is performed on the image position information of the mover to obtain the transformed first position information, including:

[0018] The camera is set relatively fixed to the magnetic drive conveying system;

[0019] photographing the calibration object in the magnetic drive conveying system based on the camera to obtain a calibration image;

[0020] Extracting feature points from the calibration image, and determining the extrinsic parameters of the camera based on position information of the feature points in a world coordinate system;

[0021] Based on the external parameters of the camera, coordinate transformation is performed on the image position information of the mover to obtain transformed first position information.

[0022] In one of the embodiments, when the converted first location information matches the actual location information, updating the identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information, then comprising:

[0023] Based on the updated identification information corresponding to the actual position information, a control instruction of the mover corresponding to the actual position information is determined, and the control instruction is used to instruct the magnetic drive device to drive the mover to operate.

[0024] In one embodiment, the training method of the recognition model includes:

[0025] Acquire a sample image set, wherein the sample image set includes annotated sample images, wherein the sample images are obtained by shooting the magnetic drive conveying system at multiple angles with a camera, and the sample images are annotated with sample movers and corresponding detection frames;

[0026] Inputting the sample image set into an initial recognition model and outputting a prediction result;

[0027] Based on the difference between the prediction result and the sample mover and the detection frame, the initial recognition model is iteratively adjusted to obtain a recognition model.

[0028] In a second aspect, the present application further provides a device for updating the identification information of a mover, comprising:

[0029] An acquisition module, used to acquire multiple frames of images of the mover in the current time period and the actual position information of the mover, wherein the multiple frames of images are obtained by continuously shooting the mover in the application scene through a camera;

[0030] A first processing module, used for inputting the multiple frames of images into a recognition model respectively, and outputting bounding box information and general identification information of the mover of each frame of the multiple frames of images;

[0031] A second processing module is used to input the bounding box information and the general identification information of the mover into the tracking model to determine the running trajectory of the mover and the unique identification information of the mover;

[0032] A conversion module, used to determine the image position information of the mover based on the running track of the mover, and perform coordinate conversion on the image position information of the mover to obtain converted first position information;

[0033] An updating module is used to update the identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information when the converted first location information matches the actual location information.

[0034] In a third aspect, the present application further provides a magnetic drive conveying system, the magnetic drive conveying system comprising:

[0035] A magnetic drive system, the magnetic drive system comprising a feedback module and a magnetic drive module; wherein the feedback module is used to determine the actual position information of the mover, and the magnetic drive module is used to drive the mover to operate;

[0036] Visual recognition system, used to photograph the moving parts in the application scenario;

[0037] A control system for executing the steps of any of the methods described above.

[0038] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of any of the methods described above when executed by a processor.

[0039] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of any of the methods described above when executed by a processor.

[0040] The above-mentioned method, device, magnetic drive conveying system, computer-readable storage medium and computer program product for updating the identification information of the mover can maintain the identification information of the mover on each running track in the image unchanged by processing multiple frames of images of the mover in the magnetic drive conveying system through pre-trained recognition models and tracking models. Through coordinate conversion and matching, the identification information of the mover in the image is assigned to the mover at the corresponding position in the application scene, so that when the actual position information of the mover in the application scene changes, the corresponding identification information remains unchanged, that is, the continuity of the position information corresponding to the identification information is guaranteed. Furthermore, the above-mentioned method is adopted in each operation cycle of the mover. Since the control of the mover in the magnetic drive conveying system is based on the position of the mover, it is possible to realize the continuous control of the mover, the distinction of the mover in complex logistics lines, the dynamic deletion of the mover, the control of the upper and lower belts of the mover, and other operations. In addition, there is no need to modify the mover, which ensures the flexibility of the magnetic drive conveying system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0042] FIG1 is an application environment diagram of a method for updating mover identification information in one embodiment;

[0043] FIG2 is a schematic flow chart of a method for updating mover identification information in one embodiment;

[0044] FIG3 is a schematic diagram of a flow chart of step S210 in one embodiment;

[0045] FIG4 is a schematic flow chart of a method for updating mover identification information in another embodiment;

[0046] FIG5 is a schematic flow chart of a method for updating mover identification information in yet another embodiment;

[0047] FIG. 6 is a structural block diagram of a device for updating the identification information of a mover in one embodiment. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0049] The method for updating the mover identification information provided in the embodiment of the present application can be applied to the application environment shown in Figure 1. Among them, the magnetic drive system 101 includes a feedback module 1011 and a magnetic drive module 1012, the feedback module 1011 is used to determine the actual position information of the mover, and the magnetic drive module 1012 is used to drive the mover to run. The camera in the visual recognition system 102 is used to continuously shoot the mover running in the application scenario. The control system 103 is used to control the magnetic drive system and the visual recognition system, and to update the identification information of the mover. Specifically, the mover runs in the magnetic drive conveying system 100, which is used for logistics transportation production and processing. The mover can be a transportation device, such as a trolley or a carrier.

[0050] In one embodiment, as shown in FIG. 2 , a method for updating the identification information of a mover is provided. In this embodiment, the method is applied to the control system 103 in FIG. 1 . The method includes the following steps:

[0051] Step S202, obtaining multiple frames of images of the mover in the current time period and the actual position information of the mover, wherein the multiple frames of images are obtained by continuously shooting the mover running in the application scene with a camera.

[0052] Exemplarily, the current time period may include the current moment and a preset time period before the current moment. The mover may be a trolley or a carrier in a magnetic drive conveying system. The application scenario may be a working scenario in a magnetic drive conveying system. Multiple frames of images may be continuous in time. The actual position information of the mover may be obtained by a position sensor, such as a laser sensor, an infrared scanning sensor, an ultrasonic sensor, etc.

[0053] Step S204 , inputting the multiple frames of images into the recognition model respectively, and outputting the bounding box information and general identification information of the mover in each frame of the multiple frames of images.

[0054] Among them, the recognition model can identify the mover in the image and output the bounding box information and general identification information of the mover. The bounding box information may include bounding box coordinates, and the bounding box information is used to determine the image position information of the mover. The general identification information may be a number, the number of movers in each frame of image may be multiple, and the general identification information of multiple movers in each frame of image may be the same. For example, images of multiple movers and other objects are input into the recognition model, and the output result may be: the mover and its location are marked by a bounding box, and carry the general identification information of the mover, such as a mover. Among them, the general identification information is used to distinguish the mover from other objects, and is not used to distinguish which specific mover among the movers.

[0055] The recognition model can be obtained through machine learning training. For example, the training method of the recognition model may include: using a camera to shoot the magnetic drive conveying system from multiple angles to obtain a large number of sample images; manually marking the movers in the sample images to obtain a sample image data set; inputting the sample image data set into the initial recognition model to output the predicted movers. Based on the difference between the predicted movers and the marked movers, the training parameters in the initial recognition model are iteratively adjusted to obtain the recognition model.

[0056] Step S206 , inputting the bounding box information and the general identification information of the mover into the tracking model to determine the running trajectory of the mover and the unique identification information of the mover.

[0057] Among them, the unique identification information can uniquely identify a specific mover, and the unique identification information can have a one-to-one correspondence with the mover. The tracking model is used to process multi-frame images with time continuity, output the movement trajectory of the mover in the multi-frame images and the unique identification information of the mover, and ensure that the corresponding identification information of the mover in the multi-frame images remains unchanged. The movement trajectory includes the image position information at the current moment and multiple historical moments.

[0058] Exemplarily, after inputting multiple frames of images into the recognition model, the data output by the recognition model can be input into the tracking model, and the running trajectory of the mover in the multiple frames of images and the unique identification information of the mover are output, so that the unique identification information of the mover on the same running trajectory in the multiple frames of images is the same. The unique identification information can be represented by characters or numbers, which is not limited in the embodiments of the present disclosure. In one possible implementation, the recognition model and the tracking model can be deployed in the control system 103.

[0059] Step S208, based on the running track of the mover, determine the image position information of the mover, and perform coordinate transformation on the image position information of the mover to obtain transformed first position information.

[0060] The image position information of the mover may be the coordinate information of the mover in the camera coordinate system.

[0061] Exemplarily, the image position information of the mover at each moment in the current time period can be determined based on the running trajectory of the mover, and then the image position information of the mover at each moment in the current time period can be subjected to coordinate transformation to obtain the transformed first position information of the mover corresponding to each moment in the current time period. Alternatively, in the case where the actual position information of the mover is collected in real time, only the image position information of the mover at the current moment can be subjected to coordinate transformation to obtain the first position information of the mover at the current moment. In one possible implementation, the image position information of the mover can be transformed into the world coordinate system to obtain the transformed first position information. In another possible implementation, the image position information of the mover can also be transformed into the same coordinate system as the actual position information. It can be understood that for each moment in the current time period, the mover corresponds one-to-one to the transformed first position information.

[0062] Step S210: When the converted first location information matches the actual location information, the identification information corresponding to the actual location information in the application scenario is updated to the unique identification information corresponding to the converted first location information.

[0063] The matching of the converted first position information and the actual position information can be used for characterization, and the actual position of the mover in the application scenario can be predicted based on the position of the mover in the image. In the case where the actual position information is the three-dimensional coordinates of the mover in the world coordinate system, the matching of the converted first position information and the actual position information can be that the distance between the converted first position information and the actual position information is within a preset range.

[0064] Exemplarily, the converted first position information corresponding to the latest moment can be matched with the actual position information, and the identification information corresponding to the actual position information in the application scenario can be updated to the unique identification information corresponding to the converted first position information matching the actual position. It can be understood that in the case where there are multiple movers, the number of converted first position information and actual position information corresponding to each moment in the current time period is multiple, and the converted first position information corresponding to the same moment and the corresponding actual position information correspond one to one.

[0065] In the above-mentioned method for updating the identification information of the mover, by processing multiple frames of images of the mover in the magnetic drive conveying system during operation through the pre-trained recognition model and tracking model, the identification information of the mover on each running track in the image can be maintained unchanged. By coordinate conversion and matching, the identification information of the mover in the image is assigned to the mover at the corresponding position in the application scenario, so that when the actual position information of the mover in the application scenario changes, the corresponding identification information remains unchanged, that is, the continuity of the position information corresponding to the identification information is guaranteed. Furthermore, the above-mentioned method is adopted in each operation cycle of the mover. Since the control of the mover in the magnetic drive conveying system is based on the position of the mover, it is possible to realize the continuous control of the mover, the distinction of the mover in complex logistics lines, the dynamic deletion of the mover, and the control of the upper and lower belts of the mover. In addition, there is no need to modify the mover, which ensures the flexibility of the magnetic drive conveying system.

[0066] In an exemplary embodiment, as shown in FIG3 , when the converted first location information matches the actual location information, updating the identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information includes:

[0067] Step A1, converting the actual position information into a target coordinate system to obtain converted second position information; wherein the target coordinate system includes the same coordinate system as the image position information conversion.

[0068] Step A2: when the distance between the converted first location information and the converted second location information is within a preset range, the identification information corresponding to the actual location information in the application scenario is updated to the unique identification information corresponding to the converted first location information.

[0069] Wherein, the target coordinate system includes the same coordinate system as the image position information conversion. In an exemplary embodiment, when the target coordinate system is the same as the coordinate system where the actual position information is located, the actual position information may not be converted. In another exemplary embodiment, when the target coordinate system is different from the coordinate system where the actual position information is located, both the image position information and the actual position information may be converted to a third coordinate system, wherein the third coordinate system may be a world coordinate system.

[0070] Furthermore, the distance between the converted first position information and the converted second position information is calculated, for example, the difference between the horizontal coordinate corresponding to the first position information and the horizontal coordinate corresponding to the second position information, or the difference between the vertical coordinate corresponding to the first position information and the vertical coordinate corresponding to the second position information, etc. When the above difference is within a preset range, it is determined that the first position information and the second position information match, then the mover corresponding to the first position information and the mover corresponding to the second position information belong to the same mover.

[0071] In an exemplary embodiment, the unique identification information of the mover corresponding to the first position information in the image is used as the identification information of the mover corresponding to the second position information.

[0072] In this embodiment, by converting the image position information and the actual position information to the same coordinate system, when the converted position information is the same, it is determined that the converted first position information matches the actual position information, that is, the image position information and the actual position information correspond to the same mover, thereby achieving rapid matching of the image position and the actual position.

[0073] In an exemplary embodiment, the target coordinate system includes a world coordinate system, the actual position information includes relative position information of the mover relative to the stator module, and the actual position information is converted to the target coordinate system to obtain the converted second position information, including:

[0074] Based on the position of the stator module in the world coordinate system, the relative position information is converted to the world coordinate system to obtain converted second position information.

[0075] In a specific implementation, the relative position information of the mover relative to the stator module can be expressed as (path, pos). According to the position expression function of the preset path in the world coordinate system, (path, pos) can be converted to the world coordinate system to obtain the second position information (X2, Y2, Z2).

[0076] In this embodiment, the actual position information of the mover can be converted to the world coordinate system through the position of the mover relative to the stator module and the position of the stator module in the world coordinate system, and then the correspondence between the mover in the image and the mover in the application scenario can be determined.

[0077] Furthermore, as shown in FIG4 , the method for updating the mover identification information may further include:

[0078] Step S2012, setting the camera and the magnetic drive conveying system to be relatively fixed.

[0079] Step S2014: photograph the calibration object in the magnetic drive conveying system based on the camera to obtain a calibration image.

[0080] Step S2016, extracting feature points in the calibration image, and determining the extrinsic parameters of the camera based on the position information of the feature points in the world coordinate system.

[0081] Step S2018: Based on the external parameters of the camera, coordinate transformation is performed on the image position information of the mover to obtain transformed first position information.

[0082] Specifically, a calibration scene is prepared, wherein the calibration scene includes a calibration object, and the calibration object can be a calibration object whose position is known and easy to measure. An image containing the calibration object is captured by a camera, ensuring that the image covers the entire calibration object, and multiple images are captured at different angles and positions to obtain a calibration image; feature points or feature areas of the calibration object are extracted from the calibration image, and the positions of these feature points or feature areas in the world coordinate system are known; the feature points or feature areas extracted from the calibration image are matched with the corresponding known positions in the world coordinate system; based on the matching information, the external parameters of the camera (the position and posture of the camera relative to the world coordinate system) are estimated by minimizing the reprojection error and other methods; and the quality of the calibration result is evaluated by using indicators such as residual analysis and reprojection error.

[0083] In an exemplary embodiment, the image position information of the mover can be converted to a world coordinate system based on the external parameters of the camera to obtain the converted first position information. For example, the external parameters of the camera can be obtained based on calibration of the camera. For example, the image position information of the mover can be expressed as (X, Y, Z). The image position information of the mover can be converted to a world coordinate system by a first formula to obtain a first coordinate (X1, Y1, Z1). The first formula may include:

[0084] ;

[0085] in, is the first coordinate, P0 is the image position information of the mover, R is the rotation matrix, the rotation matrix is ​​used to describe the rotation relationship between the camera coordinate system and the world coordinate system, T is the translation vector, and the translation vector is used to describe the position of the origin of the camera coordinate system in the world coordinate system.

[0086] In the above embodiment, by preparing a calibration scene and performing coordinate transformation on the image position information of the mover based on the external parameters of the camera, the transformed first position information can be obtained conveniently and accurately.

[0087] In an exemplary embodiment, as shown in FIG5 , the method for updating the mover identification information may further include:

[0088] Step S212: determining a control instruction of the mover corresponding to the actual position information based on the identification information corresponding to the updated actual position information, wherein the control instruction is used to instruct the magnetic drive device to drive the mover to operate.

[0089] Among them, the identification information corresponding to the updated actual position information can be used to characterize a unique mover. Moreover, since the relationship between the actual position information and the identification information of the mover in the application scenario is updated based on the identification information of the mover in the image, it is ensured that the corresponding identification information remains unchanged after the actual position information changes. At this time, the actual position information corresponding to the identification information may be a periodic change in the coordinates of the mover. Therefore, PID control of the mover can be achieved. Exemplarily, the position information and other related information pre-stored by the corresponding mover can be obtained according to the identification information corresponding to the actual position information, and a control instruction can be output based on the control algorithm to instruct the magnetic drive device to drive the mover.

[0090] In this embodiment, since the control of the mover by the magnetic drive conveying system is indexed by the identification information of the mover at the actual position, and since the identification information of the mover at the actual position remains unchanged, the continuity of the control of the mover is ensured.

[0091] In an exemplary embodiment, a camera can be used to photograph the magnetic drive conveying system from multiple angles to obtain a large number of sample images; the sample image data set can be obtained by manually annotating the movers in the sample images; and the recognition model can be obtained by training the sample image data set. Specifically, an industrial depth camera can be used to collect a large number of sample images of various angles for the magnetic drive system; accurate and complete sample images can be screened out; the sample image data set can be obtained by manually annotating the movers in the sample images; the sample image data set can be balanced so that the number of sample images at each angle in the sample image data set is balanced; the sample image data set is divided into a training set, a validation set, and a test set, the model is trained with the training set, the hyperparameters are adjusted and the model is selected with the validation set, and the performance of the model is evaluated with the test set. In one possible implementation, the training method of the above-mentioned recognition model includes:

[0092] Step B1, obtaining a sample image set, the sample image set includes annotated sample images, the sample images are obtained by shooting the magnetic drive conveying system at multiple angles with a camera, and the sample images are annotated with sample movers and corresponding detection frames.

[0093] Specifically, the sample image set includes a plurality of labeled sample images, each of which may be labeled with a sample mover and a detection frame of the sample mover.

[0094] Step B2: input the sample image set into the initial recognition model and output the prediction result.

[0095] Specifically, the prediction result may include the category of the mover and the predicted detection box corresponding to the mover. In an exemplary embodiment, the predicted detection box may include the predicted positions of the four vertices of the detection box and the general identification information corresponding to the detection box.

[0096] Step B3, based on the difference between the prediction result and the mover and the detection frame, the initial recognition model is iteratively adjusted to obtain a recognition model.

[0097] Specifically, a loss function can be constructed based on the difference between the labeled sample movers and the predicted movers, and the difference between the labeled detection frame and the predicted detection frame. In an exemplary embodiment, the training parameters in the initial recognition model are iteratively adjusted until the preset requirements are met, such as the number of iterations meets the preset number or the model accuracy meets the predicted accuracy, and the adjustment is terminated to obtain the recognition model.

[0098] Optionally, a recognition model can be trained based on the YOLO algorithm. The YOLO (You Only Look Once) algorithm divides the input image into grids and predicts the bounding box and category of the target in each grid box. The YOLO algorithm regards the target detection problem as a regression problem, and a convolutional neural network can directly obtain the bounding boxes and category probabilities of all targets in a single forward propagation. Optionally, a tracking model can be trained based on the DeepSORT algorithm. The DeepSORT (DeepSimple Online and Realtime Tracking) algorithm is used to track targets in multiple frames, track the motion trajectory of the target by calculating the target matching relationship between adjacent frames, and ensure the unique identification of the target by associating the bounding boxes of the same target in different frames.

[0099] In summary, in the above-mentioned method for updating the identification information of the mover, by processing the multi-frame images of the mover in the magnetic drive conveying system during operation through the pre-trained recognition model and tracking model, the identification information of the mover on each running track in the image can be maintained unchanged. By coordinate conversion and matching, the identification information of the mover in the image is assigned to the mover at the corresponding position in the application scenario, so that when the actual position information of the mover in the application scenario changes, the corresponding identification information remains unchanged, that is, the continuity of the position information corresponding to the identification information is guaranteed. Furthermore, the above-mentioned method is adopted in each operation cycle of the mover. Since the control of the mover in the magnetic drive conveying system is based on the position of the mover, it is possible to realize the continuous control of the mover, the distinction of the mover in complex logistics lines, the dynamic deletion of the mover, and the control of the upper and lower belts of the mover. In addition, there is no need to modify the mover, which ensures the flexibility of the magnetic drive conveying system.

[0100] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0101] Based on the same inventive concept, the embodiment of the present application also provides a device for updating the mover identification information for implementing the method for updating the mover identification information involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more mover identification information update devices provided below can refer to the limitations of the method for updating the mover identification information above, and will not be repeated here.

[0102] In an exemplary embodiment, as shown in FIG6 , a device 300 for updating the identification information of a mover is provided, comprising: an acquisition module 301 , a first processing module 302 , a second processing module 303 , a conversion module 304 and an update module 305 , wherein:

[0103] The acquisition module 301 is used to acquire multiple frames of images of the mover in the current time period and the actual position information of the mover. The multiple frames of images are obtained by continuously shooting the mover running in the application scene with a camera.

[0104] The first processing module 302 is used to input the multiple frames of images into the recognition model respectively, and output the bounding box information and general identification information of the mover in each frame of the multiple frames of images.

[0105] The second processing module 303 is used to input the boundary box information and the general identification information of the mover into the tracking model to determine the running track of the mover and the unique identification information of the mover.

[0106] The conversion module 304 is used to determine the image position information of the mover based on the running trajectory of the mover, and perform coordinate conversion on the image position information of the mover to obtain converted first position information.

[0107] The updating module 305 is used to update the identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information when the converted first location information matches the actual location information.

[0108] In an exemplary embodiment, the updating module 305 includes:

[0109] The first conversion submodule is used to convert the actual position information into a target coordinate system to obtain converted second position information; wherein the target coordinate system includes the same coordinate system as the image position information conversion.

[0110] The updating submodule is used to update the identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information when the distance between the converted first location information and the converted second location information is within a preset range.

[0111] In an exemplary embodiment, the target coordinate system includes a world coordinate system, the actual position information includes relative position information of the mover relative to the stator module, and the first conversion submodule includes:

[0112] The conversion unit is used to convert the relative position information into the world coordinate system based on the position of the stator module in the world coordinate system to obtain converted second position information.

[0113] In an exemplary embodiment, the conversion module 304 further includes:

[0114] The position setting submodule is used to set the camera and the magnetic drive conveying system to be relatively fixed.

[0115] The shooting submodule is used to shoot the calibration object in the magnetic drive conveying system based on the camera to obtain the calibration image.

[0116] The parameter determination submodule is used to extract feature points in the calibration image and determine the external parameters of the camera based on the position information of the feature points in the world coordinate system.

[0117] The second conversion submodule is used to perform coordinate conversion on the image position information of the mover based on the external parameters of the camera to obtain converted first position information.

[0118] In an exemplary embodiment, the device 300 for updating the mover identification information further includes:

[0119] An instruction determination module is used to determine a control instruction of the mover based on the unique identification information of the mover in the application scenario, and the control instruction is used to drive the mover to operate.

[0120] In an exemplary embodiment, the training method of the above recognition model includes:

[0121] Acquire a sample image set, the sample image set includes annotated sample images, the sample images are obtained by shooting the magnetic drive conveying system at multiple angles with a camera, and the sample images are annotated with sample movers and corresponding detection frames;

[0122] Input the sample image set into the initial recognition model and output the prediction result;

[0123] Based on the differences between the prediction results and the sample movers and detection frames, the initial recognition model is iteratively adjusted to obtain the recognition model.

[0124] Each module in the above-mentioned device for updating the identification information of the mover can be implemented in whole or in part by software, hardware or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0125] In an exemplary embodiment, a magnetic drive conveying system is provided, comprising:

[0126] A magnetic drive system, comprising a feedback module and a magnetic drive module; wherein the feedback module is used to determine the actual position information of the mover, and the magnetic drive module is used to drive the mover to operate.

[0127] The visual recognition system is used to photograph the moving parts in the application scenario.

[0128] The controller is used to execute the steps in the above method embodiments.

[0129] Exemplarily, please continue to refer to Figure 1, the magnetic drive module 1012 is used to receive instructions from the control system 103, change the magnetic field of the stator module, and control the movement of the mover. The feedback module 1011 is used to feedback the position of the mover on the stator module to the control system 103. Among them, the feedback module 1011 can be used to feedback the position information of the mover relative to the stator module in real time through sensors such as Hall elements. The magnetic drive module 1012 can be used in the motor to make the current pass through the winding to generate a magnetic field, interact with the magnetic field generated by the permanent magnet or other electromagnets, generate a torque, and drive the mover to run. The visual recognition system 102 is used to shoot multiple movers running in the application scenario. Among them, the visual recognition system may include an industrial depth camera. The camera can be installed at a key position of the production line to capture the image of the mover. The camera has a resolution and response energy that can adapt to the speed of the production line. In a possible implementation, the visual recognition system can access the control system 103 through Ethernet, EtherCAT, etc. for data exchange. The control system 103 is used to update the identification information of the moving part in operation in the application scenario in real time based on the multi-frame images of the moving part in the current time period taken by the visual recognition system 102, the actual position information of the moving part fed back by the magnetic drive system 101, and the pre-trained recognition model and tracking model.

[0130] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0131] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0132] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0133] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0134] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for updating the identification information of a mover, characterized in that: The method comprises: Acquire multiple frames of images of the mover in the current time period and actual position information of the mover, wherein the multiple frames of images are obtained by continuously shooting the mover in the application scene through a camera; Inputting the multiple frames of images into the recognition model respectively, and outputting the bounding box information and general identification information of the mover in each frame of the multiple frames of images; Inputting the bounding box information and the universal identification information of the mover into the tracking model to determine the running trajectory of the mover and the unique identification information of the mover; Based on the running trajectory of the mover, the image position information of the mover is determined, and based on the external parameters of the camera, the image position information of the mover is coordinate transformed to obtain transformed first position information; the transformed first position information and the actual position information are located in the same coordinate system; In a case where the converted first location information matches the actual location information, the unique identification information corresponding to the actual location information in the application scenario is updated to the unique identification information corresponding to the converted first location information.

2. The method according to claim 1, characterized in that In a case where the converted first location information matches the actual location information, updating the unique identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information includes: Converting the actual position information to a target coordinate system to obtain converted second position information; wherein the target coordinate system includes the same coordinate system as the image position information conversion; When the distance between the converted first location information and the converted second location information is within a preset range, the unique identification information corresponding to the actual location information in the application scenario is updated to the unique identification information corresponding to the converted first location information.

3. The method according to claim 2, characterized in that The target coordinate system includes a world coordinate system, the actual position information includes relative position information of the mover relative to the stator module, and the actual position information is converted to the target coordinate system to obtain the converted second position information, including: Based on the position of the stator module in the world coordinate system, the relative position information is converted to the world coordinate system to obtain converted second position information.

4. The method according to claim 1, characterized in that: The step of performing coordinate transformation on the image position information of the mover based on the external parameters of the camera to obtain the transformed first position information includes: The camera is set relatively fixed to the magnetic drive conveying system; photographing the calibration object in the magnetic drive conveying system based on the camera to obtain a calibration image; Extracting feature points from the calibration image, and determining the extrinsic parameters of the camera based on position information of the feature points in a world coordinate system; Based on the external parameters of the camera, coordinate transformation is performed on the image position information of the mover to obtain transformed first position information.

5. The method according to claim 1, characterized in that: In a case where the converted first location information matches the actual location information, updating the unique identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information, and then including: Based on the updated unique identification information corresponding to the actual position information, a control instruction of the mover corresponding to the actual position information is determined, wherein the control instruction is used to instruct the magnetic drive device to drive the mover to operate.

6. The method according to claim 1, characterized in that The training method of the recognition model includes: Acquire a sample image set, wherein the sample image set includes annotated sample images, wherein the sample images are obtained by shooting the magnetic drive conveying system at multiple angles with a camera, and the sample images are annotated with sample movers and corresponding detection frames; Inputting the sample image set into an initial recognition model and outputting a prediction result; Based on the difference between the prediction result and the sample mover and the detection frame, the initial recognition model is iteratively adjusted to obtain a recognition model.

7. A device for updating the identification information of a mover, characterized in that: The device comprises: An acquisition module, used to acquire multiple frames of images of the mover in the current time period and the actual position information of the mover, wherein the multiple frames of images are obtained by continuously shooting the mover in the application scene through a camera; A first processing module, used for inputting the multiple frames of images into a recognition model respectively, and outputting bounding box information and general identification information of the mover of each frame of the multiple frames of images; A second processing module is used to input the bounding box information and the general identification information of the mover into the tracking model to determine the running trajectory of the mover and the unique identification information of the mover; A conversion module, used to determine the image position information of the mover based on the running trajectory of the mover, and to perform coordinate conversion on the image position information of the mover based on the external parameters of the camera to obtain converted first position information; the converted first position information and the actual position information are located in the same coordinate system; An updating module is used to update the unique identification information corresponding to the actual location information in the application scenario to the unique identification information corresponding to the converted first location information when the converted first location information matches the actual location information.

8. A magnetic drive conveying system, characterized in that: The magnetic drive conveying system comprises: A magnetic drive system, the magnetic drive system comprising a feedback module and a magnetic drive module; wherein the feedback module is used to determine the actual position information of the mover, and the magnetic drive module is used to drive the mover to operate; Visual recognition system, used to photograph the moving parts in the application scenario; A control system for executing the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images