Quality control method and device based on teaching data, equipment and medium

Abstract

The application relates to the technical field of demonstration data, and particularly provides a quality control method, device and equipment based on demonstration data and a medium, aiming to solve the technical problem that massive point data generated in a demonstration process cannot be checked, thereby leading to low demonstration data quality in the prior art. To this end, the quality control method based on demonstration data comprises the following steps: based on a data acquisition task, checking data before acquisition before data acquisition, and preventing the acquisition process if the checking fails; monitoring the acquired data during the data acquisition process, and providing real-time feedback or executing preset control logic to an operator according to the monitoring result; checking the completeness and logic of the acquired data after the data acquisition is completed; verifying multiple associated data files of the same acquisition task before data uploading; and deleting the acquired data after the data uploading is completed, provided that the information fed back by a cloud service is that the uploading is successful.

Description

Technical Field

[0001] This application relates to the field of teaching data technology, specifically providing a quality control method, apparatus, equipment, and medium based on teaching data. Background Technology

[0002] With the rapid development of industrial automation technology, robots have been widely used in various industrial production scenarios such as welding, assembly, handling, and spraying. Teaching data, as the core basis for the robot to subsequently reproduce precise work trajectories, directly determines the accuracy, stability, and production safety of the robot's operations. In traditional teaching processes, operators typically manually drag the robot's end effector using a handheld teach pendant to record key parameters such as motion trajectory, point coordinates, and speed, ultimately generating teaching data that can be executed by the robot.

[0003] In existing technologies, the generation process of teaching data generally relies on the experience and judgment of operators to complete quality verification. The industry's conventional quality control method is to verify whether there are significant deviations in the trajectory and whether there is a risk of collision through manual no-load test runs after the teaching is completed. The post-event verification method can only cover a small number of typical operation scenarios and cannot perform full-dimensional compliance verification on the massive amount of point data generated during the teaching process. It is difficult to identify teaching data that meets the basic trajectory requirements but does not meet the process specifications, thus leading to the technical problem of low teaching data quality.

[0004] Accordingly, there is a need in the field for a new quality control method based on teaching data to address the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects, this application is made to provide a solution or at least a partial solution to the technical problem in the prior art that it is impossible to perform full-dimensional compliance verification on the massive point data generated during the teaching process, which leads to low teaching data quality.

[0006] In a first aspect, this application provides a quality control method based on teaching data, comprising:

[0007] Data acquisition tasks;

[0008] Based on the data acquisition task, the data before acquisition is checked, and if the check fails, the acquisition process is stopped.

[0009] During the data acquisition process, the acquired data is monitored, and real-time feedback is provided to the operator or preset control logic is executed based on the monitoring results.

[0010] After data collection is completed, the completeness and logical consistency of the collected data are verified.

[0011] Before uploading the data, verify multiple related data files for the same data collection task;

[0012] Once the data upload is complete and the cloud service confirms a successful upload, the collected data will be deleted.

[0013] In one technical solution of the above-mentioned quality control method based on teaching data, the step of checking the data before data acquisition and blocking the acquisition process if the check fails includes:

[0014] Obtain data before collection;

[0015] The data collected before collection is compared with its corresponding preset value to obtain a first comparison result;

[0016] Obtain the data collection quota and compare it with the preset maximum quota to obtain a second comparison result;

[0017] Connect to external devices and obtain connection results;

[0018] If at least one of the first comparison result, the second comparison result, or the connection result is determined to be a negative result, the acquisition process is blocked.

[0019] In one technical solution of the above-mentioned quality control method based on teaching data, the step of monitoring the collected data during the data acquisition process and providing real-time feedback to the operator or executing preset control logic based on the monitoring results includes:

[0020] Acquire tracking status of the tracking device and location data of key body parts;

[0021] Determine whether the location data has been refreshed within a preset time.

[0022] If it is determined that the data has not been refreshed and / or the tracking status is not tracked, the data acquisition process will be blocked.

[0023] In one technical solution of the above-mentioned quality control method based on teaching data, the step of monitoring the collected data during the data acquisition process and providing real-time feedback to the operator or executing preset control logic based on the monitoring results further includes:

[0024] Get the recording status within a preset time period;

[0025] If the recording status is determined to be empty, the operator will be prompted that the recording start failed, and the recording device will be shut down.

[0026] In one technical solution of the above-mentioned quality control method based on teaching data, the step of monitoring the collected data during the data acquisition process and providing real-time feedback to the operator or executing preset control logic based on the monitoring results further includes:

[0027] Acquire hand tracking signals and displacement of key hand points;

[0028] If the hand tracking signal is determined to be in a tracking loss state or the confidence level is lower than the preset confidence level, an alarm message is sent to the operator.

[0029] Based on the displacement of the key hand points and the head display displacement, the relative motion velocity of the hand after head display compensation is obtained.

[0030] The head-mounted display's moving speed is obtained based on its displacement.

[0031] If the relative movement speed of the hand and / or the movement speed of the head-mounted display are determined to be greater than a preset speed threshold, a prompt message is sent to the operator.

[0032] In one technical solution of the above-mentioned quality control method based on teaching data, the step of verifying the integrity and logic of the collected data after data acquisition includes:

[0033] Obtain the storage path of the captured video;

[0034] If the storage path is found to contain the captured video, then the file size of the captured video is obtained;

[0035] If the file size is determined to be greater than a preset file size threshold, then the captured video is deemed valid.

[0036] The captured video is decoded to obtain decoded data;

[0037] The decoded data is analyzed to obtain the analysis results;

[0038] If the analysis result is determined to be abnormal, the captured video will be marked as unusable.

[0039] In one technical solution of the above-mentioned quality control method based on teaching data, the step of verifying multiple associated data files for the same acquisition task before data upload includes:

[0040] Obtain the task identifier and the task file corresponding to the task identifier;

[0041] If it is determined that at least one file in the task file is missing and / or fails verification, the upload of that task file will be blocked.

[0042] Secondly, this application provides a quality control device, comprising:

[0043] The acquisition module is used to acquire data collection tasks;

[0044] The inspection module is used to inspect the data before data acquisition based on the data acquisition task. If the inspection fails, the acquisition process is blocked.

[0045] The inspection module is also used to monitor the collected data during the data acquisition process and provide real-time feedback to the operator or execute preset control logic based on the monitoring results.

[0046] The inspection module is also used to verify the integrity and logical consistency of the collected data after the data collection is completed.

[0047] The inspection module is also used to verify multiple associated data files for the same data acquisition task before data is uploaded;

[0048] The processing module is used to delete the collected data if the cloud service confirms that the upload was successful after the data upload is completed.

[0049] Thirdly, this application provides a quality control device, including at least one processor and a memory; wherein,

[0050] The memory stores computer-executed instructions;

[0051] The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method described in any one of the first aspects.

[0052] Fourthly, this application provides a computer-readable storage medium storing a plurality of program codes adapted to be loaded and run by a processor to perform the method described in any one of the first aspects.

[0053] This application provides a quality control method, apparatus, device, and medium based on teaching data. The method specifically includes: acquiring a data acquisition task; checking the data before acquisition based on the task, and stopping the acquisition process if the check fails; monitoring the acquired data during acquisition and providing real-time feedback to the operator or executing preset control logic based on the monitoring results; verifying the integrity and logic of the acquired data after acquisition; verifying multiple related data files for the same acquisition task before uploading; and deleting the acquired data after successful upload if the cloud service feedback indicates successful upload. This improves the quality of the teaching data. Attached Figure Description

[0054] The disclosure of this application will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Furthermore, similar numbers in the drawings are used to denote similar components, wherein:

[0055] Figure 1 This is a flowchart illustrating a quality control method based on teaching data according to an embodiment of this application.

[0056] Figure 2 This is a flowchart illustrating a second embodiment of a quality control method based on teaching data provided in this application.

[0057] Figure 3 This is a flowchart illustrating a third embodiment of a quality control method based on teaching data provided in this application.

[0058] Figure 4 This is a flowchart illustrating Embodiment 4 of a quality control method based on teaching data provided in this application.

[0059] Figure 5 This is a flowchart illustrating Embodiment 5 of a quality control method based on teaching data provided in this application.

[0060] Figure 6 This is a flowchart illustrating a sixth embodiment of a quality control method based on teaching data provided in this application.

[0061] Figure 7 This is a flowchart illustrating Embodiment Seven of a quality control method based on teaching data provided in this application.

[0062] Figure 8 This is a schematic diagram of the structure of a quality control device provided in an embodiment of this application;

[0063] Figure 9 This is a schematic diagram of the structure of a quality control device provided in an embodiment of this application.

[0064] List of reference numerals in the attached diagram:

[0065] 11: Acquisition module; 12: Inspection module; 13: Processing module; 21: Processor; 22: Memory. Detailed Implementation

[0066] Some embodiments of this application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of this application and are not intended to limit the scope of protection of this application.

[0067] In the description of this application, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and can also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "a" or "this" can also include plural forms.

[0068] Existing teaching data generation technologies have many shortcomings in ensuring data quality:

[0069] First, existing methods often lack comprehensive, cross-modal quality checks. For example, a system might only check whether the video was successfully recorded, but ignore issues such as missing hand tracking data or corrupted metadata. This means that even if the video is normal, the entire data sample cannot be used for model training due to the lack of key modalities.

[0070] Secondly, quality checks are often delayed, relying mainly on cloud-based post-processing. This method cannot provide real-time feedback to operators at the data collection site. When problems such as tracking devices freezing or excessively fast hand movements occur, operators cannot perceive and correct them in time, resulting in a significant amount of time being wasted on collecting low-quality or invalid data.

[0071] Furthermore, the process of uploading data from the acquisition device to the cloud is fraught with risks. For example, a video file might only be partially uploaded due to incomplete writing or system malfunctions, but the system might mistakenly consider the upload successful. Additionally, within a task group (a task containing multiple files such as videos, tracking data, and metadata), if some files fail to upload, existing mechanisms might still mark the task as "complete," thus contaminating the training dataset in the cloud. Even more seriously, some systems delete the original local files before confirming task completion in the cloud. If quality issues are subsequently discovered with the cloud data, retransmission or repair will be impossible, resulting in permanent data loss.

[0072] Finally, existing technologies generally lack systematic anomaly handling mechanisms. For detected quality issues, they typically only provide simple failure messages, failing to employ differentiated intelligent processing strategies such as "blocking data collection," "isolating uploads," or "holding for retrying" based on the severity and recoverability of the problem. This results in poor automation and robustness.

[0073] Based on this, in order to solve the above-mentioned technical problems, the technical concept of this application is to provide a new quality control method based on teaching data in order to improve the quality of teaching data.

[0074] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0075] Figure 1 This is a flowchart illustrating an embodiment of a quality control method based on teaching data provided in this application. Figure 1 As shown, specifically, the method includes:

[0076] Step S101: Obtain data acquisition task.

[0077] In this embodiment, the user generates a data acquisition task according to the robot's training needs and sends the data acquisition task to the acquisition device.

[0078] Step S102: Based on the data acquisition task, check the data before data acquisition. If the check fails, stop the acquisition process.

[0079] In this embodiment, after receiving a data acquisition task, the acquisition device does not directly acquire data according to the task. Instead, it checks the pre-acquisition data, which typically includes the selected task, login account, environment identifier, and task identifier. If the pre-acquisition data check fails, the acquisition process is blocked. If the check passes, the acquisition process continues.

[0080] Step S103: During the data acquisition process, the acquired data is monitored, and real-time feedback is provided to the operator or preset control logic is executed based on the monitoring results.

[0081] In this embodiment, during the data acquisition process, the validity of the acquired data is monitored in real time, for example, whether the range of the data is within a preset range, and based on the real-time monitoring results, real-time feedback is provided to the operator or preset control logic is executed.

[0082] Step S104: After the data collection is completed, verify the integrity and logic of the collected data.

[0083] In this embodiment, after the data collection is completed, it is determined whether the collected data is complete. For example, for all files under a task identifier, it is determined whether all data has been collected. Furthermore, these files are analyzed to determine whether the logic is correct.

[0084] Step S105: Before uploading the data, verify multiple associated data files for the same data acquisition task.

[0085] In this embodiment, after data collection is completed, the data needs to be uploaded to the cloud. Before uploading to the cloud, it is determined whether all associated data files for the same collection task exist and whether they all pass verification. If the verification passes, the data is uploaded to the cloud.

[0086] Step S106: After the data upload is completed, if the cloud service confirms that the upload was successful, the collected data will be deleted.

[0087] In this embodiment, after all data has been uploaded, the task completion interface in the cloud is called. When the interface returns a successful upload message, the local task is marked as completed, and the locally cached collected data is deleted. If an error occurs during the upload or interface call process, the local file is retained for subsequent retransmission or manual processing. This ensures that local data is not lost in the event of network interruption or cloud service anomalies, and the system can safely retry or recover, demonstrating strong robustness.

[0088] In this embodiment, a data acquisition task is obtained; based on the data acquisition task, the data is checked before acquisition, and the acquisition process is stopped if the check fails; during the data acquisition process, the acquired data is monitored, and real-time feedback is provided to the operator or preset control logic is executed based on the monitoring results; after the data acquisition is completed, the integrity and logic of the acquired data are verified; before the data is uploaded, multiple associated data files of the same acquisition task are verified; after the data is uploaded, if the information fed back by the cloud service is confirmed as successful, the acquired data is deleted. Compared with the prior art, which cannot perform full-dimensional compliance verification of the massive point data generated during the teaching process, resulting in low teaching data quality, this application improves the quality of teaching data by dividing the teaching data generation process into multiple stages, verifying the data corresponding to each stage in each stage, and providing corresponding solutions based on the verification results.

[0089] Figure 2This is a flowchart illustrating a second embodiment of a quality control method based on teaching data provided in this application. Based on the above embodiments, as... Figure 2 As shown, specifically, in step S102, before data acquisition, the data is checked. If the check fails, the acquisition process is stopped, including:

[0090] Step S201: Obtain the data before collection.

[0091] In this embodiment, the data collected before collection includes login status, data collection task, project, environment identifier, and scene identifier.

[0092] Step S202: Compare the data before collection with its corresponding preset value to obtain the first comparison result.

[0093] In this embodiment, the system determines whether the data collection device is logged into an account based on the login status. It determines whether to retrieve a data collection task from the server based on locally stored task data. It determines whether a project has been selected based on locally stored project data. Finally, it determines whether the current environment and scene identifiers have been obtained through methods such as scanning a QR code.

[0094] Step S203: Obtain the data collection quota and compare it with the preset maximum quota to obtain the second comparison result.

[0095] In this embodiment, the cloud service interface is called to query the data collection quota under the current environment identifier or scene identifier, and the data collection quota is compared with the preset maximum quota to obtain a second comparison result. If the second comparison result is the same, the collection quantity of the scene or environment has reached the limit, and the current collection is blocked, and the operator is prompted with the reason for the block.

[0096] Step S204: Connect to an external device and obtain the connection result.

[0097] In this embodiment, when the current acquisition task requires the acquisition device to work collaboratively with an external device, the acquisition device sends a handshake command containing a data identifier to the external device and obtains confirmation information from the external device. If a correct response is not received within a preset time, the connection result is a connection failure; if a correct response is received within a preset time, the connection result is a connection success.

[0098] Step S205: If at least one of the first comparison result, the second comparison result, or the connection result is determined to be a negative result, the acquisition process is blocked.

[0099] In this embodiment, if at least one of the first comparison result, the second comparison result, or the connection result is negative, the acquisition process is blocked.

[0100] In this embodiment, data before collection is acquired; the data before collection is compared with its corresponding preset values ​​to obtain a first comparison result; a data collection quota is acquired and compared with a preset maximum quota to obtain a second comparison result; a connection is established with an external device to obtain a connection result; if at least one of the first comparison result, the second comparison result, or the connection result is determined to be a negative result, the collection process is blocked. Thus, before data collection, multiple gating checks are performed to prevent invalid collection caused by incomplete task information, unready devices, or insufficient cloud quotas from the source, avoiding resource waste.

[0101] Figure 3 This is a flowchart illustrating a third embodiment of a quality control method based on teaching data provided in this application. Based on the above embodiments, as... Figure 3 As shown, the specific implementation of step S103 includes:

[0102] Step S301: Obtain the tracking status of the tracking device and the location data of key body parts.

[0103] In this embodiment, the key body parts may be the head, left wrist, and right wrist.

[0104] Step S302: Determine whether the location data is refreshed within a preset time.

[0105] In this embodiment, the location data of key body parts is acquired at preset time intervals, and it is determined whether the location data is the same before and after the preset time interval. If they are the same, no refresh is performed; if they are different, it is determined that a refresh has been performed.

[0106] In this embodiment, for example, the preset time is 3 seconds.

[0107] Step S303: If it is determined that the data collection process is not refreshed and / or the tracking status is not tracked, then the data collection process is blocked.

[0108] In this embodiment, if the tracking status of the device is determined to be untracked and / or key body parts are not refreshed, it is determined that the tracker is abnormal and cannot collect data normally. At this time, it is necessary to block the data collection process.

[0109] In this embodiment, the tracking status of the tracking device and the location data of key body parts are acquired; it is determined whether the location data is refreshed within a preset time; if it is determined that the data has not been refreshed and / or the tracking status is not tracked, the data acquisition process is blocked to avoid invalid data acquisition, thereby improving data acquisition efficiency.

[0110] Figure 4This is a flowchart illustrating a fourth embodiment of a quality control method based on teaching data provided in this application. Based on the above embodiments, as... Figure 4 As shown, specifically, step S103 further includes:

[0111] Step S401: Obtain the recording status within the preset time period.

[0112] In this embodiment, a preset time is set before starting video recording, and the recording status returned by the video recording module is obtained within the preset time.

[0113] Step S402: If the recording status is found to be empty, the operator will be prompted that the recording start failed, and the recording device will be shut down.

[0114] In this embodiment, if the video recording module does not return to the recording state within a preset time, it is determined that the recording state is empty, the acquisition process is stopped, the camera and other related recording devices are turned off, and then the operator is prompted with a recording start failure message.

[0115] In this embodiment, the recording status within a preset time period is obtained; if the recording status is determined to be empty, the operator is prompted that the recording start failed and the recording device is shut down. During the data collection process, the recording status is monitored at preset time intervals, and if the recording status is determined to be abnormal, a prompt message is sent to the operator and the recording device is shut down to avoid wasting time by continuing to collect data when problems occur during the recording process.

[0116] Figure 5 This is a flowchart illustrating Embodiment 5 of a quality control method based on teaching data provided in this application. Based on the above embodiments, as follows... Figure 5 As shown, specifically, step S103 further includes:

[0117] Step S501: Acquire hand tracking signals and hand key point displacements.

[0118] During the data acquisition process, the displacement of key hand points in the world coordinate system is calculated in real time, and hand tracking signals are acquired. The key hand points are the wrist joint and the five fingertips.

[0119] Step S502: If the hand tracking signal is determined to be in a tracking loss state or the confidence level is lower than the preset confidence level, an alarm message is sent to the operator.

[0120] In this embodiment, if it is determined that the hand tracking signal is lost, then the hand tracking signal is in a tracking loss state; or, if the hand tracking signal is normal, but the confidence level is lower than the preset confidence level, then it is determined that the hand tracking signal is abnormal, and an alarm message is sent to the operator.

[0121] Step S503: Based on the displacement of the key points of the hand and the displacement of the head display, obtain the relative motion speed of the hand after compensation by the head display.

[0122] In this embodiment, the relative motion speed of the hand after head-mounted display compensation is obtained by subtracting the displacement of the key points of the hand from the displacement of the head-mounted display within the same time period.

[0123] Step S504: Obtain the head-mounted display movement speed based on the head-mounted display displacement.

[0124] In this embodiment, the ratio of the head-mounted display displacement to the time period is the head-mounted display movement speed.

[0125] Step S505: If the relative movement speed of the hand and / or the movement speed of the head display are greater than a preset speed threshold, a prompt message is sent to the operator.

[0126] In this embodiment, if the relative movement speed of the hand is greater than a preset speed threshold, and / or the head-mounted display movement speed is greater than a preset speed threshold, a prompt message indicating that the movement is too fast needs to be sent to the operator.

[0127] In this embodiment, hand tracking signals and hand key point displacements are acquired; if the hand tracking signal is found to be in a tracking loss state or the confidence level is lower than a preset confidence level, an alarm message is sent to the operator; based on the hand key point displacements and head-mounted display (HMD) displacements, the relative hand movement speed after HMD compensation is obtained; based on the HMD displacement, the HMD movement speed is obtained; if the relative hand movement speed and / or HMD movement speed are determined to be greater than a preset speed threshold, a prompt message is sent to the operator, enabling the operator to correct erroneous actions in a timely manner during data acquisition, thereby improving the success rate of data acquisition.

[0128] Figure 6 This is a flowchart illustrating a sixth embodiment of a quality control method based on teaching data provided in this application. Based on the above embodiments, as... Figure 6 As shown, the specific implementation of step S104 includes:

[0129] Step S601: Obtain the storage path of the captured video.

[0130] In this embodiment, if the recording device stops recording, it obtains the storage path of the captured video from the configuration file, and all the files recorded by the recording device are stored in that storage path.

[0131] Step S602: If the storage path contains the captured video, then obtain the file size of the captured video.

[0132] In this embodiment, all files under the storage path are retrieved, and their file extensions are obtained. Based on the file extensions, it is determined that a video capture exists. Then, the file size of the captured video is obtained.

[0133] Step S603: If the file size is greater than the preset file size threshold, then the video capture is considered valid.

[0134] In this embodiment, for example, the preset file size threshold is 1Kb.

[0135] In this embodiment, files with 0 bytes are excluded by determining that the file size is greater than a preset file size threshold.

[0136] In this embodiment, the size of the collected data can be detected three times consecutively. When the file size obtained from the three detections is the same, it is determined that the video file has been completely written to the memory.

[0137] Step S604: Decode the captured video to obtain decoded data.

[0138] In this embodiment, for the captured video that has been confirmed to be written, the video decoding component is called to read the metadata of the captured video to obtain the decoded data, which includes the frame rate, total number of frames and duration.

[0139] Step S605: Analyze the decoded data to obtain the analysis results.

[0140] In this embodiment, if decoding fails, times out, or the decoded data is illogical, the analysis result is abnormal.

[0141] Step S606: If the analysis result is determined to be abnormal, the acquired video will be marked as unusable.

[0142] In this embodiment, if the analysis result is abnormal, the captured video will be marked as unusable.

[0143] In this embodiment, after the data acquisition is completed, all relevant context information, including task information, environment scene ID, device serial number, video parameters, application version number, operator information, UTC start and end time, clock synchronization round-trip time, device temperature, and acquisition method, is written into a structured metadata file.

[0144] In this embodiment, the storage path of the captured video is obtained; if the captured video exists in the storage path, the file size of the captured video is obtained; if the file size is greater than a preset file size threshold, the captured video is determined to be valid; the captured video is decoded to obtain decoded data; the decoded data is analyzed to obtain analysis results; if the analysis results are determined to be abnormal, the captured video is marked as unusable. Through multiple checks on the existence, size, write stability, and resolvability of the video file, the upload of 0-byte, corrupt, or incomplete video files is effectively prevented, ensuring the physical and logical availability of the video file.

[0145] Figure 7 This is a flowchart illustrating Embodiment Seven of a quality control method based on teaching data provided in this application. Based on the above embodiments, as... Figure 7 As shown, the specific implementation of step S105 includes:

[0146] Step S701: Obtain the task identifier and the task file corresponding to the task identifier.

[0147] In this embodiment, the task file includes video files, tracking data files, and metadata files.

[0148] Step S702: If it is determined that at least one file in the task file is missing and / or fails verification, then the upload of the task file will be blocked.

[0149] In this embodiment, each task file is checked to determine whether all of them exist and whether they have all passed their respective quality checks. If at least one file in a task file is missing and / or fails the check, then the task file is considered erroneous, and uploading of that task file is prevented.

[0150] In this embodiment, a task identifier and the corresponding task file are obtained; if it is determined that at least one file in the task file is missing and / or fails verification, the upload of the task file is blocked. Through task file integrity verification and comprehensive metadata records, it is ensured that the multimodal data uploaded to the cloud is complete and contains rich contextual information, which greatly enhances the consistency and traceability of the dataset.

[0151] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of this application, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the scope of protection of this application.

[0152] Furthermore, this application also provides a quality control device.

[0153] Figure 8 This is a schematic diagram of a quality control device provided in an embodiment of this application. Figure 8As shown, the quality control device in this embodiment mainly includes an acquisition module 11, an inspection module 12, and a processing module 13. In some embodiments, one or more of the acquisition module 11, inspection module 12, and processing module 13 can be combined into a single module. In some embodiments, the acquisition module 11 can be configured to acquire data acquisition tasks. The inspection module 12 can be configured to check the data before acquisition based on the data acquisition task, and block the acquisition process if the check fails. The inspection module 12 can also be configured to monitor the acquired data during the data acquisition process and provide real-time feedback to the operator or execute preset control logic based on the monitoring results. The inspection module 12 can also be configured to verify the integrity and logic of the acquired data after the data acquisition is completed. The inspection module 12 can also be configured to verify multiple associated data files of the same acquisition task before data upload. The processing module 13 can also be configured to delete the acquired data if the cloud service feedback indicates successful upload after data upload is completed.

[0154] The aforementioned quality control device is used for performing Figures 1 to 7 The illustrated embodiments of the quality control method based on teaching data are similar in technical principle, the technical problems solved, and the technical effects produced. Those skilled in the art can clearly understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the quality control device can be found in the embodiments of the quality control method based on teaching data, and will not be repeated here.

[0155] Those skilled in the art will understand that all or part of the processes in the method of the above-described embodiment can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0156] Furthermore, this application also provides a quality control device.

[0157] Figure 9 This is a schematic diagram of the structure of a quality control device provided in an embodiment of this application. Figure 9As shown, the quality control device includes a processor 21 and a memory 22. The memory 22 can be configured to store a program for executing the quality control method based on teach data in the above-described method embodiments. The processor 21 can be configured to execute the program stored in the memory, which includes, but is not limited to, a program for executing the quality control method based on teach data in the above-described method embodiments. For ease of explanation, only the parts related to the embodiments of this application are shown. For specific technical details not disclosed, please refer to the method section of the embodiments of this application. The quality control device can be a control device device comprising various electronic devices.

[0158] Furthermore, this application also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to this application, the computer-readable storage medium can be configured to store a program that performs the quality control method based on teaching data described in the above-described method embodiments. This program can be loaded and run by a processor to implement the above-described quality control method based on teaching data. For ease of explanation, only the parts related to the embodiments of this application are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of this application. The computer-readable storage medium can be a storage device comprising various electronic devices. Optionally, in the embodiments of this application, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0159] Furthermore, it should be understood that since the various modules are only provided to illustrate the functional units of the device described in this application, the physical devices corresponding to these modules may be the processor itself, or a part of the processor's software, hardware, or a combination of both. Therefore, the number of modules shown in the figures is merely illustrative.

[0160] Those skilled in the art will understand that the various modules in the device can be adaptively split or combined. Such splitting or combining of specific modules will not cause the technical solution to deviate from the principles of this application; therefore, the technical solutions after splitting or combining will fall within the protection scope of this application.

[0161] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. Without departing from the principles of this application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of this application.

Claims

1. A quality control method based on teaching data, characterized in that, include: Data acquisition tasks; Based on the data acquisition task, the data before acquisition is checked, and if the check fails, the acquisition process is stopped. During the data acquisition process, the acquired data is monitored, and real-time feedback is provided to the operator or preset control logic is executed based on the monitoring results. After data collection is completed, the completeness and logical consistency of the collected data are verified. Before uploading the data, verify multiple related data files for the same data collection task; Once the data upload is complete and the cloud service confirms a successful upload, the collected data will be deleted.

2. The method according to claim 1, characterized in that, Before data collection, the data is checked, and the collection process is stopped if the check fails. This includes: Obtain data before collection; The data collected before collection is compared with its corresponding preset value to obtain a first comparison result; Obtain the data collection quota and compare it with the preset maximum quota to obtain a second comparison result; Connect to external devices and obtain connection results; If at least one of the first comparison result, the second comparison result, or the connection result is determined to be a negative result, the acquisition process is blocked.

3. The method according to claim 2, characterized in that, During the data acquisition process, the acquired data is monitored, and real-time feedback is provided to the operator or preset control logic is executed based on the monitoring results, including: Acquire tracking status of the tracking device and location data of key body parts; Determine whether the location data has been refreshed within a preset time. If it is determined that the data has not been refreshed and / or the tracking status is not tracked, the data acquisition process will be blocked.

4. The method according to claim 2, characterized in that, The process of monitoring the collected data during data acquisition and providing real-time feedback to the operator or executing preset control logic based on the monitoring results also includes: Get the recording status within a preset time period; If the recording status is determined to be empty, the operator will be prompted that the recording start failed, and the recording device will be shut down.

5. The method according to claim 2, characterized in that, The process of monitoring the collected data during data acquisition and providing real-time feedback to the operator or executing preset control logic based on the monitoring results also includes: Acquire hand tracking signals and displacement of key hand points; If the hand tracking signal is determined to be in a tracking loss state or the confidence level is lower than the preset confidence level, an alarm message is sent to the operator. Based on the displacement of the key hand points and the head display displacement, the relative motion velocity of the hand after head display compensation is obtained. The head-mounted display's moving speed is obtained based on its displacement. If the relative movement speed of the hand and / or the movement speed of the head-mounted display are determined to be greater than a preset speed threshold, a prompt message is sent to the operator.

6. The method according to claim 5, characterized in that, After data collection is completed, the integrity and logical consistency of the collected data are verified, including: Obtain the storage path of the captured video; If the storage path is found to contain the captured video, then the file size of the captured video is obtained; If the file size is determined to be greater than a preset file size threshold, then the captured video is deemed valid. The captured video is decoded to obtain decoded data; The decoded data is analyzed to obtain the analysis results; If the analysis result is determined to be abnormal, the captured video will be marked as unusable.

7. The method according to claim 6, characterized in that, The step of verifying multiple associated data files for the same data acquisition task before uploading the data includes: Obtain the task identifier and the task file corresponding to the task identifier; If it is determined that at least one file in the task file is missing and / or fails verification, the upload of that task file will be blocked.

8. A quality control device, characterized in that, include: The acquisition module is used to acquire data collection tasks; The inspection module is used to inspect the data before data acquisition based on the data acquisition task. If the inspection fails, the acquisition process is blocked. The inspection module is also used to monitor the collected data during the data acquisition process and provide real-time feedback to the operator or execute preset control logic based on the monitoring results. The inspection module is also used to verify the integrity and logical consistency of the collected data after the data collection is completed. The inspection module is also used to verify multiple associated data files for the same data acquisition task before data is uploaded; The processing module is used to delete the collected data if the cloud service confirms that the upload was successful after the data upload is completed.

9. A quality control device, comprising at least one processor and a memory; wherein, The memory stores computer-executed instructions; The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by a processor to perform the method of any one of claims 1 to 7.

Images