A detection task scheduling method and device

Abstract

The present invention discloses a method and device for scheduling detection tasks, the method comprising: updating the mode parameters corresponding to the best operation mode of each executor in the operation mode database; obtaining the mode parameters corresponding to the best operation mode of all executors performing the detection task from the operation mode database; obtaining the current active state of all executors performing the detection task, and determining the current execution state of each executor in combination with the mode parameters of all executors performing the detection task; matching the materials to be detected based on the current execution state of all executors, and scheduling the materials to be detected to realize the detection execution of the materials to be detected. The present invention significantly improves the work efficiency and work quality of the executors, while reducing the management cost and error rate, and realizes intelligent and personalized task scheduling management.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and in particular relates to a detection task scheduling method and device. Background Art

[0002] With the development of industrial automation and intelligent manufacturing, task scheduling and executor management in production workshops have become increasingly complex and important. Traditional task scheduling methods usually rely on fixed rules and executors' self-adjustment, which is difficult to adapt to complex production environments and individual differences, resulting in low production efficiency, uneven executor load, high fatigue and other problems.

[0003] At present, most factories adopt two main executor management methods: one is manual scheduling based on experience, and the other is automatic scheduling based on fixed algorithms. The former relies on the experience and intuition of the workshop manager, and it is difficult to achieve refined and personalized management; although the latter can achieve a certain degree of automation, due to the lack of consideration of individual differences of executors, it is often unable to give full play to the best working ability of each executor. For example, the patent application text CN105893263A discloses a test task scheduling method. The method includes: setting a certain number of test task queues, specifying the priority range of the test tasks that can be listed for each test task queue; adding the test task to the corresponding test task queue according to the priority range; when scheduling the test task, taking the test task with the highest priority among all the test tasks as the task to be switched; after the test task is scheduled, the new priority of all test tasks is obtained according to the current priority of the test task, the time of the last priority update, the time to be executed, and the priority weight of the test task queue, and adding them to the queue of the corresponding test task according to the new priority of each test task. Since the executor's operating mode involves multiple parameters such as work frequency, working hours, and rest time, traditional methods are difficult to accurately obtain and dynamically adjust, resulting in unreasonable task scheduling and affecting production efficiency and product quality.

[0004] The present invention aims to improve factory production efficiency, optimize executor workload, reduce fatigue, and improve product quality through intelligent and personalized task scheduling methods, so as to provide strong support for the development of intelligent manufacturing and Industry 4.0. Summary of the invention

[0005] In view of the defects existing in the above-mentioned prior art, the present invention provides a detection task scheduling method, the method comprising:

[0006] Update the mode parameters corresponding to the best operation mode of each executor in the operation mode database;

[0007] Acquire mode parameters corresponding to the best operation modes of all executors performing the detection task from the operation mode database;

[0008] Obtain the current active status of all executors of the detection task, and determine the current execution status of each executor in combination with the mode parameters of all executors of the detection task;

[0009] Based on the current execution status of all executors, the materials to be tested are matched and scheduled to realize the detection execution of the materials to be tested.

[0010] Wherein, updating the mode parameters corresponding to the best operation mode of each executor in the operation mode database includes:

[0011] Authentication of the executor who enters the corresponding work area of ​​the work site;

[0012] Based on the identity authentication of the executor and the work site where the executor is located, determining whether the operation mode database has the best operation mode for the work site corresponding to the executor;

[0013] Initiate unaware operation mode testing for executors that do not have optimal operation mode;

[0014] By analyzing the data collected in the imperceptible operation mode test, the best operation mode of the executor is identified, and the mode parameters corresponding to the best operation mode are stored in the operation mode database.

[0015] Among them, identity authentication is performed on the executor entering the work area corresponding to the work site, including: starting a face recognition process for the executor entering the work area; comparing the features extracted by the face recognition process with the pre-stored facial features of the executor to give a recognition result, and after the recognition is passed, the executor identity is associated with the work site where the executor is located.

[0016] Specifically, a camera is installed in each work area, and when the executor enters the work area, the system will start the face recognition process;

[0017] The extracted features are compared with the facial features of the performer stored in the database, and once the identity of the performer is confirmed, the identity of the performer is associated with the work site where the camera that captured the facial image of the performer is located.

[0018] The operation mode database includes an executor information table, a face recognition information table, a workstation information table, and an operation mode information table, among which:

[0019] The performer information table stores the identity information of each performer, including a unique performer ID, name, and an identifier associated with facial features;

[0020] The face recognition information table stores information related to face recognition, and each entry includes an executor ID and data associated with facial features;

[0021] The work site information table stores the work site information of each performer, and each entry includes the performer ID and the work site ID;

[0022] The operation mode information table stores the operation mode of each performer, and each entry will include a performer ID and mode parameter data associated with its optimal operation mode.

[0023] Among them, the unaware operation mode test is initiated for the executor who does not have the best operation mode, including:

[0024] Continuously monitor the performers in the work area and obtain the performers' video stream data;

[0025] The video stream data is divided into a number of image data according to the number of frames, and the actions and behaviors of the performer in each image data are identified;

[0026] According to the actions and behaviors of the performer in each image data, the optimal operation mode corresponding to the mode parameters of the performer is determined.

[0027] According to the actions and behaviors of the performer in each image data, the corresponding mode parameters of the best operation mode of the performer are determined, including:

[0028] According to the actions and behaviors of the performer in each image data, all the start time points and end time points of the performer's operations are determined;

[0029] Determine an operation time interval between two consecutive operations based on the timing of the video stream data;

[0030] All operation time intervals are arranged in chronological order to form time series data;

[0031] The time series data is analyzed and processed to extract the corresponding mode parameters of the executor's optimal operation mode.

[0032] The optimal operation mode parameters of each executor include: the fluctuation range of the operation time interval, the periodic change of the operation time interval, the initial offset of the operation time interval and the average operation time interval; specifically,

[0033] A i : The amplitude of the operation time interval of executor i, which indicates the fluctuation range of the operation time interval;

[0034] ω i : The operation time interval frequency of executor i, indicating the periodic change of the operation time interval;

[0035] : The operation time interval phase of executor i, indicating the initial offset of the operation time interval;

[0036] B i : The baseline operation time interval of executor i, which represents the average operation time interval.

[0037] Among them, based on the current execution status of all executors, the materials to be tested are matched and scheduled to realize the detection execution of the materials to be tested, including:

[0038] S41, obtaining a list of materials to be tested, and determining a set of executors corresponding to each material to be tested that has completed testing based on all executors who have performed the testing task;

[0039] S43, sorting all executors according to the current execution status of each executor;

[0040] S45, based on the list of materials to be tested, allocating the materials to be tested to the executor with the highest current execution status in the corresponding executor set for testing;

[0041] S47. Repeat steps S43-S45 until all materials to be tested have been prepared.

[0042] Wherein, based on the list of materials to be tested, the materials to be tested are allocated to the executor with the highest current execution status in the set of executors corresponding to the completed testing of the materials to be tested for testing, including:

[0043] S451, determining the materials to be tested that are currently to be deployed according to the order of the list of materials to be tested, and providing a set of executors corresponding to the completed testing of the materials to be tested that are currently to be deployed;

[0044] S453, combining the sorting results of all executors, selecting the executor with the highest current execution status in the executor set;

[0045] S455, obtaining the continuous working time, maximum working time and the completion time of the inspection of the materials to be inspected to be deployed currently of the executor with the highest current execution status in the executor set;

[0046] S457. When the continuous working time and the accumulation of the continuous working time and the completion inspection time do not exceed the maximum working time, the materials to be inspected currently to be allocated are allocated to the selected executor for inspection; otherwise, the selected executor is removed from the set of executors corresponding to the materials to be inspected currently to be allocated, and the process returns to step S453.

[0047] The present invention also discloses a detection task scheduling device, which adopts the above detection task scheduling method, and the device comprises:

[0048] A parameter updating module, which is used to update the mode parameters corresponding to the best operation mode of each executor in the operation mode database;

[0049] A parameter acquisition module, which is used to acquire mode parameters corresponding to the best operation modes of all executors performing the detection task from the operation mode database;

[0050] An execution determination module, which is used to obtain the current active status of all executors performing the detection task, and determine the current execution status of each executor in combination with the mode parameters of all executors performing the detection task;

[0051] The material scheduling module is used to match the materials to be tested based on the current execution status of all executors, and schedule the materials to be tested to realize the detection execution of the materials to be tested.

[0052] The present invention records the best operation mode of each executor through the operation mode database, realizes personalized task scheduling, ensures that each executor can work at the work intensity and rhythm that best suits him / her, and improves overall work efficiency. This dynamic scheduling and method of detection tasks significantly improves the work efficiency and work quality of the executors, reduces fatigue and error rate. In addition, the use of deep learning face recognition algorithm for executor identity authentication can effectively prevent identity fraud and misidentification, and improve the safety and management efficiency of the work area.

[0053] The present invention solves the problems of executor identity authentication, personalized task scheduling, and optimization of task intensity and work / rest intervals through deep learning face recognition algorithm, operation mode database and non-perceptual operation mode test, significantly improves the work efficiency and work quality of executors, reduces management costs and error rates, and realizes intelligent and personalized task scheduling management. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present disclosure will become readily understood. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0055] Figure 1 is a flow chart showing a method for scheduling detection tasks according to an embodiment of the present invention;

[0056] Figure 2 is a flow chart showing a method for scheduling detection tasks according to an embodiment of the present invention;

[0057] Figure 3 is a flow chart showing the completion of material detection according to an embodiment of the present invention;

[0058] Figure 4 is a flow chart showing the preparation of materials to be tested according to an embodiment of the present invention;

[0059] Figure 5 It is a schematic diagram showing a detection task scheduling device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technical executors in the field without creative work are within the scope of protection of the present invention.

[0061] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0062] It should be understood that although the terms first, second, third, etc. may be used to describe ... in the embodiments of the present invention, these ... should not be limited to these terms. These terms are only used to distinguish .... For example, without departing from the scope of the embodiments of the present invention, the first ... may also be referred to as the second ..., and similarly, the second ... may also be referred to as the first ....

[0063] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0064] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0065] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0066] In power detection, different executors have different operating habits and work efficiency. Reasonable arrangement of task intensity and work / rest intervals can significantly improve the work efficiency and satisfaction of executors, but traditional methods are difficult to dynamically adjust and optimize. Traditional task scheduling methods are difficult to adapt to the individual differences of each executor, resulting in low work efficiency. The present invention can automatically identify and record the operation mode of each executor through the operation mode database and non-perceptual operation mode test, thereby improving the personalization and effectiveness of task scheduling.

[0067] like Figure 1 As shown, the present invention discloses a detection task scheduling method, the method comprising:

[0068] Update the mode parameters corresponding to the best operation mode of each executor in the operation mode database;

[0069] Acquire mode parameters corresponding to the best operation modes of all executors performing the detection task from the operation mode database;

[0070] Obtain the current active status of all executors of the detection task, and determine the current execution status of each executor in combination with the mode parameters of all executors of the detection task;

[0071] Based on the current execution status of all executors, the materials to be tested are matched and scheduled to realize the detection execution of the materials to be tested.

[0072] Wherein, updating the mode parameters corresponding to the best operation mode of each executor in the operation mode database includes:

[0073] Authentication of the executor who enters the corresponding work area of ​​the work site;

[0074] Based on the identity authentication of the executor and the work site where the executor is located, determining whether the operation mode database has the best operation mode for the work site corresponding to the executor;

[0075] Initiate unaware operation mode testing for executors that do not have optimal operation mode;

[0076] By analyzing the data collected in the imperceptible operation mode test, the best operation mode of the executor is identified, and the mode parameters corresponding to the best operation mode are stored in the operation mode database.

[0077] Specifically, Figure 2 As shown, when the executor enters the workspace, a deep learning face recognition algorithm is used to authenticate the identity, and the authentication result will be associated with the specific work station.

[0078] The system will query the operation mode database for the current executor's operation mode. If there is a record, the task will be scheduled directly. If there is no record, the non-perceived operation mode test will be started.

[0079] In the non-perceptual operation mode test, during the normal work of the executor, the system collects work data within the test period through the camera, including operation frequency, operation duration, and rest time, and records the mode parameters of the executor based on the work data.

[0080] During the test phase of the operation mode, the system gradually adjusts the intensity of the work tasks and the work / rest intervals, while recording the work efficiency and work quality of the performer.

[0081] Among them, in the operation mode test, the system schedules tasks of different intensities to the executors at different time points, or changes the ratio of work and rest time.

[0082] By analyzing the collected data, the system can identify the best operating mode of the executor and save the mode parameters corresponding to the best operating mode in the operating mode database.

[0083] Task scheduling is performed based on the corresponding mode parameters of the executor in the operation mode database.

[0084] It can be seen that the present invention records the best operation mode of each executor through the operation mode database, realizes personalized task scheduling, ensures that each executor can work at the work intensity and rhythm that best suits him / her, and improves overall work efficiency. This dynamic scheduling and method of detection tasks significantly improves the work efficiency and work quality of the executors, reduces fatigue and error rate. In addition, the use of deep learning face recognition algorithm for executor identity authentication can effectively prevent identity fraud and misidentification, and improve the safety and management efficiency of the work area.

[0085] Optionally, when the performer enters the work area, the system will start the face recognition process. Authentication of the performer who enters the work area corresponding to the work site includes: starting the face recognition process for the performer who enters the work area; comparing the features extracted by the face recognition process with the pre-stored facial features of the performer to give a recognition result, and after the recognition is passed, associating the performer identity with the work site where the performer is located.

[0086] In actual application scenarios, a camera is installed in each workstation area, and the facial image of the executor is captured by the camera when the executor enters the work area. To ensure image quality, the contrast and brightness of the image need to be adjusted, or it needs to be cropped to include only the facial area.

[0087] The system will use a pre-trained deep learning model, the Deep Residual Network (ResNet) algorithm, to process the captured images.

[0088] The system compares the extracted features with the facial features of the performer stored in the database. If a match is found, the system can confirm the identity of the performer.

[0089] Once the performer is identified, the system associates that performer with the workstation where the camera that captured the performer's facial image is located, indicating that the system now knows which performer operates that specific workstation.

[0090] When a performer enters a work area for the first time or when a performer is registered, their facial image is recorded in an identity database, which will be used for subsequent identity verification and work site association.

[0091] When the executor enters the workspace again, the facial recognition system will recognize their facial features and match them with the records in the database to verify their identity.

[0092] Once the performer's identity is confirmed, the system retrieves their record in the database that associates them with a workstation. The association can be static (e.g., each performer has a fixed workstation) or dynamic (e.g., a performer's workstation may change based on the decision of the scheduling system).

[0093] When a dynamic association relationship occurs and the executor's workstation changes, the system needs to update the association information between the executor and the workstation in the database.

[0094] When querying the operation mode, the system will access the operation mode database and search for the record of the current executor. If the executor's operation mode has been recorded, this information will be used for direct task scheduling.

[0095] If the operator's operation mode record is not found in the database, the system will start the unconscious operation mode test. This test process is carried out during the operator's normal work, which means that the operator will not be aware of the test. During this process, the system will collect various parameters, including the operator's operation frequency, operation duration, rest time, and evaluate the operation intensity.

[0096] As the test progresses, the system may gradually adjust the intensity of the work tasks and the work / rest intervals to find the best work pattern for the performer. This process may involve scheduling tasks of different intensities to the performer at different times within a preset time period, or changing the ratio of work and rest time.

[0097] After the test process is over, the system will analyze the collected data, identify the best operating mode for the performer, and save the parameters corresponding to this mode in the operating mode database. In this way, the next time the performer enters the workplace, the system will immediately know their best operating mode, thereby performing effective task scheduling.

[0098] The system database includes an executor information table, a face recognition information table, a workstation information table, and an operation mode information table, wherein:

[0099] Performer Information Table: This table will store the identity information of each performer, including their unique performer ID, name, and an identifier associated with their facial features.

[0100] Face Recognition Information Table: This table will store information related to face recognition. Each entry will include the performer ID and data associated with their facial features.

[0101] Worksite information table: This table will store the worksite information of each performer. Each entry will include performer ID and worksite ID.

[0102] Operation Mode Information Table: This table will store the operation mode of each performer. Each entry will include the performer ID and data associated with their operation mode.

[0103] Among them, the index performer ID can be used to connect to other performer feature information. Therefore, these tables cannot query each other, ensuring the independence and security of information.

[0104] Optionally, sensitive data (such as facial features and operation mode data) is encrypted to prevent unauthorized access, and only authenticated and authorized users are allowed to access specific information.

[0105] Initiate unaware operation mode testing for performers that do not have optimal operation mode, including:

[0106] Continuously monitor the performers in the work area and obtain the performers' video stream data;

[0107] The video stream data is divided into a number of image data according to the number of frames, and the actions and behaviors of the performer in each image data are identified;

[0108] According to the actions and behaviors of the performer in each image data, the best operation mode corresponding to the mode parameters of the performer is determined, including:

[0109] According to the actions and behaviors of the performer in each image data, all the start time points and end time points of the performer's operations are determined;

[0110] Determine an operation time interval between two consecutive operations based on the timing of the video stream data;

[0111] All operation time intervals are arranged in chronological order to form time series data;

[0112] The time series data is analyzed and processed to extract the corresponding mode parameters of the executor's optimal operation mode.

[0113] In actual application scenarios, cameras are used to collect the performer’s work data and record mode parameters, including:

[0114] Install high-resolution cameras in the performer's work area. Configure the cameras to continuously monitor and record the performer's activities.

[0115] The camera continuously captures images of the performer's activities. Video processing software is used to segment the live video stream into frames and analyze each frame. Computer vision and deep learning algorithms (such as convolutional neural networks, CNNs) are used to identify the performer's movements and behaviors in each frame. These algorithms can be pre-trained to recognize specific operating behaviors, such as operating a machine, manually handling materials, taking a break, etc.

[0116] Count the number of times the actor performs a specific detection operation (related to the corresponding detection test step) within a specific time period. For example, use repeated actions detected in image frames to count the operation frequency, including:

[0117] Each frame is processed to detect the time points (frames) when the performer starts and ends the operation.

[0118] Mark the start and end time of each operation.

[0119] Calculates the time interval between two consecutive operations.

[0120] Assume The end time of this operation is , No. The operation started at .

[0121] Operation time interval .

[0122] Time series construction, all operations are separated by time intervals Stored in chronological order, forming time series data.

[0123] Record the time it takes for the performer to perform an action from the beginning to the end. For example, the operation duration can be calculated by detecting the start frame and the end frame of the action.

[0124] Identify time periods when the performer is not performing any operation. Through image analysis, detect frames where the performer stops performing any operation and calculate the total duration of these time periods.

[0125] The identified operation frequency, operation duration, rest time and other data are stored in the database.

[0126] By analyzing the collected data, the operating mode parameters of the performers are extracted.

[0127] By analyzing the collected data, the parameters of the operator's operation mode are extracted, including the calculation of the following statistical characteristics based on time series analysis:

[0128] Average operation time interval ;

[0129] Standard deviation of the operating time interval ;

[0130] Operation density: the reciprocal of operation frequency, that is, the number of operations per time unit.

[0131] Afterwards, a sine function is fitted based on the sine function model.

[0132] Sine function model: ;

[0133] in:

[0134] It's time The operation time interval on.

[0135] is the amplitude, which indicates the range of variation of the operating time interval.

[0136] is the angular frequency, which represents the periodicity (frequency) of the operating time interval.

[0137] is the phase, indicating the starting point of the cycle.

[0138] is the offset, representing the average operation time interval.

[0139] Initial parameter estimates: Amplitude : Estimated as half the difference between the maximum and minimum values ​​of the operating time interval data: .

[0140] Angular frequency : Estimate the main frequency components of the data through fast Fourier transform (FFT). ,in, It is the main cycle in the data, that is, the large cycle of the operating frequency adjustment. The larger T is, the lower the perception of the operating frequency adjustment is, and the smaller T is, the higher the perception of the operating frequency adjustment is.

[0141] Phase : It can be estimated by observing the periodic changes in the data. Usually, the phase The value of is 0.

[0142] Offset : estimated as the mean of the operating time interval data, ,when Greater than offset When , it indicates that the current operation intensity is low. Less than offset , it indicates that the current operation intensity is high.

[0143] In addition, nonlinear least squares methods (such as Levenberg-Marquardt algorithm) can be used to optimize the fitting parameters to minimize the error between the fitting function and the actual operation time interval data. In the fitting model verification, a goodness of fit test is required.

[0144] Among them, based on the calculation of the coefficient of determination ( value) to assess the explanatory power of the fitted model, The closer the value is to 1, the better the model fit is: ,in, is the operation time interval predicted by the model, is the average value of the operation time interval. Analyze the residuals of the fitted model (the difference between the actual data and the fitted data) and check the distribution and trend of the residuals to verify the applicability of the model.

[0145] Through the above method, the operation time interval data of the executor can be systematically collected and analyzed, and the sine function model can be used for fitting. The model parameters after fitting reflect the amplitude, periodicity and average level of the operation time interval, which can be used for further operation optimization and task scheduling.

[0146] The operation mode test phase includes initial setting of initial parameters, specifically, predicting the operator's operation time interval using a sine function model based on previously collected data.

[0147] ;

[0148] Task intensity and rest time ratio: Set different task intensities (low, medium, high) and different work / rest time ratios (for example, 25:5, 50:10, 75:15 minutes). Different task intensities correspond to different operation time intervals.

[0149] Data collection - record executor performance: record the executor's work efficiency (the number of tasks completed per unit time) and work quality (the accuracy of task completion) at different task intensities and work / rest time ratios.

[0150] Then, the operation mode is optimized to optimize the work efficiency under the premise that the work quality (accuracy) meets the requirements. Define the objective function, which is Finally, the main focus is on optimizing work efficiency.

[0151] Assume the following parameters:

[0152] is work efficiency (the number of tasks completed per unit time);

[0153] For work quality (accuracy of task completion);

[0154] Minimum requirements for work quality;

[0155] Then the objective function is defined as:

[0156] Objective Function

[0157] That is, when the quality of work Meet or exceed When the objective function is equal to the work efficiency ; Otherwise, the objective function value is negative infinity.

[0158] Optionally, the best mode parameters are quickly found based on preliminary optimization and refined optimization.

[0159] Among them, the optimal interval is quickly found based on preliminary optimization, including the use of optimization algorithms to quickly search for the initial optimal interval of task intensity and work / rest time ratio. The goal is to maximize work efficiency while ensuring that the work quality is not lower than

[0160] In the initial optimal range, more detailed parameter adjustments are made, and local search algorithms are used to further optimize task intensity and work / rest time ratio.

[0161] Determine the best operating mode under different task intensities and work / rest time ratios, and extract and record the corresponding model parameters

[0162] Optionally, a genetic algorithm (GA) is used for preliminary optimization. Genetic algorithm is a search algorithm based on natural selection and genetic mechanism, which is applicable to global optimization problems and specifically includes the following processes:

[0163] Randomly generate several combinations (chromosomes) of task intensity and work / rest time ratio.

[0164] Fitness function: Calculate the fitness value of each combination (chromosome), that is, the objective function value.

[0165] The better combination is selected for reproduction according to the fitness value.

[0166] Perform a crossover operation on the selected chromosomes to generate new combinations.

[0167] Mutate some chromosomes to increase population diversity.

[0168] Fitness calculation, selection, crossover, and mutation are repeated until a predetermined number of iterations is reached or the fitness value converges.

[0169] Output the preliminary optimization results, that is, output the best combination of task intensity and work / rest time ratio found by the genetic algorithm.

[0170] Optionally, based on the preliminary optimization, more detailed parameter adjustments are performed, including setting a fine adjustment range.

[0171] A fine-tuning range is set around the best combination obtained by preliminary optimization, for example, the task intensity is within ±10% and the work / rest time ratio is within ±5 minutes.

[0172] Use the simulated annealing algorithm SA for fine tuning. The simulated annealing algorithm is a global optimization algorithm based on the principles of thermodynamics, suitable for the combination of local search and global search, including:

[0173] The best combination obtained by preliminary optimization is selected as the initial solution.

[0174] Set initial temperature and cooling rate .

[0175] Randomly generate a new solution within the fine-tuning range of the current solution.

[0176] Calculate the objective function value of the new solution.

[0177] The decision on whether to accept the new solution is based on the following criteria:

[0178]

[0179] Cooling down: Gradually lower the temperature .

[0180] When the temperature drops to a certain threshold or the number of iterations reaches a predetermined value, the iteration is stopped and the operation mode at this time is the optimal operation mode.

[0181] Among them, the optimal operation mode parameters for each executor include:

[0182] : Executor The operating time interval amplitude.

[0183] : Executor The frequency of the operating interval.

[0184] : Executor The operating time interval phase.

[0185] : Executor The baseline operation time interval.

[0186] like Figure 3 As shown, based on the current execution status of all executors, the materials to be tested are matched and scheduled to be tested, so as to implement the testing execution of the materials to be tested, including:

[0187] S41, obtaining a list of materials to be tested, and determining a set of executors corresponding to each material to be tested that has completed testing based on all executors who have performed the testing task;

[0188] S43, sorting all executors according to the current execution status of each executor;

[0189] S45, based on the list of materials to be tested, allocating the materials to be tested to the executor with the highest current execution status in the set of executors corresponding to the completed tests of the materials to be tested for testing;

[0190] S47. Repeat steps S43-S45 until all materials to be tested have been prepared.

[0191] like Figure 4As shown, in step S45, based on the list of materials to be tested, the materials to be tested are allocated to the executor with the highest current execution status in the set of executors corresponding to the completed testing of the materials to be tested for testing, which may include:

[0192] S451, determining the materials to be tested that are currently to be deployed according to the order of the list of materials to be tested, and providing a set of executors corresponding to the completed testing of the materials to be tested that are currently to be deployed;

[0193] S453, combining the sorting results of all executors, selecting the executor with the highest current execution status in the executor set;

[0194] S455, obtaining the continuous working time, maximum working time and the completion time of the inspection of the materials to be inspected to be deployed currently of the executor with the highest current execution status in the executor set;

[0195] S457. When the continuous working time and the accumulation of the continuous working time and the completion inspection time do not exceed the maximum working time, the materials to be inspected currently to be allocated are allocated to the selected executor for inspection; otherwise, the selected executor is removed from the set of executors corresponding to the materials to be inspected currently to be allocated, and the process returns to step S453.

[0196] In actual application scenarios, when there is a new list of materials to be tested, the system will schedule tasks according to the executor's mode parameters. Materials to be tested and The detection task scheduling process is as follows:

[0197] Step 1, read the executor mode parameters: read the mode parameters of all executors from the operation mode database.

[0198] Step 2. List of materials to be tested: Get the current list of materials to be tested.

[0199] Step 3. Get the executor status: Get the status of each executor, including the current number of tasks, operation time interval, rest time, etc.

[0200] Step 4. Task scheduling loop:

[0201] Sorting executors: Sort the executors according to their current execution status, and give priority to executors with higher current execution status.

[0202] Scheduling tasks: dispatch the materials to be tested to the sorted executors in sequence.

[0203] Update executor status: After each task is scheduled, update the current number of tasks of the executor.

[0204] Optionally, adjust the task scheduling based on the executor's accumulated working hours. For example, if the operation time interval of an executor is close to its maximum working time (e.g., in a 75:15 work / rest time distribution, the executor is about to work continuously for 75 units of time), consider scheduling it to other executors or arranging a rest for the executor.

[0205] Step 5. End of loop: When all materials to be tested have been scheduled, the task scheduling loop ends.

[0206] Indicates the executor In time The operation time interval is a time interval model based on the sine function. Although it is in the form of a sine function, it represents the operation time interval of each executor at different time points, rather than the direct work intensity. It can reflect the changes in the work pace of the executor rather than the direct work intensity.

[0207] Each workstation has a display unit that can display the current work intensity. Optionally, the selection function The operation time interval can be Work intensity index converted into success To express it, we can define the work intensity index The reciprocal of the operation time interval: work intensity index , indicating that the smaller the operation time interval, the greater the work intensity index The higher it is, the higher the intensity of the current work.

[0208] The display unit can show the work intensity index Display, the executor can see the current work intensity index If the executor thinks that the work intensity needs to be adjusted, he can slide on the interactive display unit to adjust it. After receiving the executor's feedback on the work intensity adjustment, the system can adjust the work intensity based on the work intensity index. Modify the task scheduling plan, that is, based on the modified work intensity index Determine the corresponding operation time interval , based on this operation time interval Instead of the operation time interval stored in the database Perform task scheduling. After the executor actively adjusts the operation time interval, the operation time interval The current detection cycle can be kept as a fixed value until the detection task of the current time period is completed, that is, short-term temporary feedback will not affect the executor mode parameters stored in the database.

[0209] Among them, the executors are sorted according to their current execution status, and the executors with higher current execution status are preferentially scheduled, including task scheduling based on the following process:

[0210] First define the following parameter definitions:

[0211] Executor model parameters, including:

[0212] : Executor The operation time interval amplitude indicates the fluctuation range of the operation time interval.

[0213] : Executor The operation time interval frequency indicates the periodic change of the operation time interval and the specific situation of the work / rest time allocation.

[0214] : Executor The operation time interval phase indicates the initial offset of the operation time interval and is generally set to 0.

[0215] : Executor The baseline operation time interval of , which represents the average operation time interval.

[0216] Executor status parameters, including:

[0217] : Executor The current number of tasks.

[0218] Task scheduling related parameters, including:

[0219] : Current time.

[0220] : Executor In time The operation time interval.

[0221] : Executor In time The current execution status of .

[0222] : Executor The expected workload in a future time period.

[0223] Current execution status The calculation formula is:

[0224] ,

[0225] Among them, the function Used to adjust the impact of future workloads on the current execution state.

[0226] Predicting future workload,The future workload can be calculated by predicting the operation time interval in the future. Suppose we predict the future time period , then the future workload can be expressed as:

[0227] ;

[0228] in, ,

[0229] Based on the following adjustment function Reflecting the impact of future workloads:

[0230] ;

[0231] When the future workload is large, will be significantly reduced, reflecting the future work pressure of the executors.

[0232] Based on the above parameter definitions, determine the task scheduling of testing materials, including:

[0233] Get the status of each executor, including the current number of tasks, operation time interval, rest time, etc.

[0234] The task scheduling cycle includes sorting the executors, selecting the appropriate executor, and finally updating the executor status. Among them, selecting the executor means selecting the inspection station corresponding to the executor, and delivering the inspection materials to the corresponding inspection station through the AGV equipment.

[0235] Among them, the executors are sorted according to their current execution status, and priority is given to executors with higher current execution status.

[0236] Among them, the materials to be tested are dispatched to the sorted executors in turn, including:

[0237] Calculate the current working energy: ,

[0238] Select the appropriate executor: Traverse the list of materials to be tested and schedule tasks according to the current execution status of the testing process that handles the same step. For each material to be tested, select the executor with the highest current execution status for scheduling.

[0239] Update executor status: After each task is scheduled, update the current number of tasks for all executors.

[0240] Update the executor status, including: For the executor scheduled to the task, update the current number of tasks: .

[0241] The present invention solves the problems of executor identity authentication, personalized task scheduling, and optimization of task intensity and work / rest intervals through deep learning face recognition algorithm, operation mode database and non-perceptual operation mode test, significantly improves the work efficiency and work quality of executors, reduces management costs and error rates, and realizes intelligent and personalized task scheduling management.

[0242] like Figure 5 As shown, the present invention also provides a detection task scheduling device, which adopts the above detection task scheduling method, and the device includes:

[0243] A parameter updating module, which is used to update the mode parameters corresponding to the best operation mode of each executor in the operation mode database;

[0244] A parameter acquisition module, which is used to acquire mode parameters corresponding to the best operation modes of all executors performing the detection task from the operation mode database;

[0245] An execution determination module, which is used to obtain the current active status of all executors performing the detection task, and determine the current execution status of each executor in combination with the mode parameters of all executors performing the detection task;

[0246] The material scheduling module is used to match the materials to be tested based on the current execution status of all executors, and schedule the materials to be tested to realize the detection execution of the materials to be tested.

[0247] The present invention also provides a detection task scheduling system, the system comprising a memory and a processor, the memory storing a computer program, and the processor being used to execute the computer program in the memory to implement the above detection task scheduling method.

[0248] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0249] The computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0250] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0251] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0252] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0253] The above introduces the preferred embodiments of the present invention, which is intended to make the spirit of the present invention clearer and easier to understand, but is not intended to limit the present invention. All modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection outlined by the claims attached to the present invention.

Claims

1. A detection task scheduling method, characterized in that: The method comprises: Update the mode parameters corresponding to the best operation mode of each executor in the operation mode database; Acquire mode parameters corresponding to the best operation modes of all executors performing the detection task from the operation mode database; Obtain the current active status of all executors of the detection task, and determine the current execution status of each executor in combination with the mode parameters of all executors of the detection task; Based on the current execution status of all executors, the materials to be tested are matched and scheduled to implement the testing execution of the materials to be tested, including: S41, obtaining a list of materials to be tested, and determining a set of executors corresponding to each material to be tested that has completed testing based on all executors who have performed the testing task; S43, sorting all executors according to the current execution status of each executor; S45, based on the list of materials to be tested, allocating the materials to be tested to the executor with the highest current execution status in the corresponding executor set for testing; S47, repeat steps S43-S45 until all materials to be tested have been prepared; Among them, the current execution status of each executor satisfies the following relationship: ； In the formula, C i (t) represents the current execution status of the i-th executor, tasks i is the current number of detection tasks of the i-th executor, ΔT i (t) is the operation time interval of the ith executor at time t, g() is the adjustment function, is the future workload of the ith executor in the future time interval T, A i is the fluctuation range of the operation time interval of the i-th executor, ω i is the periodic change of the operation time interval of the i-th executor, is the initial offset of the operation time interval of the i-th executor, B i is the average operation time interval of the i-th executor.

2. The detection task scheduling method according to claim 1, characterized in that: Update the mode parameters corresponding to the best operation mode of each executor in the operation mode database, including: Authentication of the executor who enters the corresponding work area of ​​the work site; Based on the identity authentication of the executor and the work site where the executor is located, determining whether the operation mode database has the best operation mode for the work site corresponding to the executor; Initiate unaware operation mode testing for executors that do not have optimal operation mode; By analyzing the data collected in the imperceptible operation mode test, the best operation mode of the executor is identified, and the mode parameters corresponding to the best operation mode are stored in the operation mode database.

3. The detection task scheduling method according to claim 2, characterized in that: Authentication is performed on the executor who enters the work area corresponding to the work site, including: starting a face recognition process for the executor who enters the work area; comparing the features extracted by the face recognition process with the pre-stored facial features of the executor to give a recognition result, and after the recognition is passed, the identity of the executor is associated with the work site where the executor is located.

4. The detection task scheduling method according to claim 1, characterized in that: The operation mode database includes the executor information table, face recognition information table, workstation information table, and operation mode information table, among which: The performer information table stores the identity information of each performer, including a unique performer ID, name, and an identifier associated with facial features; The face recognition information table stores information related to face recognition, and each entry includes an executor ID and data associated with facial features; The work site information table stores the work site information of each performer, and each entry includes the performer ID and the work site ID; The operation mode information table stores the operation mode of each performer, and each entry will include a performer ID and mode parameter data associated with its optimal operation mode.

5. The detection task scheduling method according to claim 2, characterized in that: Initiate unaware operation mode testing for performers that do not have optimal operation mode, including: Continuously monitor the performers in the work area and obtain the performers' video stream data; The video stream data is divided into a number of image data according to the number of frames, and the actions and behaviors of the performer in each image data are identified; According to the actions and behaviors of the performer in each image data, the optimal operation mode corresponding to the mode parameters of the performer is determined.

6. The detection task scheduling method according to claim 5, characterized in that: According to the actions and behaviors of the performer in each image data, the best operation mode corresponding to the mode parameters of the performer is determined, including: According to the actions and behaviors of the performer in each image data, all the start time points and end time points of the performer's operations are determined; Determine an operation time interval between two consecutive operations based on the timing of the video stream data; All operation time intervals are arranged in chronological order to form time series data; The time series data is analyzed and processed to extract the corresponding mode parameters of the executor's optimal operation mode.

7. The detection task scheduling method according to claim 1, characterized in that: Based on the list of materials to be tested, the materials to be tested are allocated to the executors with the highest current execution status in their corresponding executor set, including: S451, determining the materials to be tested that are currently to be deployed according to the order of the list of materials to be tested, and providing a set of executors corresponding to the completed testing of the materials to be tested that are currently to be deployed; S453, combining the sorting results of all executors, selecting the executor with the highest current execution status in the executor set; S455, obtaining the continuous working time, maximum working time and the completion time of the inspection of the materials to be inspected to be deployed currently of the executor with the highest current execution status in the executor set; S457. When the continuous working time and the accumulation of the continuous working time and the completion inspection time do not exceed the maximum working time, the materials to be inspected currently to be allocated are allocated to the selected executor for inspection; otherwise, the selected executor is removed from the set of executors corresponding to the materials to be inspected currently to be allocated, and the process returns to step S453.

8. A detection task scheduling device, characterized in that: According to any one of claims 1 to 7, the detection task scheduling method is adopted, wherein the device comprises: A parameter updating module, which is used to update the mode parameters corresponding to the best operation mode of each executor in the operation mode database; A parameter acquisition module, which is used to acquire mode parameters corresponding to the best operation modes of all executors performing the detection task from the operation mode database; An execution determination module, which is used to obtain the current active status of all executors performing the detection task, and determine the current execution status of each executor in combination with the mode parameters of all executors performing the detection task; The material scheduling module is used to match the materials to be tested based on the current execution status of all executors, and schedule the materials to be tested to realize the detection execution of the materials to be tested.

Images