Robot cooperation task allocation method and system

Abstract

The invention relates to a robot cooperative task allocation method and system, and relates to the field of task allocation optimization, and the method comprises the steps: carrying out basic monitoring, obtaining an implicit abnormal position when an abnormality occurs, and carrying out the prediction to obtain a plurality of implicit abnormal angles and a plurality of angle probabilities; generating a plurality of adaptation coefficients and a plurality of cost coefficients according to the plurality of angle probabilities; and a plurality of real-time positions and a plurality of real-time angles of the plurality of monitoring robots are obtained, task optimization distribution is performed according to the hidden abnormal positions, the plurality of adaptation coefficients and the plurality of cost coefficients according to the sequence of the probabilities of the plurality of angles from large to small, and a plurality of monitoring tasks are obtained for verification monitoring. According to the invention, the technical problems of low monitoring efficiency, insufficient accuracy and waste of monitoring resources due to the fact that a traditional method cannot accurately allocate inspection tasks according to actual monitoring requirements and the current state of the robot can be solved; the scheduling cost can be effectively reduced, the accuracy and efficiency of anomaly detection can be improved, and efficient and accurate factory inspection can be realized.

Description

Technical Field

[0001] The present invention relates to the field of task allocation optimization, and in particular to a robot collaborative task allocation method and system. Background Art

[0002] With the rapid development of industrial automation technology, inspection robots, as an important tool for automated production management and safety monitoring, have gradually been widely used in various industrial fields, such as petroleum refining, steel metallurgy, chemical plants, power plants, etc. In these complex and large-scale industrial plants, the application of inspection robots helps to realize automated and intelligent equipment monitoring and fault diagnosis, greatly improving work efficiency and safety.

[0003] However, although modern inspection robots have achieved significant improvements in functionality and performance, they still face many challenges in collaborative work and task allocation. Especially in multi-robot collaborative mode, how to efficiently and accurately allocate inspection tasks remains a problem that needs to be solved. Summary of the invention

[0004] The present invention provides a robot collaborative task allocation method and system to solve the technical problems that traditional inspection robots cannot accurately allocate inspection tasks according to actual monitoring needs and the current status of the robots when collaborating, and there are low monitoring efficiency, insufficient accuracy and waste of monitoring resources.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides a robot collaborative task allocation method, comprising: performing basic monitoring of a target space by multiple monitoring robots, and when an abnormality occurs, obtaining an implicit abnormal position, predicting an abnormal angle, and obtaining multiple implicit abnormal angles and multiple angle probabilities; generating multiple adaptation coefficients and multiple cost coefficients for the multiple implicit abnormal angles according to the multiple angle probabilities; obtaining multiple real-time positions and multiple real-time angles of the multiple monitoring robots, and optimizing the allocation of monitoring tasks for the multiple implicit abnormal angles according to the implicit abnormal positions, multiple adaptation coefficients and multiple cost coefficients in order of the multiple angle probabilities from large to small, to obtain multiple monitoring tasks, and controlling the multiple monitoring robots to perform verification monitoring.

[0006] Preferably, the robot collaborative task allocation method also includes: performing basic monitoring of the target space by multiple monitoring robots, and when an abnormality occurs, obtaining the location coordinates of the abnormality as an implicit abnormal location; predicting the abnormal angle based on the implicit abnormal location to obtain multiple implicit abnormal angles and multiple angle probabilities.

[0007] Preferably, the robot collaborative task allocation method also includes: collecting abnormal angle prediction data based on historical monitoring abnormal data in the target space; using the abnormal angle prediction data to construct an abnormal angle prediction table, wherein the abnormal angle prediction table includes a mapping relationship between abnormal positions, abnormal angles and abnormal angle probabilities; inputting the implicit abnormal position into the abnormal angle prediction table, and mapping output to obtain multiple implicit abnormal angles and multiple angle probabilities.

[0008] Preferably, the robot collaborative task allocation method also includes: collecting a sample abnormal position set based on the historical monitoring abnormality data in the target space; collecting the angle at which the abnormality occurs at each sample abnormal position to obtain multiple sample abnormal angle sets, and collecting the proportion of each sample abnormal angle to obtain multiple sample angle probability sets; integrating the sample abnormal position set, multiple sample abnormal angle sets and multiple sample angle probability sets to obtain abnormal angle prediction data.

[0009] Preferably, the robot collaborative task allocation method further includes: using the multiple angle probabilities as multiple adaptation coefficients of the multiple implicit abnormal angles; and subtracting the multiple adaptation coefficients from 1 to obtain multiple cost coefficients.

[0010] Preferably, the robot collaborative task allocation method also includes: obtaining multiple real-time positions and multiple real-time angles of the multiple monitoring robots; selecting a first implicit abnormal angle in order of the probabilities of the multiple angles from large to small, and generating a first verification monitoring angle relative to the first implicit abnormal angle; performing task optimization allocation based on the first verification monitoring angle according to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, and obtaining a first monitoring task; continuing to optimize the monitoring task allocation for other multiple implicit abnormal angles in order of the probabilities of the multiple angles from large to small, obtaining multiple monitoring tasks, and controlling the multiple monitoring robots to perform verification monitoring.

[0011] Preferably, the robot collaborative task allocation method further comprises: constructing a first task allocation optimization function according to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, and the first verification monitoring angle, as shown in the following formula: ; Among them, FITM is the task assignment score, is the first adaptation coefficient, is the first cost coefficient, is the real-time angle of the robot selected in the optimization, For the first verification monitoring angle, is the deviation between the robot's real-time angle and the first verification monitoring angle, is the distance between the real-time position of the robot selected in the optimization and the implicit abnormal position, is a preset distance; among the multiple monitoring robots, a first monitoring robot is randomly selected; according to the first real-time position and the first real-time angle of the first monitoring robot, combined with the first verification monitoring angle and the implicit abnormal position, a first task allocation score is calculated based on the first task allocation optimization function; continue to traverse and select monitoring robots and calculate task scores, select the monitoring robot with the largest task allocation score, perform verification monitoring of the first implicit abnormal angle, and obtain the first monitoring task.

[0012] In the second aspect, the present invention provides a robot collaborative task allocation system, including: an implicit abnormality information acquisition module, which is used to perform basic monitoring of the target space through multiple monitoring robots, and when an abnormality occurs, obtain the implicit abnormality position, predict the abnormal angle, and obtain multiple implicit abnormality angles and multiple angle probabilities; a characteristic coefficient analysis module, which is used to generate multiple adaptation coefficients and multiple cost coefficients for the multiple implicit abnormality angles according to the multiple angle probabilities; a monitoring task optimization allocation module, which is used to obtain multiple real-time positions and multiple real-time angles of the multiple monitoring robots, and optimize the monitoring task allocation for the multiple implicit abnormality angles according to the implicit abnormality positions, multiple adaptation coefficients and multiple cost coefficients in the order of the multiple angle probabilities from large to small, to obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

[0013] The beneficial effects of the present invention are as follows: basic monitoring is performed on the target space by multiple monitoring robots, and when an abnormality occurs, the implicit abnormal position is obtained, the abnormal angle is predicted, and multiple implicit abnormal angles and multiple angle probabilities are obtained; then, according to the multiple angle probabilities, multiple adaptation coefficients and multiple cost coefficients of the multiple implicit abnormal angles are respectively generated; then, multiple real-time positions and multiple real-time angles of the multiple monitoring robots are obtained, and in the order of the multiple angle probabilities from large to small, the multiple implicit abnormal angles are optimally allocated monitoring tasks according to the implicit abnormal position, multiple adaptation coefficients and multiple cost coefficients, and multiple monitoring tasks are obtained; finally, according to the multiple monitoring tasks, the multiple monitoring robots are controlled to perform verification monitoring; that is, by optimizing the robot task allocation, the matching degree between the inspection task and the actual monitoring demand and the current state of the robot can be improved, thereby improving the scientificity and accuracy of the inspection task allocation, ensuring that high-risk areas are adequately monitored, and reducing invalid or inefficient resource waste, thereby effectively reducing scheduling costs and improving the accuracy and efficiency of abnormality detection, and realizing efficient and accurate factory inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1A schematic diagram of a flow chart of a robot collaborative task allocation method provided by the present invention; Figure 2 A schematic diagram of the structure of a robot collaborative task allocation system provided by the present invention.

[0015] In the accompanying drawings, the components represented by the reference numerals are described as follows: Implicit abnormal information acquisition module 10, characteristic coefficient analysis module 20, monitoring task optimization allocation module 30. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0017] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0018] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0019] Embodiment 1, as Figure 1 As shown, an embodiment of the present invention provides a robot collaborative task allocation method, which specifically includes the following steps: S100: Perform basic monitoring in the target space through multiple monitoring robots. When an abnormality occurs, obtain the implicit abnormal position, predict the abnormal angle, and obtain multiple implicit abnormal angles and multiple angle probabilities.

[0020] Furthermore, step S100 of the present invention further includes: S110: Perform basic monitoring in the target space by using multiple monitoring robots, and when an abnormality occurs, obtain the coordinates of the location where the abnormality occurs as an implicit abnormal location.

[0021] Specifically, the present invention relates to factory inspections through inspection robots, for example, automated monitoring and anomaly detection tasks in high-risk, complex environments such as oil refining production areas and steel and metallurgical plants. In these complex industrial environments, there are many production equipment and complex operating conditions. Manual inspections not only pose safety hazards but are also inefficient. Therefore, the use of robots for automated inspections can effectively improve inspection efficiency and accuracy and reduce the risk of safety accidents.

[0022] First, multiple monitoring robots are used to conduct basic monitoring of the target space (such as oil refining production areas, steel and metallurgical plants, etc., which can be set according to the actual inspection scene). The inspection robots carry various sensors (such as temperature sensors, pressure sensors, gas sensors, etc.) to conduct basic automated monitoring of the target plant area. At this stage, the monitoring is usually low-resolution, that is, the robot quickly scans the production environment, covering a large area, but the collected data is of low accuracy. Then the sensor data is compared with the preset standard parameters (such as the normal operating value of the equipment). If the parameters at a certain location deviate, it will be considered abnormal. These abnormal areas may represent equipment failures, signs of danger or other potential problems. On the other hand, due to the limitations of low-resolution monitoring, there may be misjudgments when anomalies occur. That is to say, changes in certain places may not be real anomalies (such as environmental factors, sensor errors, etc.), but due to the limitations of the monitoring system, they are still judged as anomalies. Therefore, in order to avoid false alarms caused by misjudgments, all areas initially detected as abnormal are regarded as implicit abnormal locations. These locations are not the final anomalies, but only a potential problem area, indicating that there may be problems in these areas. The implicit abnormal location is obtained. The main function of the implicit abnormal location is to provide basic data for subsequent verification monitoring. At this step, the implicit abnormal location does not immediately trigger an emergency response, but serves as a key area for subsequent high-resolution monitoring. When the implicit abnormal location is marked, these areas can be detected more accurately.

[0023] Through the collaboration of multiple robots, a large range of target space can be quickly scanned, and the coordinates of the abnormal location can be obtained when an abnormality occurs. It will be regarded as an implicit abnormal location, which can make monitoring resources more inclined to high-risk areas and improve the efficiency and accuracy of production area inspections.

[0024] S120: Predicting abnormal angles according to the implicit abnormal position to obtain multiple implicit abnormal angles and multiple angle probabilities.

[0025] Furthermore, step S120 of the present invention further includes: S121: Collecting abnormal angle prediction data based on historical monitoring abnormal data in the target space.

[0026] Furthermore, step S121 of the present invention further includes: S1211: Collect a set of sample abnormal position data based on historical monitoring abnormal data in the target space; S1212: Collect the angle at which the abnormality occurs at each sample abnormal position to obtain multiple sample abnormal angle sets, and collect the proportion of each sample abnormal angle to obtain multiple sample angle probability sets; S1213: Integrate the sample abnormal position set, multiple sample abnormal angle sets and multiple sample angle probability sets to obtain abnormal angle prediction data.

[0027] Specifically, first, the abnormal positions are collected through the historical monitoring abnormal data in the target space. These positions refer to the abnormal areas found in the past monitoring. The positioning of these abnormal areas is usually obtained by collecting data from robot sensors and comparing them with standard values. Then, the abnormal positions in all historical monitoring abnormal data are extracted to form a sample abnormal position set. This set contains all the specific position coordinates that were judged to be abnormal in the past monitoring. Then, the angles at which the abnormality occurs under each sample abnormal position are collected, that is, when a certain position is judged to be abnormal, the abnormality may appear at different angles of the position. These angles represent the possible locations of abnormal phenomena under different monitoring directions or viewing angles. For example, the air leakage of a certain device may be visible from multiple directions, and may be observed from different angles such as the front, side or top. Different abnormal angles under each abnormal position are collected, and a sample abnormal angle set is constructed, that is, each abnormal position corresponds to multiple possible abnormal angles, and multiple sample abnormal angle sets are obtained, wherein the sample abnormal position and the sample abnormal angle set correspond one to one.

[0028] On the other hand, since the frequency or proportion of each abnormal angle may be different in historical data, by calculating the frequency of occurrence of each angle in historical monitoring, we can obtain the probability distribution of each abnormal angle. For example, one side of the equipment may be more prone to abnormalities than other sides. Therefore, for each abnormal angle, its proportion of occurrence in past data is counted, and the proportion (i.e., probability) of each angle is recorded to form a sample angle probability set. This set represents the probability of abnormality at each angle corresponding to each abnormal position, and multiple sample angle probability sets are obtained, in which the sample abnormal angles and sample angle probabilities correspond one to one.

[0029] Next, the sample anomaly position set, multiple sample anomaly angle sets and multiple sample angle probability sets are integrated, that is, the sample anomaly positions, sample anomaly angle sets and sample angle probability sets that have corresponding relationships are integrated into a set of sample data, and multiple sets of sample data are obtained as anomaly angle prediction data.

[0030] S122: Using the abnormal angle prediction data, construct an abnormal angle prediction table, wherein the abnormal angle prediction table includes a mapping relationship between abnormal positions, abnormal angles and abnormal angle probabilities; S123: inputting the implicit abnormal position into the abnormal angle prediction table, and mapping output to obtain multiple implicit abnormal angles and multiple angle probabilities.

[0031] Specifically, multiple groups of sample data in the abnormal angle prediction data are obtained, wherein each group of sample data includes a sample abnormal position and a corresponding sample abnormal angle set and a sample angle probability set; then, based on the decision tree principle, a decision tree is a tree-structured data modeling method, wherein each node represents a decision judgment on a certain feature, and each child node represents a different branch of the decision, and the final leaf node contains the final result of the decision. According to the multiple groups of sample data, the sample abnormal position in each group of sample data is used as a child node, and the corresponding sample abnormal angle set and sample angle probability set (for each abnormal angle under each abnormal position, the probability of its occurrence is mapped) are used as leaf nodes of the child node to construct an abnormal angle prediction table, and the abnormal angle prediction table is shown in Table 1 below: Table 1: Abnormal angle prediction table

[0032] Finally, the implicit abnormal position is input into the abnormal angle prediction table for matching. According to the mapping rules in the table, the implicit abnormal position is input to find all possible abnormal angles at that position in the table, and the probability of abnormality at each angle is obtained, and multiple implicit abnormal angles and multiple angle probabilities are output.

[0033] Constructing an abnormal angle prediction table based on the decision tree principle can effectively organize and map abnormal positions, abnormal angles and their corresponding occurrence probabilities through a hierarchical structure, helping inspection robots make more accurate task allocation decisions based on historical data, improving the efficiency and accuracy of inspection tasks, while avoiding resource waste and misjudgment and optimizing the monitoring process.

[0034] S200: Generate a plurality of adaptation coefficients and a plurality of cost coefficients for the plurality of implicit abnormal angles respectively according to the plurality of angle probabilities.

[0035] Furthermore, step S200 of the present invention further includes: S210: Using the multiple angle probabilities as multiple adaptation coefficients of the multiple implicit abnormal angles; S220: Subtracting the multiple adaptation coefficients from 1 respectively to obtain multiple cost coefficients.

[0036] Specifically, the multiple angle probabilities are used as multiple adaptation coefficients of the multiple implicit abnormal angles, wherein each implicit abnormal angle has a probability of occurrence. These probability values ​​reflect the possibility of occurrence of abnormal angles. In order to optimize the inspection task allocation, the angle probability is used as the adaptation coefficient of the implicit abnormal angle. The adaptation coefficient represents the adaptability weight between the selected robot and the implicit abnormal angle. The higher the probability of the angle, the larger the adaptation coefficient, which means that when the robot selects this angle for verification monitoring, the monitoring quality and accuracy are higher, thereby allowing the robot to give priority to these high-probability angles for verification monitoring.

[0037] Next, 1 is subtracted from the multiple adaptation coefficients, and the difference between 1 and the adaptation coefficient is used as the cost coefficient corresponding to the implicit abnormal angle to obtain multiple cost coefficients. The cost coefficient reflects the weight of the scheduling distance cost between the selected robot and the implicit abnormal position. That is, the larger the adaptation coefficient, the smaller the cost coefficient, that is, the lower the weight of the cost of scheduling the robot in verification monitoring, which means that the robot can be dispatched to high-risk areas for monitoring at a lower cost.

[0038] By calculating the adaptation coefficient and the cost coefficient, the inspection task allocation can be dynamically adjusted according to the size of the adaptation coefficient and the cost coefficient. For example, a large adaptation coefficient indicates that the probability of an abnormality at this angle is high, and the robot should give priority to verifying this angle; a small cost coefficient indicates that the scheduling cost is low, and the robot should give priority to tasks with lower scheduling costs. This method can not only effectively improve the inspection efficiency, but also reduce the waste of invalid resources, making the inspection task allocation more accurate and intelligent.

[0039] S300: Acquire multiple real-time positions and multiple real-time angles of the multiple monitoring robots, optimize the allocation of monitoring tasks for the multiple implicit abnormal angles in the order of the probability of the multiple angles from large to small, according to the implicit abnormal positions, multiple adaptation coefficients and multiple cost coefficients, obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

[0040] Furthermore, step S300 of the present invention further includes: S310: Acquire multiple real-time positions and multiple real-time angles of the multiple monitoring robots; S320: Select a first implicit abnormal angle in descending order of probability of the multiple angles, and generate a first verification monitoring angle relative to the first implicit abnormal angle.

[0041] Specifically, first, multiple real-time positions and multiple real-time angles of the multiple monitoring robots are obtained, which can be obtained through GPS or IMU (inertial measurement unit). This information determines whether the robot can cover the target area and whether it can effectively monitor from the correct angle. The real-time angle represents the direction or viewing angle of the robot, usually expressed in degrees, for example, 0° represents due east, 90° represents due north, 180° represents west, and 270° represents south. Then, according to the order of the multiple angle probabilities from large to small, the implicit abnormal angle corresponding to the maximum angle probability is selected as the first implicit abnormal angle. This angle is the angle where the abnormality is most likely to occur, so it needs to be verified first. Then determine the first verification monitoring angle relative to the first implicit abnormal angle. For example, assuming that the first implicit abnormal angle is due east, the degree calculated clockwise from the positive direction of the X-axis is 0°, then the angle of the relative monitoring acquisition of the angle image should be due west, that is, the relative direction, and the degree is 180°, then the first verification monitoring angle is 180°.

[0042] S330: According to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, based on the first verification monitoring angle, perform task optimization allocation to obtain a first monitoring task.

[0043] Further, step S330 of the present invention further includes: S331: Construct a first task allocation optimization function according to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, and the first verification monitoring angle, as shown in the following formula: ; Among them, FITM is the task assignment score, is the first adaptation coefficient, is the first cost coefficient, is the real-time angle of the robot selected in the optimization, For the first verification monitoring angle, is the deviation between the robot's real-time angle and the first verification monitoring angle, is the distance between the real-time position of the robot selected in the optimization and the implicit abnormal position, is a preset distance; S332: randomly selecting a first monitoring robot from the multiple monitoring robots; S333: according to the first real-time position and the first real-time angle of the first monitoring robot, combined with the first verification monitoring angle and the implicit abnormal position, calculating the first task allocation score based on the first task allocation optimization function; S334: continue to traverse and select monitoring robots and calculate task scores, filter the monitoring robot with the largest task allocation score, perform verification monitoring of the first implicit abnormal angle, and obtain the first monitoring task.

[0044] Specifically, in the inspection task, the robot needs to optimize the task allocation according to its current position, angle and target monitoring angle. The goal is to improve the inspection efficiency and reduce resource waste by optimizing the task allocation. The angle of the monitoring robot is generally pre-set to monitor the predetermined area. Therefore, in the verification monitoring, the robot angle is not adjusted, but the robot position is moved and adjusted to verify the hidden position.

[0045] Firstly, according to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, and the first verification monitoring angle, a first task allocation optimization function is constructed. The first task allocation optimization function is used to evaluate the pros and cons of task allocation, aiming to balance adaptability, cost, angle deviation and robot scheduling distance, so as to make the monitoring task more efficient and accurate, avoid waste of resources and improve inspection efficiency.

[0046] In the first task allocation optimization function, FITM is a task allocation score, which is used to measure the quality of task allocation. The larger the value, the more appropriate the task allocation. is the first adaptation coefficient, is the first cost coefficient, is the real-time angle of the robot selected in the optimization, For the first verification monitoring angle, It is the deviation between the real-time angle of the robot and the first verification monitoring angle. The smaller it is, the more adapted it is, the better the monitoring effect is, and the greater the task allocation score is. is the distance between the real-time position of the robot selected in the optimization and the implicit abnormal position, which can be calculated based on the coordinates of the real-time position and the implicit abnormal position. The larger it is, the greater the cost of the moving distance for verification monitoring, and the smaller the task allocation score. It is a preset distance, which indicates the optimal monitoring distance between the robot and the hidden abnormal position. It can be set according to the actual scenario, for example, 5m.

[0047] Next, randomly select any one of the multiple monitoring robots as the first monitoring robot, and obtain the first real-time position and the first real-time angle of the first monitoring robot; then use the first task allocation optimization function to calculate the first task allocation score according to the first real-time position, the first real-time angle, the first verification monitoring angle and the implicit abnormal position. Further continue to randomly select monitoring robots from the multiple monitoring robots and calculate the task score until all monitoring robots are analyzed, select the robot with the highest task score, and assign it to perform the verification monitoring task of the first implicit abnormal angle to obtain the first monitoring task. Through this optimization process, the accuracy and efficiency of task allocation can be ensured, so that abnormal monitoring work can be completed efficiently.

[0048] S340: Continue to optimize the allocation of monitoring tasks for other multiple implicit abnormal angles in the order of the multiple angle probabilities from large to small, obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

[0049] Specifically, after completing the monitoring task assignment for the first implicit abnormal angle, the next step is to optimize the task assignment for the other multiple implicit abnormal angles in the order of the probabilities of the multiple angles from large to small, that is, for each implicit abnormal angle, select the robot with the highest task score from the remaining multiple monitoring robots to assign the verification monitoring task of the implicit abnormal angle, and obtain multiple monitoring tasks; after assigning the corresponding robot to each implicit abnormal angle, control the multiple monitoring robots to perform the verification monitoring task.

[0050] This approach ensures that each potential abnormal angle is verified by the most suitable robot, optimizes resource allocation during the inspection process, and improves the efficiency and accuracy of anomaly detection.

[0051] A robot collaborative task allocation method provided by an embodiment of the present invention has at least the following technical effects: A plurality of monitoring robots are used to perform basic monitoring in the target space. When an abnormality occurs, the implicit abnormal position is obtained, the abnormal angle is predicted, and a plurality of implicit abnormal angles and a plurality of angle probabilities are obtained; then, according to the plurality of angle probabilities, a plurality of adaptation coefficients and a plurality of cost coefficients of the plurality of implicit abnormal angles are respectively generated; then, a plurality of real-time positions and a plurality of real-time angles of the plurality of monitoring robots are obtained, and in the order of the plurality of angle probabilities from large to small, the plurality of implicit abnormal angles are optimally allocated monitoring tasks according to the implicit abnormal position, the plurality of adaptation coefficients and the plurality of cost coefficients, and a plurality of monitoring tasks are obtained; finally, according to the plurality of monitoring tasks, the plurality of monitoring robots are controlled to perform verification monitoring; that is, by optimizing the allocation of robot tasks, the matching degree between the inspection tasks and the actual monitoring requirements and the current status of the robots can be improved, thereby improving the scientificity and accuracy of the inspection task allocation, ensuring that high-risk areas are adequately monitored, and reducing invalid or inefficient resource waste, thereby effectively reducing scheduling costs and improving the accuracy and efficiency of abnormality detection, and realizing efficient and accurate factory inspections.

[0052] Embodiment 2, as Figure 2As shown, based on the same inventive concept as a robot collaborative task allocation method provided in Example 1, an embodiment of the present invention further provides a robot collaborative task allocation system, including: an implicit abnormality information acquisition module 10, used to perform basic monitoring of the target space through multiple monitoring robots, and when an abnormality occurs, obtain the implicit abnormality position, predict the abnormal angle, and obtain multiple implicit abnormality angles and multiple angle probabilities; a characteristic coefficient analysis module 20, used to generate multiple adaptation coefficients and multiple cost coefficients for the multiple implicit abnormal angles according to the multiple angle probabilities; a monitoring task optimization allocation module 30, used to obtain multiple real-time positions and multiple real-time angles of the multiple monitoring robots, and optimize the monitoring task allocation for the multiple implicit abnormal angles according to the implicit abnormal positions, multiple adaptation coefficients and multiple cost coefficients in the order of the multiple angle probabilities from large to small, to obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

[0053] Furthermore, the robot collaborative task allocation system is also used to: perform basic monitoring of the target space through multiple monitoring robots, and when an abnormality occurs, obtain the location coordinates of the abnormality as an implicit abnormal position; predict the abnormal angle based on the implicit abnormal position to obtain multiple implicit abnormal angles and multiple angle probabilities.

[0054] Furthermore, the robot collaborative task allocation system is also used to: collect abnormal angle prediction data based on historical monitoring abnormal data in the target space; use the abnormal angle prediction data to construct an abnormal angle prediction table, wherein the abnormal angle prediction table includes a mapping relationship between abnormal positions, abnormal angles and abnormal angle probabilities; input the implicit abnormal position into the abnormal angle prediction table, and obtain multiple implicit abnormal angles and multiple angle probabilities through mapping output.

[0055] Furthermore, the robot collaborative task allocation system is also used to: collect a set of sample abnormal position data based on historical monitoring abnormality data in the target space; collect the angle at which the abnormality occurs at each sample abnormal position to obtain multiple sample abnormal angle sets, and collect the proportion of each sample abnormal angle to obtain multiple sample angle probability sets; integrate the sample abnormal position set, multiple sample abnormal angle sets and multiple sample angle probability sets to obtain abnormal angle prediction data.

[0056] Furthermore, the robot collaborative task allocation system is also used to: use the multiple angle probabilities as multiple adaptation coefficients of the multiple implicit abnormal angles; and subtract the multiple adaptation coefficients from 1 to obtain multiple cost coefficients.

[0057] Furthermore, the robot collaborative task allocation system is also used to: obtain multiple real-time positions and multiple real-time angles of the multiple monitoring robots; select a first implicit abnormal angle in order of the probabilities of the multiple angles from large to small, and generate a first verification monitoring angle relative to the first implicit abnormal angle; optimize task allocation based on the first verification monitoring angle according to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, and obtain a first monitoring task; continue to optimize monitoring task allocation for other multiple implicit abnormal angles in order of the probabilities of the multiple angles from large to small, obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

[0058] Furthermore, the robot collaborative task allocation system is also used to construct a first task allocation optimization function according to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, and the first verification monitoring angle, as shown in the following formula: ; Among them, FITM is the task assignment score, is the first adaptation coefficient, is the first cost coefficient, is the real-time angle of the robot selected in the optimization, For the first verification monitoring angle, is the deviation between the robot's real-time angle and the first verification monitoring angle, is the distance between the real-time position of the robot selected in the optimization and the implicit abnormal position, is a preset distance; among the multiple monitoring robots, a first monitoring robot is randomly selected; according to the first real-time position and the first real-time angle of the first monitoring robot, combined with the first verification monitoring angle and the implicit abnormal position, a first task allocation score is calculated based on the first task allocation optimization function; continue to traverse and select monitoring robots and calculate task scores, select the monitoring robot with the largest task allocation score, perform verification monitoring of the first implicit abnormal angle, and obtain the first monitoring task.

[0059] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0060] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A robot collaborative task allocation method, characterized in that: The method is applied to multiple monitoring robots, and the method comprises: Use multiple monitoring robots to conduct basic monitoring in the target space. When an abnormality occurs, obtain the implicit abnormal position, predict the abnormal angle, and obtain multiple implicit abnormal angles and multiple angle probabilities. According to the multiple angle probabilities, respectively generate multiple adaptation coefficients and multiple cost coefficients for the multiple implicit abnormal angles; Acquire multiple real-time positions and multiple real-time angles of the multiple monitoring robots, optimize the allocation of monitoring tasks for the multiple implicit abnormal angles in order of the probability of the multiple angles from large to small, based on the implicit abnormal positions, multiple adaptation coefficients and multiple cost coefficients, obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

2. The robot collaborative task allocation method according to claim 1, characterized in that: Through multiple monitoring robots, basic monitoring is performed in the target space. When an abnormality occurs, the implicit abnormal position is obtained, the abnormal angle is predicted, and multiple implicit abnormal angles and multiple angle probabilities are obtained, including: Use multiple monitoring robots to perform basic monitoring in the target space. When an abnormality occurs, obtain the coordinates of the abnormal location as the implicit abnormal location; Anomaly angle prediction is performed according to the implicit abnormal position to obtain multiple implicit abnormal angles and multiple angle probabilities.

3. The robot collaborative task allocation method according to claim 2, characterized in that: An abnormal angle is predicted according to the implicit abnormal position to obtain multiple implicit abnormal angles and multiple angle probabilities, including: Collecting abnormal angle prediction data based on historical monitoring abnormal data in the target space; Using the abnormal angle prediction data, constructing an abnormal angle prediction table, wherein the abnormal angle prediction table includes a mapping relationship between abnormal positions, abnormal angles, and abnormal angle probabilities; The implicit abnormal position is input into the abnormal angle prediction table, and a mapping output is obtained to obtain multiple implicit abnormal angles and multiple angle probabilities.

4. The robot collaborative task allocation method according to claim 3, characterized in that: According to the historical monitoring abnormal data in the target space, abnormal angle prediction data is collected, including: Collecting a set of sample abnormal positions according to historical monitoring abnormal data in the target space; The abnormal angles at each sample abnormal position are collected to obtain multiple sample abnormal angle sets, and the proportion of each sample abnormal angle is collected to obtain multiple sample angle probability sets; The sample abnormal position set, multiple sample abnormal angle sets and multiple sample angle probability sets are integrated to obtain abnormal angle prediction data.

5. The robot collaborative task allocation method according to claim 1, characterized in that: According to the multiple angle probabilities, a plurality of adaptation coefficients and a plurality of cost coefficients of the multiple implicit abnormal angles are respectively generated, including: Using the multiple angle probabilities as multiple adaptation coefficients of the multiple implicit abnormal angles; The multiple adaptation coefficients are respectively subtracted from 1 to obtain multiple cost coefficients.

6. The robot collaborative task allocation method according to claim 1, characterized in that: Acquire multiple real-time positions and multiple real-time angles of the multiple monitoring robots, optimize the allocation of monitoring tasks for the multiple implicit abnormal angles in descending order of probability of the multiple angles according to the implicit abnormal positions, multiple adaptation coefficients and multiple cost coefficients, obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring, including: Acquiring multiple real-time positions and multiple real-time angles of the multiple monitoring robots; Selecting a first implicit abnormal angle according to the order of the plurality of angle probabilities from large to small, and generating a first verification monitoring angle relative to the first implicit abnormal angle; According to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, based on the first verification monitoring angle, perform task optimization allocation to obtain a first monitoring task; Continue to optimize the allocation of monitoring tasks for other multiple implicit abnormal angles in the order of the probabilities of the multiple angles from large to small, obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

7. The robot collaborative task allocation method according to claim 6, characterized in that: According to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, based on the first verification monitoring angle, performing task optimization allocation to obtain a first monitoring task includes: According to the first adaptation coefficient and the first cost coefficient corresponding to the first implicit abnormal angle, and the first verification monitoring angle, a first task allocation optimization function is constructed as follows: ； Among them, FITM is the task allocation score, is the first adaptation coefficient, is the first cost coefficient, is the real-time angle of the robot selected in the optimization, For the first verification monitoring angle, is the deviation between the robot's real-time angle and the first verification monitoring angle, is the distance between the real-time position of the robot selected in the optimization and the implicit abnormal position, is the preset distance; Randomly selecting a first monitoring robot among the plurality of monitoring robots; Calculate a first task allocation score based on the first task allocation optimization function according to the first real-time position and the first real-time angle of the first monitoring robot, in combination with the first verification monitoring angle and the implicit abnormal position; Continue to traverse and select monitoring robots and calculate task scores, select the monitoring robot with the largest task assignment score, perform verification monitoring of the first implicit abnormal angle, and obtain the first monitoring task.

8. A robot collaborative task allocation system, characterized in that: The steps for implementing a robot collaborative task allocation method as described in any one of claims 1 to 7 include: The implicit abnormal information acquisition module is used to perform basic monitoring in the target space through multiple monitoring robots. When an abnormality occurs, the implicit abnormal position is obtained, the abnormal angle is predicted, and multiple implicit abnormal angles and multiple angle probabilities are obtained. A characteristic coefficient analysis module, used to generate a plurality of adaptation coefficients and a plurality of cost coefficients of the plurality of implicit abnormal angles respectively according to the plurality of angle probabilities; A monitoring task optimization allocation module is used to obtain multiple real-time positions and multiple real-time angles of the multiple monitoring robots, optimize the allocation of monitoring tasks for the multiple implicit abnormal angles in the order of the probability of the multiple angles from large to small, based on the implicit abnormal positions, multiple adaptation coefficients and multiple cost coefficients, obtain multiple monitoring tasks, and control the multiple monitoring robots to perform verification monitoring.

Images