State self-checking system of unhooking robot

Abstract

The invention provides a state self-checking system for an unhooking robot, and relates to the technical field of automatic machinery, and the system comprises a real-time collection module which collects a real-time working data set of the unhooking robot; the operation evaluation module is used for comprehensively evaluating the operation of the unhooking robot based on the real-time working data set; the fault detection module is used for carrying out point cloud matching and data comparison on the unhooking robot based on the operation state evaluation information, carrying out fault detection according to a comparison result and generating a fault self-detection report; the fault tracing module is used for synchronizing the fault self-checking report to a human-computer interface for fault tracing; and the fault processing module is used for formulating fault processing suggestions and sending the suggestions to the human-computer interface to perform remote control on the unhooking robot. The technical problem that in the prior art, due to insufficient fault detection precision, tiny or complex faults cannot be found in time, and the operation stability of the robot is affected is solved, the fault detection precision and the response speed are improved, and the operation reliability and safety are enhanced.

Description

Technical Field

[0001] The present application relates to the field of automated mechanical technology, and in particular to a state self-checking system for a hook-removing robot. Background Art

[0002] In railway transportation and marshalling yard operations, unhooking robots have become one of the key equipment for realizing automated operations. They can automatically complete unhooking operations and improve production efficiency.

[0003] At present, most hook-removing robots rely on single sensor data for fault detection and lack comprehensive comprehensive evaluation capabilities. This detection method is difficult to accurately identify small mechanical deviations or complex multi-component coordinated failures, especially in complex working conditions, such as deformation of the coupler handle or a variety of carriage models. The detection accuracy and adaptability of existing technologies are obviously insufficient. In addition, after a fault occurs, existing methods require manual on-site troubleshooting and processing, which is not only time-consuming and labor-intensive, but also seriously affects the stability and production efficiency of the equipment. These problems have, to a certain extent, limited the application and promotion of hook-removing robots in complex environments. Summary of the invention

[0004] The present application provides a status self-check system for a hook-removing robot, which solves the technical problem that the prior art relies on single sensor data and has insufficient fault detection accuracy, resulting in the inability to promptly detect minor or complex faults, which in turn affects the operational stability of the robot. This achieves the technical effect of improving the fault detection accuracy and response speed of the hook-removing robot and enhancing the operational reliability and safety of the hook-removing robot.

[0005] In view of the above problems, the present application provides a state self-checking system for a hook-removing robot, the system comprising: A real-time acquisition module is used to traverse multiple components of the unhooking robot for sensing through a sensing unit to collect a real-time working data set of the unhooking robot; an operation evaluation module is used to perform a comprehensive operation evaluation on the unhooking robot based on the real-time working data set to generate operation status evaluation information; a fault detection module is used to perform point cloud matching on the unhooking robot based on the operation status evaluation information, perform data comparison according to the point cloud matching result, generate a comparison deviation coefficient, perform fault detection according to the comparison deviation coefficient, and generate a fault self-check report; a fault tracing module is used to synchronize the fault self-check report to the human-machine interface for fault tracing and determine the fault source data; a fault handling module is used to monitor the status of the unhooking robot according to the fault source data, formulate fault handling suggestions, send the fault handling suggestions to the human-machine interface, and remotely control the unhooking robot through the human-machine interface.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The real-time acquisition module traverses multiple components of the unhooking robot through the sensor unit to collect real-time working data. Through the detailed collection at the component level, it can more accurately monitor every detail of the unhooking robot in actual operation, provide basic data for the entire self-checking system, and ensure that fault detection and evaluation are based on accurate real-time information, thereby improving detection accuracy and data integrity. The operation evaluation module performs a comprehensive operation evaluation of the unhooking robot based on the real-time working data set, generates operation status evaluation information, grasps the operation status of the robot as a whole, and provides a comprehensive reference for fault detection. The fault detection module performs point cloud matching and data comparison on the unhooking robot based on the operation status evaluation information, accurately detects faults, and further converts the operation status evaluation information into specific fault detection results, so that potential faults can be quantified and identified. The fault tracing module synchronizes the fault self-check report to the human-machine interface for fault tracing, deeply explores the root cause of the fault, and determines the fault source data, which helps to fundamentally understand the cause of the fault and provide accurate direction for subsequent fault handling. Based on the fault tracing results, the fault handling module monitors the status of the unhooking robot and formulates corresponding fault handling suggestions. It also supports remote control of the robot through the human-machine interface to achieve intelligent fault response, so that the self-test system can not only detect faults, but also provide solutions and implement fault handling remotely, reducing manual intervention and improving the availability of the unhooking robot and operation continuity.

[0007] In summary, this application uses a real-time acquisition module to fully perceive the operating status of the unhooking robot, an operation evaluation module to conduct comprehensive analysis and judgment, a fault detection module to accurately locate the fault, a fault tracing module to quickly guide the operator to find the root cause of the fault, and a fault handling module to provide solutions and support remote operation, thereby achieving comprehensive monitoring and efficient maintenance of the unhooking robot status. This solution not only improves the accuracy and efficiency of fault detection, but also significantly shortens the fault handling time through intelligent fault handling suggestions and remote control functions, improves the operating reliability and production efficiency of the unhooking robot, and enables it to better adapt to production needs in complex industrial environments.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic diagram of the structure of the status self-check system of the unhooking robot provided in an embodiment of the present application.

[0010] Figure 2A flowchart of incremental learning of long-term state scores and short-term state scores to generate learning results in the state self-check system of the unhooking robot provided in an embodiment of the present application.

[0011] Explanation of reference numerals: real-time acquisition module 10 , operation evaluation module 20 , fault detection module 30 , fault tracing module 40 , fault processing module 50 . DETAILED DESCRIPTION

[0012] The embodiment of the present application solves the technical problem that the prior art relies on a single sensor data and has insufficient fault detection accuracy, resulting in the inability to promptly detect minor or complex faults, which in turn affects the stability of the robot's operation, by providing a status self-check system for the unhooking robot. This achieves the technical effect of improving the fault detection accuracy and response speed of the unhooking robot and enhancing the reliability and safety of the unhooking robot's operation.

[0013] like Figure 1 As shown, the embodiment of the present application provides a state self-checking system for a hook-removing robot, the system comprising: The real-time acquisition module 10 is used to perform sensing by traversing multiple components of the hook-removing robot through a sensing unit, and to collect the real-time working data set of the hook-removing robot.

[0014] Specifically, the sensing unit is a device that can sense physical quantities (such as position, speed, force, etc.) and convert them into measurable signals, usually including sensors and signal conditioning circuits. The real-time working data set is a collection of relevant working information of each component of the unhooking robot at the current moment, such as the position coordinates of the robotic arm, the movement speed value, the torque magnitude, and other information.

[0015] The real-time acquisition module 10 realizes data acquisition by installing various sensor units on different parts of the unhooking robot. These sensor units will read data at certain time intervals or under specific trigger conditions. For example, for the robotic arm part, angle sensors may be installed at the joints, displacement sensors may be installed on the moving parts, and force sensors may be installed on the drive devices. These sensors will continuously sense the state changes of the components and convert the physical quantities into electrical signals or digital signals. Then, the signals are converted into a form suitable for acquisition through signal conditioning circuits (such as amplification, filtering, etc.), and finally collected by the acquisition equipment (such as data acquisition cards) to form a real-time working data set.

[0016] By collecting the real-time working data set of the unhooking robot, comprehensive and accurate operating data is provided for the entire self-inspection system, which is the basis for subsequent modules to carry out evaluation, detection and other operations.

[0017] The operation evaluation module 20 is used to perform a comprehensive operation evaluation on the unhooking robot based on the real-time working data set to generate operation status evaluation information.

[0018] Specifically, the operation evaluation module 20 receives the real-time working data set from the real-time acquisition module 10, and classifies and organizes the data, for example, classifies the position data, speed data, torque data, etc. of the robot arm. Then, these data are analyzed according to the preset evaluation rules. These evaluation rules can be formulated based on the ideal operation model of the unhooking robot. For example, for the movement of the robot arm, according to the design parameters of the robot, its motion trajectory should be within a certain spatial range, and the speed should be within a reasonable range. By comparing and comprehensively analyzing the actual data collected with these ideal values, the overall operation state of the unhooking robot is evaluated to generate operation state evaluation information. For example, based on the position and speed data of the robot arm, it is judged whether its motion trajectory meets expectations, and based on the torque data, it is judged whether the mechanical structure is under abnormal pressure. This operation state evaluation information is a comprehensive description of the overall operation status of the unhooking robot, including whether the coordination between the various components of the robot is normal, whether the overall performance is within the normal range, and other information.

[0019] Through comprehensive evaluation of the real-time working data set, the operating status of the unhooking robot can be grasped as a whole, providing a preliminary judgment basis for subsequent fault detection.

[0020] The fault detection module 30 is used to perform point cloud matching on the unhooking robot based on the operating status evaluation information, perform data comparison according to the point cloud matching result, generate a comparison deviation coefficient, perform fault detection according to the comparison deviation coefficient, and generate a fault self-check report.

[0021] Specifically, the comparison deviation coefficient is a quantitative value used to indicate the degree of difference between the current data and the normal data. The fault self-check report is a report on whether there is a fault in the unhooking robot and related information of the fault (such as the component where the fault may occur, the severity of the fault, etc.).

[0022] The fault detection module 30 performs point cloud matching based on the operation status evaluation information generated by the operation evaluation module 20. For the key components of the unhooking robot, a point cloud model of the robot in normal operation is pre-established. For example, the three-dimensional shape of the robot arm in normal operation can be represented by a point cloud. Then, the component data under the current operation state collected is converted into a point cloud form and matched with the pre-stored normal point cloud model. In the point cloud matching process, an iterative closest point (ICP) algorithm can be used. For example, the relative position between the current point cloud and the normal point cloud is continuously adjusted by the ICP algorithm to minimize the matching error. After the point cloud matching is completed, data comparison is performed according to the matching results, such as comparing the differences in parameters such as distance and angle between the point clouds, and generating a comparison deviation coefficient. If this coefficient exceeds a preset threshold, it is determined that there is a fault, and finally a fault self-check report is generated, which contains information such as the components where the fault may occur, the type of fault (such as structural deformation, abnormal motion, etc.).

[0023] Through precise point cloud matching and data comparison technology, the fault condition of the unhooking robot can be accurately detected, and the operating status assessment information can be further converted into specific fault detection results, thereby improving the accuracy of fault detection.

[0024] The fault tracing module 40 is used to synchronize the fault self-check report to the human-machine interface for fault tracing and determine the fault source data.

[0025] Specifically, the human-machine interface is an interface for the operator to interact with the unhooking robot, which can be a combination of a display screen and an operation button, or a touch screen, etc. The fault tracing module 40 synchronizes the fault self-test report to the human-machine interface. On the human-machine interface, the operator can view the detailed information in the fault self-test report. Then, the fault tracing module 40 will trace the source of the fault based on the fault-related information in the report, combined with the structure and working principle of the unhooking robot. For example, if the fault self-test report shows that the motion trajectory of the robot arm is abnormal, the fault tracing module 40 will analyze factors such as the drive system, control system, and mechanical structure of the robot arm, such as checking the working parameters of the drive motor, the command output of the control system, the connection status of the joints of the robot arm, etc., and through step-by-step investigation, determine which component or link caused the fault, thereby determining the fault source data.

[0026] By displaying the fault self-check report on the human-machine interface and conducting source tracing analysis, the root cause of the fault can be deeply explored, providing accurate direction for subsequent fault handling.

[0027] The fault handling module 50 is used to monitor the status of the unhooking robot according to the fault source data, formulate fault handling suggestions, send the fault handling suggestions to the human-machine interface, and remotely control the unhooking robot through the human-machine interface.

[0028] Specifically, the fault handling module 50 monitors the state of the unhooking robot according to the fault source data, and sets a monitoring strategy for the faulty components or fault-related parameters. For example, if the fault source is the wear of the robot arm joint, then the position, force and other parameters of the joint will be monitored in detail. During the monitoring process, fault handling suggestions are formulated according to the specific circumstances of the fault. These fault handling suggestions are solutions and operational suggestions for the fault situation. If the parameters of a certain component are abnormal, the handling suggestion may be to adjust the control parameters of the component; if the component is damaged, the handling suggestion may be to replace the component. The fault handling suggestions are then sent to the human-machine interface, and the operator can remotely control the unhooking robot through the human-machine interface to perform fault handling operations.

[0029] Targeted status monitoring and processing suggestions are formulated based on the fault source data, and remote control of the unhooking robot is achieved through the human-machine interface, which improves the efficiency and accuracy of fault handling.

[0030] Furthermore, the operation evaluation module 20 of the embodiment of the present application further includes: The cluster analysis unit is used to traverse the real-time working data set to perform cluster analysis, determine multiple operating data, use the multiple operating data as cluster centers to perform data division on the real-time working data set, and generate multiple operating data clusters.

[0031] The long-term operation evaluation unit is used to identify the coupler type of the unhooking robot, perform long-term operation evaluation on the multiple operation data clusters according to the coupler type, and generate a long-term status score.

[0032] The short-term operation evaluation unit is used to perform short-term operation evaluation on the multiple operation data clusters according to the coupler type to generate a short-term status score.

[0033] The state verification unit is used to perform incremental learning on the long-term state score and the short-term state score to generate a learning result, perform state verification based on the learning result, and obtain a state verification token.

[0034] A reverse screening unit is used to perform reverse screening on the multiple operation data clusters according to the status verification token to generate the operation status evaluation information.

[0035] Specifically, the cluster analysis unit first performs cluster analysis on the real-time working data set, and classifies similar data points in the data set into one category. The entire real-time working data set is traversed, and multiple operating data are determined as cluster centers by calculating the similarity between data points (such as Euclidean distance, etc.). These cluster centers are representative data points in the data set. Then, based on these cluster centers, the data of the real-time working data set is divided into different categories, thereby generating multiple operating data clusters. Exemplarily, K-means clustering, DBSCAN clustering algorithm, etc. can be used. Taking the K-means clustering algorithm as an example, K initial cluster centers are set (K is the number of clusters to be divided), and then the positions of these cluster centers are continuously adjusted until the convergence conditions are met, thereby obtaining multiple operating data clusters, such as when K=3, operating data clusters representing normal operation, low-speed operation and high-speed operation can be obtained. Through cluster analysis, complex real-time working data sets are classified, so that subsequent evaluations can be carried out for different types of data clusters, thereby improving the pertinence and efficiency of the evaluation.

[0036] The long-term operation evaluation unit determines the type of coupler by reading the configuration information of the hook-off robot or identifying certain features of the coupler (such as the shape of the coupler, the connection method, etc.), and extracts the operation data cluster related to the current coupler type from the historical data according to the determined coupler type for long-term operation evaluation, that is, evaluates and analyzes the operation data of the hook-off robot over a long period of time (such as the past hours, days or even weeks), analyzes the movement trajectory of the mechanical arm of the hook-off robot in the long-term operation, the force condition and other operation data rules, such as the average value and change trend of the data, and generates a long-term state score for each operation data cluster according to the statistical characteristics of the long-term data (such as the average value, standard deviation, etc.). This long-term state score reflects the state of the hook-off robot in various aspects under long-term operation. In the long-term operation evaluation, the time series analysis method can be used. For example, for the force data cluster of the coupler, the autoregressive moving average (ARMA) model is used to analyze the change trend of the force data in the long term, so as to generate a long-term state score for this data cluster.

[0037] The short-term operation evaluation unit extracts the recent operation data cluster related to the current coupler type from the real-time working data set according to the coupler type for short-term operation evaluation. The short-term operation evaluation focuses on the operation data cluster related to the coupler in a short period of time (such as a few minutes to a few hours). Similar to the long-term operation evaluation, by analyzing the data change rules in the data cluster, each operation data cluster is evaluated and a short-term status score is generated. In the short-term operation evaluation, a sliding window algorithm can be used. For example, when analyzing the short-term robot arm motion speed data cluster, a short time window (such as 10 minutes) is set, and the speed mean, variance and other statistics are calculated within this window, and then a short-term status score is generated based on these statistics. The short-term operation evaluation can timely reflect the operation status of the unhooking robot in a short period of time. Combined with the long-term operation evaluation, the operation of the robot can be more comprehensively evaluated.

[0038] The state verification unit conducts incremental learning on the long-term state score and the short-term state score, inputs the long-term state score and the short-term state score into the incremental learning model, and learns the relationship between the two. For example, the long-term force of the manipulator in the long-term state score and the recent force fluctuation in the short-term state score are learned through the incremental learning algorithm to generate a learning result. Then, based on this learning result, the state verification is performed by comparing the learning result with a pre-set standard or model, for example, comparing the learned long-term and short-term relationship of the manipulator force with the model under normal working conditions, to generate a state verification token. This state verification token is a result identifier for verifying whether the operating state of the equipment is normal, usually a binary value (normal or abnormal). Through incremental learning and state verification, the relationship between the long-term and short-term operating state scores can be better utilized to improve the accuracy of the evaluation of the operating state of the dehooking robot and timely discover abnormal conditions in the operating state.

[0039] The reverse screening unit reversely screens multiple operation data clusters according to the status verification token, removes abnormal data clusters, and then aggregates the screened data clusters to generate the final operation status evaluation information. Through reverse screening, the data can be optimized according to the verification results of the operation status, and the operation status evaluation information under the normal operation state can be generated, providing a more effective basis for subsequent fault detection and other operations.

[0040] Further, such as Figure 2 As shown, the state verification unit is also used to perform the following steps: Step 1: Perform state feature analysis on the unhooking robot based on the multiple operation data clusters to construct multiple operation feature vectors.

[0041] Step 2: Standardize and learn the long-term state score and the short-term state score according to the multiple operating feature vectors to generate multiple standard sample parameters, wherein the multiple standard sample parameters include long-term standard sample parameters and short-term standard sample parameters.

[0042] Step three: Use stochastic gradient descent to adjust the parameters of the long-term standard sample to generate a long-term training parameter set, and use stochastic gradient descent to adjust the parameters of the short-term standard sample to generate a short-term training parameter set.

[0043] Step 4: Based on the long-term training parameter set, the long-term standard sample parameters are incrementally learned in a time series to generate long-term learning results; based on the short-term training parameter set, the short-term standard sample parameters are incrementally learned in a time series to generate short-term learning results.

[0044] Step 5: Add the long-term learning result and the short-term learning result to the learning result.

[0045] Specifically, state feature analysis is performed on multiple operation data clusters to extract features that can reflect their operation status, such as the mean, variance, extreme value, and change trend of the data. Taking the operation data cluster containing the robot arm motion speed data as an example, the mean, variance, maximum, and minimum speed features are calculated. Then, these features are combined into operation feature vectors in a certain order. These operation feature vectors are used to represent the operation status characteristics of the unhooking robot in a certain aspect. By constructing the operation feature vector, complex operation data clusters can be represented in a concise and structured way, which is convenient for subsequent learning and analysis operations, and provides an effective data representation for standardized learning.

[0046] The long-term state score and the short-term state score are standardized and learned according to multiple running feature vectors. First, the long-term state score and the short-term state score are standardized (such as the z-score standardization formula) according to the statistics related to the running feature vectors (such as the mean and standard deviation of the features), and multiple standard sample parameters are obtained, including long-term standard sample parameters and short-term standard sample parameters. Through standardized learning, the long-term state score and the short-term state score are comparable at the same scale.

[0047] For the long-term standard sample parameters, the stochastic gradient descent algorithm is used to adjust the parameters. First, a loss function (such as the mean square error loss function) is defined. Then, according to the update formula of the stochastic gradient descent algorithm, in each iteration, the gradient of the loss function to the long-term standard sample parameters is calculated, and the parameters are adjusted according to the update formula. After multiple iterations, a long-term training parameter set is generated. For the short-term standard sample parameters, the same method is used to generate a short-term training parameter set. By adjusting the standard sample parameters through the stochastic gradient descent algorithm, the parameters can be optimized to make them more in line with the actual operation status evaluation requirements, improve the effect of subsequent incremental learning, and thus more accurately reflect the long-term and short-term operation status of the unhooking robot.

[0048] For the long-term training parameter set, the state verification unit performs incremental learning of the long-term standard sample parameters according to the time series based on this parameter set. At each time step, an incremental learning algorithm (such as an incremental neural network algorithm) is used to learn based on the existing learning results and the current standard sample parameters. For example, in a neural network, the output is calculated based on the input long-term standard sample parameters and the existing long-term training parameter set, and then the long-term training parameter set is adjusted based on the difference between the output and the target value (such as the target value determined based on the long-term state score), and the long-term learning result is generated after learning the entire time series. For the short-term training parameter set and the short-term standard sample parameters, the same method is used to generate the short-term learning result.

[0049] The long-term learning results and short-term learning results are added to the previously generated learning results, and the operating status information at different time scales is integrated to form a more comprehensive learning result, providing a richer and more accurate basis for subsequent status verification.

[0050] Furthermore, the fault detection module 30 in the embodiment of the present application includes: The point cloud data acquisition unit is used to activate the scanning unit to scan the hook removing robot and draw the point cloud data set of the hook removing robot.

[0051] The motion analysis unit is used to perform motion analysis on the hook-removing robot based on the point cloud data set and extract a three-dimensional motion trajectory.

[0052] The preset motion trajectory determination unit is used to map the running status assessment information to the point cloud data set for motion analysis to determine a preset motion trajectory.

[0053] The point cloud matching unit is used to match the three-dimensional motion trajectory with the preset motion trajectory according to the point cloud data set to generate a point cloud matching result.

[0054] The deviation identification unit is used to determine whether there is a deviation between the three-dimensional motion trajectory and the preset motion trajectory according to the point cloud matching result. If there is a deviation, multiple deviation points are determined, and a comparison deviation coefficient between the three-dimensional motion trajectory and the preset motion trajectory is calculated based on the multiple deviation points.

[0055] Specifically, the scanning unit is a device or component used to scan the unhooking robot to obtain information such as its surface shape and spatial position, such as a laser scanner. The point cloud data acquisition unit activates the scanning unit to start scanning the unhooking robot to obtain the position information of each point on the surface of the unhooking robot. For each scanning point of the unhooking robot, the scanning unit records its three-dimensional coordinates (x, y, z) and possible other properties (such as reflectivity, etc.). As the scanning proceeds, the information of these points is continuously collected, and finally the point cloud data set of the unhooking robot is drawn. The point cloud data set provides an accurate three-dimensional spatial data foundation for subsequent operations such as motion analysis and trajectory comparison of the unhooking robot, and can intuitively reflect the appearance shape and spatial position information of the unhooking robot.

[0056] The motion analysis unit analyzes the motion process of the unhooking robot based on the point cloud data set, including analysis of motion trajectory, speed, acceleration, etc. First, the point cloud data of continuous frames are analyzed. For each point cloud data frame, the positions of each key point (such as joint points, endpoints, etc.) on the unhooking robot are determined. Then, as time goes by, the positions of these key points are connected to extract the three-dimensional motion trajectory of the unhooking robot. For example, if the robotic arm of the unhooking robot is moving, the three-dimensional motion trajectory of the robotic arm endpoint is constructed by analyzing the position changes of the robotic arm endpoint in different point cloud data frames. In the motion analysis process, an algorithm based on feature point matching can be used to identify feature points (such as corner points, edge points, etc.) in the point cloud data set and match them between different frames, and combine the interpolation algorithm to construct a continuous motion trajectory between discrete point cloud data points.

[0057] The preset motion trajectory determination unit maps the operation status evaluation information to the point cloud data set for motion analysis. The operation status evaluation information contains the normal operating parameters of the unhooking robot in different working modes, such as speed range, torque limit, etc. Based on this information, the structure and kinematic model of the unhooking robot are analyzed in combination with the point cloud data set. For example, if the operation status evaluation information indicates that the movement speed of the robot arm should be between v1 and v2 in a certain working mode, and the structural information such as the length of the robot arm is known, then the ideal motion path of the robot arm under these speed restrictions is determined based on the point cloud data set, thereby determining the preset motion trajectory, providing a standard reference for subsequent trajectory comparison.

[0058] The point cloud matching unit uses the iterative closest point (ICP) algorithm to match the 3D motion trajectory with the preset motion trajectory according to the point cloud data set. First, select the initial corresponding point pair, and then calculate the transformation (such as translation, rotation, etc.) between the two point clouds to minimize the sum of the distances between the corresponding point pairs. This process is iterated until the convergence conditions are met, and the point cloud matching results are generated, including the degree of matching, the correspondence between the matching points, and other information. The point cloud matching results can accurately reflect the relationship between the 3D motion trajectory and the preset motion trajectory, and provide a direct basis for judging whether there is a deviation in the trajectory.

[0059] The deviation identification unit determines whether there is a deviation between the three-dimensional motion trajectory and the preset motion trajectory according to the point cloud matching results. If there is a deviation, the distances between the corresponding points in the matched point cloud data are compared to determine multiple deviation points, that is, the positions of the points with deviations. For example, if the difference between the coordinates of a point in the three-dimensional motion trajectory and the coordinates of the corresponding point in the preset motion trajectory exceeds a certain threshold, it is determined as a deviation point. Then, based on these deviation points, the overall error degree is quantified using methods such as mean square error (MSE) and maximum deviation calculation, and the comparison deviation coefficient between the three-dimensional motion trajectory and the preset motion trajectory is calculated. By calculating the comparison deviation coefficient, the degree of deviation between the three-dimensional motion trajectory and the preset motion trajectory can be quantitatively described, which provides a quantitative basis for judging whether there is a fault in the unhooking robot and the severity of the fault, thereby improving the accuracy of fault identification.

[0060] Furthermore, the fault detection module 30 in the embodiment of the present application further includes: The judgment unit is used to set a deviation critical value according to the preset motion trajectory and judge whether the comparison deviation coefficient is greater than the deviation critical value.

[0061] The detection unit is used to generate a detection instruction when the comparison deviation coefficient is greater than the deviation critical value, and to perform continuous fault detection on the unhooking robot through the detection instruction combined with the three-dimensional motion trajectory to generate a fault detection result.

[0062] An abnormality analysis unit is used to perform abnormality analysis based on the fault detection result in combination with the comparison deviation coefficient, identify the fault mode of the unhooking robot, and add the fault mode to the fault self-check report.

[0063] The troubleshooting unit is used to generate a troubleshooting instruction when the comparison deviation coefficient is less than or equal to the deviation critical value, and continuously troubleshoot the unhooking robot through the troubleshooting instruction in combination with the three-dimensional motion trajectory to generate a troubleshooting result.

[0064] A state analysis unit is used to perform state analysis based on the troubleshooting result in combination with the comparison deviation coefficient, identify the state type of the unhooking robot, and add the state type to the fault self-check report.

[0065] Specifically, the judgment unit sets a deviation critical value according to the characteristics of the preset motion trajectory and the normal operation requirements of the hook removal robot, which is used to measure the acceptable degree of deviation between the three-dimensional motion trajectory and the preset motion trajectory. Then, the previously calculated comparison deviation coefficient is compared with the deviation critical value to determine whether the comparison deviation coefficient is greater than the deviation critical value.

[0066] When the comparison deviation coefficient is greater than the deviation critical value, the detection unit generates a detection instruction. This detection instruction contains the specific content to be detected, such as detailed inspection requirements for the joints, sensors and other components of the robotic arm. Then, the unhooking robot is continuously detected for faults according to the detection instruction combined with the three-dimensional motion trajectory. For example, the detection instruction requires the detection of the force conditions during the movement of the robotic arm, and the position and speed information of the robotic arm in the three-dimensional motion trajectory are combined to analyze which positions are abnormally stressed. Through sensor data acquisition, mechanical structure inspection and other detection methods, a fault detection result is generated. This result includes the specific location of the fault (such as wear of a joint), the type of fault (such as mechanical fault or electrical fault), etc., which can accurately determine whether the unhooking robot has a fault and the specific situation of the fault, providing an important basis for subsequent fault mode identification and maintenance.

[0067] The abnormality analysis unit performs abnormality analysis based on the fault detection results combined with the comparison deviation coefficient. The comparison deviation coefficient can reflect the severity of the fault, and the fault detection result provides specific information about the fault. A fault tree can be constructed based on the fault detection results, and the fault mode can be identified by gradually analyzing the top event (such as abnormal movement of the robot arm) to the basic event (such as joint wear). For example, if the fault detection result shows that a joint of the robot arm moves abnormally and the comparison deviation coefficient is large, by analyzing the kinematic model and force conditions of the robot arm, it may be possible to identify the fault mode as decreased motion accuracy caused by joint wear. Then, the identified fault mode is added to the fault self-check report, so that maintenance personnel can intuitively understand the fault condition of the unhooking robot.

[0068] When the comparison deviation coefficient is less than or equal to the deviation critical value, the troubleshooting unit generates a troubleshooting instruction. The troubleshooting instruction includes inspection requirements for some key components or operating parameters of the unhooking robot. Then, the unhooking robot is continuously troubleshooted according to the troubleshooting instruction combined with the three-dimensional motion trajectory. For example, the troubleshooting instruction may require checking the sensor accuracy of the robotic arm, and analyzing whether the sensor data is accurate in combination with the motion data of the robotic arm in the three-dimensional motion trajectory. Through the troubleshooting operation, a troubleshooting result is finally generated, which may indicate whether the unhooking robot has potential fault hazards or whether the fault has been completely eliminated.

[0069] The state analysis unit performs state analysis based on the troubleshooting results combined with the comparison deviation coefficient, and determines the state type according to pre-set rules (such as different ranges of the comparison deviation coefficient correspond to different state types). If the troubleshooting results show that the unhooking robot has no potential fault hazards and the comparison deviation coefficient is within the normal range, then the state type may be identified as a normal state. If there are some small deviations but have not reached the level of failure, it may be identified as a potential fault state. Then, the identified state type is added to the fault self-check report so that the operator can fully understand the operating status of the unhooking robot. The state type is identified through state analysis and added to the fault self-check report to facilitate monitoring and management of the overall operating status of the unhooking robot.

[0070] Furthermore, the fault tracing module 40 in the embodiment of the present application further includes: A fault mode extraction unit is used to extract the fault mode of the unhooking robot based on the fault self-check report for data analysis to obtain a fault self-check data set.

[0071] The evolution path construction unit is used to synchronize the fault self-check data set to the human-machine interface for time evolution and construct a fault evolution path.

[0072] The multi-dimensional correlation analysis unit is used to perform multi-dimensional correlation analysis on the unhooking robot according to the fault evolution path through a human-machine interface to generate a correlation decay factor.

[0073] The source tracing and locating unit is used to trace and locate the fault evolution path according to the associated decay factor, and determine multiple fault source locating information.

[0074] The fault source data determining unit is used to add the plurality of fault source location information to the fault source data.

[0075] Specifically, the fault mode extraction unit extracts key information from the fault self-test report, such as the fault mode, comparison deviation coefficient, fault detection result, etc., and organizes the extracted data into a structured data set to obtain a fault self-test data set for subsequent analysis.

[0076] The fault evolution path describes the development and change of the fault mode over time, including the characteristics of each stage in the process from the initial state to the final state of the fault. The evolution path construction unit synchronizes the fault self-check data set to the human-machine interface, and arranges and displays the data in chronological order on the human-machine interface. For example, if the fault self-check data set contains data on the wear of the robot joints at different time points, then the human-machine interface will draw a curve or display a data table with time as the horizontal axis and the degree of wear, related working parameters, etc. as the vertical axis. In this way, a fault evolution path is constructed to intuitively present the development of the fault over time, such as whether the wear is gradually increasing or suddenly deteriorating. The fault evolution path can clearly show the development process of the fault, which helps to better understand the generation and development mechanism of the fault.

[0077] The multi-dimensional correlation analysis unit performs a multi-dimensional correlation analysis on the unhooking robot according to the fault evolution path on the human-machine interface, analyzes the fault evolution path from multiple dimensions such as time, space, and different fault characteristics, and finds the correlation between the dimensions. For example, the correlation between the fault development speed and the working environment temperature of the robot is analyzed from the time dimension, and the correlation between the wear of different joints of the robot arm is analyzed from the space dimension. By analyzing the mutual influence and change rules between data of different dimensions, the correlation decay factor is calculated. The correlation decay factor is used to measure the correlation changes during the development of the fault, and can reflect the changes in the correlation between each dimension during the development of the fault, providing an important basis for fault tracing and positioning. Exemplarily, this correlation decay factor can be obtained by quantifying the changes in the correlation coefficient (such as the Pearson correlation coefficient) between data of different dimensions over time or other factors.

[0078] The source tracing and positioning unit traces and locates the fault evolution path according to the associated decay factor to determine the source location or starting cause of the fault, that is, to find out where the fault first occurred or what caused the fault to start. If the associated decay factor shows that the correlation between two dimensions has changed significantly (such as from strong correlation to weak correlation) at a certain point in time or under a certain working state, then an in-depth analysis is carried out starting from this change point. For example, if the correlation between the wear degree of the robot arm and the workload suddenly decays at a certain moment, the working condition of the robot arm before and after this moment, the load change, the surrounding environment and other factors are further analyzed to determine multiple fault source location information. It may be that the sudden damage of a component, the abnormal increase in workload and other reasons caused the start of the fault. This process can be completed using a causal analysis algorithm, such as the Granger causality test algorithm, which is used to determine the causal relationship between different factors to determine the source of the fault. Determining the fault source location information through source tracing and positioning helps to fundamentally solve the fault problem of the unhooking robot and improve maintenance efficiency and robot reliability.

[0079] The fault source data determination unit adds the determined fault source location information to the fault source data in a specific format (such as a data record). For example, if the fault source location information includes the name of the fault source (such as a sensor fault), the location of the fault source (such as at the joint of the robot arm), the time when the fault source occurs, etc., this information is added to the fault source data according to a certain field structure, which facilitates the unified management and analysis of the fault source and provides data support for subsequent fault prevention and improvement measures.

[0080] Through the above process, not only can the evolution process of the fault be displayed intuitively, but also the correlation between the fault and other factors can be clarified through multi-dimensional correlation analysis, and finally the root cause of the fault can be accurately located, which provides clear guidance for subsequent fault handling and significantly improves the efficiency and accuracy of fault handling.

[0081] Furthermore, the fault processing module 50 in the embodiment of the present application includes: The fault risk assessment unit is used to traverse multiple components of the unhooking robot to perform fault risk assessment according to the multiple fault source location information and generate multiple fault risk levels.

[0082] A fault sorting unit is used to prioritize the multiple fault source location information based on the multiple fault risk levels and generate a fault to-be-processed sequence.

[0083] The operation status matching unit is used to match the operation status of the unhooking robot according to the fault to be processed sequence, and generate a fault status alarm signal according to the matching result.

[0084] The status tracking unit is used to retrieve historical maintenance records, perform status tracking according to the fault status alarm signal, obtain fault status trend data, and formulate the fault handling suggestions for the unhooking robot based on the fault status trend data and the historical maintenance records.

[0085] Specifically, the fault risk assessment unit traverses multiple related components of the unhooking robot based on the fault source location information, conducts risk assessment on each fault source, and considers the impact of the fault on the operation, safety, and production of the unhooking robot. Based on the assessment results, a fault risk level is assigned to each fault source. For example, it can be divided into three levels: low, medium, and high.

[0086] The fault sorting unit sorts multiple fault source location information according to the generated multiple fault risk levels. For example, using the bubble sorting algorithm, starting from the first fault source location information, compare it with the fault risk level of the next fault source location information. If the risk level of the previous one is lower than that of the next one, swap their positions, and so on. After multiple rounds of comparison, the fault source location information with a high risk level can be placed in front, thereby generating a sequence of faults to be processed. This sequence clarifies the order of fault processing. During the sorting process, if there are multiple fault source location information with the same risk level, they can also be sorted again according to other factors (such as the maintenance cost of the components involved in the fault source location information, and the ones with low maintenance costs are processed first). Through priority sorting, the order of fault processing is clarified and the efficiency of fault processing is improved.

[0087] The operation state matching unit starts to match the operation state of the unhooking robot according to the content in the fault-to-be-processed sequence to check whether there is a corresponding fault state. First, the current operation state data of the unhooking robot should be collected, such as the temperature, movement speed, current and other parameters of each component. For example, if there is information about the motor fault in the fault-to-be-processed sequence, check whether the current current of the motor is abnormal and whether the temperature is too high. The state monitoring system can be used to collect data, and then the pattern recognition algorithm can be used for matching. If the current of the motor exceeds the normal range and the temperature is also high, it is determined to match the motor fault in the fault-to-be-processed sequence. A fault state alarm signal is generated based on this matching result. This signal can be a specific digital code (such as 1 for motor fault alarm) or an audio-visual signal (such as a flashing red light for a fault). Through the operation state matching, a fault state alarm signal is generated in time to remind the operator to handle the fault, thereby reducing downtime and improving the fault response speed.

[0088] The state tracking unit retrieves the historical maintenance records related to the fault from the database, including the fault handling history, maintenance time, maintenance content, etc. Then, the state of the unhooking robot is tracked according to the fault state alarm signal, and fault state trend data is generated. These data reflect the development trend of the fault state over time, such as whether the fault parameters (temperature, pressure, etc.) of a certain component are rising or falling. For example, if the fault state alarm signal is that the temperature of a certain component is too high, the temperature change of this component is continuously monitored, and the temperature data is recorded every 5 minutes. These data are plotted into a curve, which is the fault state trend data. According to the fault state trend data and historical maintenance records, fault handling suggestions are formulated. For example, if the fault state trend data shows that the fault of a certain component develops gradually, and there are multiple ways to handle the fault in the historical maintenance records, such as repair, replacement of parts, adjustment of working parameters, etc., then different fault handling suggestions can be formulated. Then, according to the current work requirements, maintenance resources and other actual conditions of the unhooking robot, the most appropriate fault handling suggestion is selected. Combining historical maintenance records and state tracking, more reasonable fault handling suggestions can be formulated, which improves the efficiency of fault handling and the availability of the unhooking robot, and provides strong support for the stable operation of the unhooking robot.

[0089] Furthermore, the fault processing module 50 in the embodiment of the present application further includes: The simulation execution unit is used to simulate the execution of fault handling suggestions on the unhooking robot and generate a fault handling report.

[0090] A real-time feedback unit is used to provide real-time feedback on the fault handling suggestion based on the fault handling report, update the fault handling suggestion, and generate a fault handling optimization suggestion.

[0091] The fault visualization unit is used to send the fault handling optimization suggestion to the human-machine interface for fault visualization and generate a fault visualization interface.

[0092] The remote dynamic monitoring unit is used to perform remote dynamic monitoring by executing the fault handling optimization suggestion through the fault visualization interface and generate a remote control command.

[0093] A fault repair control unit is used to execute the remote control command through a human-machine interface to perform fault repair control on the hook removal robot.

[0094] Specifically, the simulation execution unit creates a virtual environment similar to the actual unhooking robot through simulation software (such as MATLAB / Simulink, V-REP), operates according to the fault handling suggestions in the simulation environment, records the results of the simulation execution, and generates a fault handling report. This fault handling report records the results of the simulation execution, including the processing process, expected results, and potential problems. By simulating the execution of fault handling suggestions on the unhooking robot, the effectiveness of the handling suggestions can be pre-evaluated without actually affecting the normal operation of the robot (if it is still running) or avoiding unnecessary risks (such as avoiding more serious failures due to incorrect handling), providing detailed reference information for subsequent optimization.

[0095] The real-time feedback unit analyzes the data in the report through a data analysis algorithm. For example, if the fault handling report shows that the pressure of a component in the simulation execution exceeds the normal range after processing, it can be determined that there is a problem with the original fault handling suggestion. Data mining technology can be used to extract key information from the report, and then the original fault handling suggestion can be modified based on this information. For example, if the original suggestion is to increase the power of a component, and the report shows that the increase in power has caused other problems, it can be adjusted to appropriately reduce the power and add other auxiliary measures, thereby generating fault handling optimization suggestions. Through timely feedback and optimization, the accuracy and effectiveness of fault handling suggestions can be improved, the fault handling plan can be made more scientific and reasonable, and the secondary faults and other problems that may occur when handling faults can be reduced.

[0096] The fault visualization unit converts the fault handling optimization suggestions into data format and displays them on the human-machine interface in the form of intuitive graphics, images or visualization interfaces, which is easy for operators to understand. For example, if the optimization suggestions involve some complex parameter adjustments, these parameters can be presented in the form of charts or graphs. Graphic drawing tools (such as OpenGL, etc.) can be used to create visualization elements, and the various parts of the fault handling optimization suggestions (such as operating steps, involved components, etc.) can be combined in an intuitive way on the fault visualization interface to facilitate operators to quickly understand the fault handling plan, improve the efficiency of fault handling, and reduce operational errors caused by misunderstandings.

[0097] The remote dynamic monitoring unit establishes a connection with the unhooking robot through a network communication protocol (such as TCP / IP protocol), and then uses the fault visualization interface to execute fault handling optimization suggestions. During the execution process, remote dynamic monitoring is performed, and sensors (such as temperature sensors, pressure sensors, etc.) are used to collect real-time operating data of the unhooking robot, such as the force of the mechanical arm, the temperature of the motor, etc. These data are transmitted to the remote monitoring center through the network. The monitoring center uses data analysis software to analyze the data. If the data is found to be abnormal (such as the motor temperature is too high), a remote control command is generated according to the preset rules (such as reducing the motor power when the motor temperature exceeds 80°C).

[0098] The fault repair control unit is on the human-machine interface. After the operator confirms the remote control command, the human-machine interface sends the command to the control system of the unhooking robot through a communication link (such as industrial Ethernet, etc.) to realize the fault repair control of the unhooking robot. In this process, a signal conversion device can be used to convert the digital command into an electrical signal suitable for the device to receive. The remote control command is executed through the human-machine interface to realize the precise fault repair control of the unhooking robot, ensuring that the fault can be effectively handled and improving the availability and reliability of the equipment.

[0099] In summary, the state self-checking system of the unhooking robot provided in the embodiment of the present application has the following technical effects: In general, the embodiment of the present application realizes comprehensive monitoring and efficient maintenance of the state of the unhooking robot by using the real-time acquisition module 10 to fully perceive the operating status of the unhooking robot, the operation evaluation module 20 to conduct comprehensive analysis and judgment, the fault detection module 30 to accurately locate the fault, the fault tracing module 40 to quickly guide the operator to find the root cause of the fault, and the fault handling module 50 to provide solutions and support remote operation. This system not only improves the accuracy and efficiency of fault detection, but also significantly shortens the fault handling time through intelligent fault handling suggestions and remote control functions, improves the operating reliability and production efficiency of the unhooking robot, and enables it to better adapt to production needs in complex industrial environments.

[0100] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The self-checking system of the unhooking robot is characterized by: The system comprises: A real-time acquisition module is used to perform sensing by traversing multiple components of the hook-removing robot through a sensing unit, and to collect a real-time working data set of the hook-removing robot; An operation evaluation module, used to perform a comprehensive operation evaluation on the unhooking robot based on the real-time working data set and generate operation status evaluation information; A fault detection module, used for performing point cloud matching on the unhooking robot based on the operation status evaluation information, performing data comparison according to the point cloud matching result, generating a comparison deviation coefficient, performing fault detection according to the comparison deviation coefficient, and generating a fault self-check report; A fault tracing module is used to synchronize the fault self-check report to the human-machine interface for fault tracing and determine the fault source data; The fault handling module is used to monitor the status of the unhooking robot according to the fault source data, formulate fault handling suggestions, send the fault handling suggestions to the human-machine interface, and remotely control the unhooking robot through the human-machine interface.

2. The state self-checking system of the hook-removing robot according to claim 1, characterized in that: The operation evaluation module also includes: A cluster analysis unit, configured to traverse the real-time working data set to perform cluster analysis, determine a plurality of operating data, divide the real-time working data set into data using the plurality of operating data as cluster centers, and generate a plurality of operating data clusters; A long-term operation evaluation unit, used for identifying the coupler type of the unhooking robot, performing long-term operation evaluation on the plurality of operation data clusters according to the coupler type, and generating a long-term status score; A short-term operation evaluation unit, configured to perform short-term operation evaluation on the plurality of operation data clusters according to the coupler type to generate a short-term status score; A state verification unit, configured to perform incremental learning on the long-term state score and the short-term state score to generate a learning result, perform state verification based on the learning result, and obtain a state verification token; A reverse screening unit is used to perform reverse screening on the multiple operation data clusters according to the status verification token to generate the operation status evaluation information.

3. The state self-checking system of the hook-removing robot according to claim 2, characterized in that: The state verification unit is also used to perform the following steps: Performing state feature analysis on the unhooking robot based on the multiple operation data clusters, and constructing multiple operation feature vectors; Performing standardized learning on the long-term state score and the short-term state score according to the multiple operation feature vectors to generate multiple standard sample parameters, wherein the multiple standard sample parameters include long-term standard sample parameters and short-term standard sample parameters; The long-term standard sample parameters are adjusted by using stochastic gradient descent to generate a long-term training parameter set, and the short-term standard sample parameters are adjusted by using stochastic gradient descent to generate a short-term training parameter set; Based on the long-term training parameter set, the long-term standard sample parameters are incrementally learned in a time series to generate a long-term learning result; based on the short-term training parameter set, the short-term standard sample parameters are incrementally learned in a time series to generate a short-term learning result; The long-term learning result and the short-term learning result are added to the learning result.

4. The state self-checking system of the hook-removing robot according to claim 1, characterized in that: The fault detection module comprises: A point cloud data acquisition unit is used to activate the scanning unit to scan the hook removing robot and draw a point cloud data set of the hook removing robot; A motion analysis unit, used to perform motion analysis on the hook-removing robot based on the point cloud data set and extract a three-dimensional motion trajectory; A preset motion trajectory determination unit, used for mapping the running status evaluation information to the point cloud data set to perform motion analysis and determine a preset motion trajectory; A point cloud matching unit, used to match the three-dimensional motion trajectory with the preset motion trajectory according to the point cloud data set to generate a point cloud matching result; The deviation identification unit is used to determine whether there is a deviation between the three-dimensional motion trajectory and the preset motion trajectory according to the point cloud matching result. If there is a deviation, multiple deviation points are determined, and a comparison deviation coefficient between the three-dimensional motion trajectory and the preset motion trajectory is calculated based on the multiple deviation points.

5. The state self-checking system of the hook-removing robot according to claim 4, characterized in that: The fault detection module also includes: A judgment unit, configured to set a deviation critical value according to the preset motion trajectory, and judge whether the comparison deviation coefficient is greater than the deviation critical value; A detection unit, used for generating a detection instruction when the comparison deviation coefficient is greater than the deviation critical value, performing continuous fault detection on the hook-removing robot through the detection instruction combined with the three-dimensional motion trajectory, and generating a fault detection result; an abnormality analysis unit, configured to perform abnormality analysis based on the fault detection result in combination with the comparison deviation coefficient, identify the fault mode of the unhooking robot, and add the fault mode to the fault self-check report; A troubleshooting unit, used for generating a troubleshooting instruction when the comparison deviation coefficient is less than or equal to the deviation critical value, and continuously troubleshooting the unhooking robot through the troubleshooting instruction in combination with the three-dimensional motion trajectory to generate a troubleshooting result; A state analysis unit is used to perform state analysis based on the troubleshooting result in combination with the comparison deviation coefficient, identify the state type of the unhooking robot, and add the state type to the fault self-check report.

6. The state self-checking system of the hook-removing robot according to claim 5, characterized in that: The fault tracing module comprises: A fault mode extraction unit, used for extracting the fault mode of the unhooking robot based on the fault self-check report for data analysis to obtain a fault self-check data set; An evolution path construction unit, used to synchronize the fault self-check data set to a human-machine interface for time evolution, and to construct a fault evolution path; A multi-dimensional correlation analysis unit, used to perform a multi-dimensional correlation analysis on the unhooking robot according to the fault evolution path through a human-machine interface to generate a correlation decay factor; A source tracing and locating unit, configured to trace and locate the fault evolution path according to the associated decay factor, and determine multiple fault source locating information; The fault source data determining unit is used to add the plurality of fault source location information to the fault source data.

7. The state self-checking system of the hook-removing robot according to claim 6, characterized in that: The fault processing module comprises: A fault risk assessment unit, configured to traverse multiple components of the unhooking robot to perform fault risk assessment according to the multiple fault source location information, and generate multiple fault risk levels; A fault sorting unit, configured to prioritize the plurality of fault source location information based on the plurality of fault risk levels, and generate a sequence of faults to be processed; An operation status matching unit is used to match the operation status of the unhooking robot according to the fault to be processed sequence, and generate a fault status alarm signal according to the matching result; The status tracking unit is used to retrieve historical maintenance records, perform status tracking according to the fault status alarm signal, obtain fault status trend data, and formulate the fault handling suggestions for the unhooking robot based on the fault status trend data and the historical maintenance records.

8. The state self-checking system of the hook-removing robot according to claim 1, characterized in that: The fault processing module also includes: A simulation execution unit is used to simulate and execute fault handling suggestions on the unhooking robot and generate a fault handling report; A real-time feedback unit, configured to provide real-time feedback on the fault handling suggestion based on the fault handling report, update the fault handling suggestion, and generate a fault handling optimization suggestion; A fault visualization unit, used for sending the fault handling optimization suggestion to a human-machine interface for fault visualization and generating a fault visualization interface; A remote dynamic monitoring unit, used to perform remote dynamic monitoring and generate remote control commands by executing the fault handling optimization suggestions through the fault visualization interface; A fault repair control unit is used to execute the remote control command through a human-machine interface to perform fault repair control on the hook removal robot.

Images