A method and system for assessing the health of intelligent warehousing and logistics robots

Abstract

This invention provides a method and system for assessing the health of intelligent warehousing and logistics robots, relating to the field of equipment assessment technology. The method includes: acquiring historical operational data of logistics robots under different operating scenarios, preprocessing and classifying the data; clustering and filtering historical samples using the DBSCAN algorithm to acquire real-time operational data, calculating the deviation between real-time operational data and historical samples to obtain the aging degree; calculating external fault factors of the logistics robots based on real-time environmental data, and calculating the anomaly degree based on the external fault factors and real-time performance data; constructing a correlation weight matrix for the logistics robots based on the collaborative relationships between them; and comprehensively assessing the health of the logistics robots based on the correlation weight matrix, aging degree, and anomaly degree. This invention solves the technical problem of low accuracy in assessing health using a single operational parameter in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of equipment evaluation technology, specifically to a method and system for evaluating the health of intelligent warehousing and logistics robots. Background Technology

[0002] With the development of smart logistics, logistics robots such as AGV robots and automated warehousing equipment (stacking cranes, sorting machines) have become core assets, and their health status directly affects the efficiency, cost and safety of the logistics system.

[0003] Currently, the health assessment of intelligent warehousing and logistics robots mostly relies on single operating parameters of the logistics robot itself (such as temperature and current), ignoring the environmental interference unique to the logistics scenario, which makes it difficult for the assessment results to reflect the actual operating status of the logistics robot. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for assessing the health of intelligent warehousing and logistics robots, thereby solving the technical problem of low accuracy in assessing health based on a single operating parameter in existing technologies.

[0005] A first aspect of the present invention is to provide a method for assessing the health of an intelligent warehousing and logistics robot, the method comprising:

[0006] Historical operating data of logistics robots under different operating scenarios are obtained. The historical operating data includes historical performance data, historical environmental data and historical task data of logistics robots. The historical operating data is preprocessed and divided according to different operating scenarios.

[0007] Then, based on the task load in the historical task data under different operating scenarios, the historical operating data is divided into historical operating data, and the historical operating data is clustered and historical samples are filtered through the DBSCAN algorithm to obtain the real-time operating data of the logistics robot. The deviation between the real-time operating data and the historical samples is calculated to obtain the aging degree of the logistics robot. The real-time operating data includes the real-time performance data, real-time environmental data and real-time task data of the logistics robot.

[0008] Based on real-time environmental data under different operating scenarios, the external fault factors of the logistics robot are calculated, and then based on the external fault factors and the real-time performance data, the degree of abnormality of the logistics robot is calculated.

[0009] Based on the collaborative relationships between various logistics robots, a correlation weight matrix for logistics robots is constructed. Then, based on the correlation weight matrix, the degree of aging, and the degree of abnormality, the health of the logistics robots is comprehensively evaluated.

[0010] Compared with existing technologies, the beneficial effects of this invention are as follows: By acquiring and preprocessing historical operating data of logistics robots in multiple scenarios, and using the DBSCAN algorithm to cluster and filter samples based on task load, the aging degree of the equipment can be accurately obtained, providing a reliable historical benchmark for subsequent evaluation. By mining the correlation between the environment and faults using historical operating data, external fault factors can be calculated, the impact of the environment on logistics robot faults can be quantified, and the accuracy of fault prediction can be improved. Based on the collaborative relationship of logistics robots, a correlation weight matrix is ​​constructed, and the correlation weight, aging degree, and anomaly degree are integrated to comprehensively evaluate the health. This breaks through the limitation of isolated evaluation of a single logistics robot, realizes system-level health evaluation, accurately reflects the chain effect between logistics robots, and improves the accuracy of health evaluation. Thus, it solves the technical problem of low accuracy of evaluation results when using a single operating parameter to evaluate health in existing technologies.

[0011] According to one aspect of the above technical solution, historical operational data is further divided based on the task load in historical task data under different operating scenarios. The DBSCAN algorithm is then used to cluster and filter historical samples to obtain real-time operational data of the logistics robot. The deviation between the real-time operational data and historical samples is calculated to determine the aging degree of the logistics robot. The real-time operational data includes the logistics robot's real-time performance data, real-time environmental data, and real-time task data. Specifically, this includes:

[0012] The historical running data is then divided according to the task load in the historical task data under different running scenarios, and the historical running data is clustered and historical samples are filtered by the DBSCAN algorithm. The task load includes light-load tasks, medium-load tasks and heavy-load tasks.

[0013] Real-time operational data of the logistics robot is acquired, including real-time performance data, real-time environmental data, and real-time task data. Based on the task load corresponding to the real-time task data, a task load correction factor is calculated using the following formula:

[0014] ,

[0015] in, for Task load correction factor at any time, for The basic weight of the task load corresponding to real-time task data. For parameters, for The task load in real-time task data. for The base load corresponding to the real-time task data;

[0016] Based on the task load correction factor, the deviation between real-time performance data and historical samples is calculated to obtain the aging degree of the logistics robot.

[0017] According to one aspect of the above technical solution, the step of calculating the deviation between real-time performance data and historical samples based on the task load correction factor to obtain the aging degree of the logistics robot specifically includes:

[0018] Based on the task load correction factor, the deviation between real-time performance data and historical samples is calculated using the following formula:

[0019] ,

[0020] in, for Deviation in time, for Real-time performance data at any given moment. This is the mean vector of the historical samples. The covariance matrix of the historical samples;

[0021] Based on the aforementioned deviation, the aging degree of the logistics robot is calculated using the following formula:

[0022] ,

[0023] in, for The aging level of logistics robots at all times. This is the failure threshold under the task load in this operating scenario.

[0024] According to one aspect of the above technical solution, the steps for calculating the external failure factors of a logistics robot based on real-time environmental data under different operating scenarios specifically include:

[0025] Based on historical environmental data and historical operational data under different operating scenarios, a gradient boosting decision tree model is used to establish the correlation between environmental parameters and fault types, and to calculate and output the contribution weight of each environmental parameter to the fault.

[0026] The external failure factor is calculated by weighted summation based on real-time environmental data and contribution weights under different operating scenarios.

[0027] According to one aspect of the above technical solution, the step of calculating the degree of abnormality of the logistics robot based on the external fault factor and the real-time performance data specifically includes:

[0028] Using dynamic density clustering algorithms in historical performance data and historical environment data, core dense regions of normal features are mined, and core clusters of normal features are output.

[0029] Obtain external fault factors within a preset time period, calculate time-series statistical characteristics, and calculate time-domain and frequency-domain characteristics of real-time performance data;

[0030] The time-series statistical features, time-domain features, and frequency-domain features are mapped to the anomaly space to obtain an anomaly feature vector;

[0031] The outlier degree between the abnormal feature vector and the core cluster is calculated to obtain the degree of abnormality.

[0032] According to one aspect of the above technical solution, the step of constructing the association weight matrix of logistics robots based on the collaborative relationship between various logistics robots specifically includes:

[0033] Data on collaboration anomalies among logistics robots is obtained from historical performance data, and task dependencies among logistics robots are obtained from historical task data. Based on the system anomaly data and task dependencies, the association weights of the logistics robots are calculated, and the formula is as follows:

[0034] ,

[0035] in, For logistics robots For logistics robots Association weights For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies To eliminate logistics robots External logistics robots For logistics robots With logistics robots Task dependency between them This represents the maximum value of the task dependency among all logistics robots. For associated parameters;

[0036] Based on the association weights between various logistics robots, an association weight matrix for logistics robots is constructed.

[0037] Based on one aspect of the above technical solution, and further based on the correlation weight matrix, the degree of aging, and the degree of anomaly, the health of the logistics robot is comprehensively evaluated, and the calculation formula is as follows:

[0038] ,

[0039] in, For logistics robots health To eliminate logistics robots All other logistics robots, , They are respectively , In logistics robots With logistics robots The allocation coefficients on the correlation dimension between them , They are respectively Time Logistics Robot Logistics robots The degree of aging, , They are respectively Time Logistics Robot Logistics robots The degree of abnormality, , They are respectively , In logistics robots The distribution coefficient on, , This is the health weighting coefficient.

[0040] A second aspect of this application is to provide a health assessment system for intelligent warehousing and logistics robots, the system being used to execute the aforementioned health assessment method for intelligent warehousing and logistics robots, the system comprising:

[0041] The data acquisition module is used to acquire historical operating data of logistics robots under different operating scenarios. The historical operating data includes historical performance data, historical environmental data, and historical task data of logistics robots. The historical operating data is preprocessed and divided according to different operating scenarios.

[0042] The aging degree calculation module is used to further divide the historical operation data according to the task load in the historical task data under different operating scenarios, and to cluster and filter the historical operation data through the DBSCAN algorithm to obtain the real-time operation data of the logistics robot, calculate the deviation between the real-time operation data and the historical samples, and obtain the aging degree of the logistics robot. The real-time operation data includes the real-time performance data, real-time environmental data and real-time task data of the logistics robot.

[0043] The anomaly degree calculation module is used to calculate the external fault factors of the logistics robot based on real-time environmental data under different operating scenarios, and then calculate the anomaly degree of the logistics robot based on the external fault factors and the real-time performance data.

[0044] The health assessment module is used to construct a correlation weight matrix for logistics robots based on the collaborative relationship between them, and then comprehensively assess the health of the logistics robots based on the correlation weight matrix, the degree of aging, and the degree of abnormality.

[0045] A third aspect of this application is to provide a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described intelligent warehousing and logistics robot health assessment method.

[0046] A fourth aspect of this application is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the above-described intelligent warehousing and logistics robot health assessment method. Attached Figure Description

[0047] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0048] Figure 1 This is a flowchart illustrating the health assessment method for intelligent warehousing and logistics robots in Embodiment 1 of the present invention.

[0049] Figure 2 This is a structural block diagram of the intelligent warehousing and logistics robot health assessment system in Embodiment 2 of the present invention;

[0050] Component symbol explanation in the attached diagram:

[0051] Data acquisition module 100, aging degree calculation module 200, abnormality degree calculation module 300, health assessment module 400;

[0052] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0053] To make the objectives, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Several embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of the present invention will be more thorough and complete.

[0054] Example 1

[0055] Please see Figure 1The first embodiment of the present invention provides a method for assessing the health of an intelligent warehousing and logistics robot, the method comprising steps S10-S13:

[0056] Step S10: Obtain historical operating data of the logistics robot under different operating scenarios. The historical operating data includes historical performance data, historical environmental data, and historical task data of the logistics robot. The historical operating data is preprocessed and divided according to different operating scenarios.

[0057] Historical performance data includes performance data extracted from various sensors mounted on logistics robots (such as AGVs, stacker cranes, and transport vehicles). These data may include battery SOC (remaining charge), motor operating current, operating speed, fork positioning error, engine speed, fuel consumption, tire pressure, braking system temperature, lifting motor torque, slewing mechanism speed, and fork load capacity.

[0058] Furthermore, historical environmental data can be obtained from environmental monitoring equipment deployed in logistics parks and along transportation routes, such as data recorded by temperature and humidity sensors, dust concentration sensors, and light sensors in warehouses.

[0059] Furthermore, historical task data can be extracted from the enterprise's logistics management system (such as warehouse management system, WMS), including: task type (such as inbound, outbound, transportation, sorting, etc.), task execution time, information on the goods involved in the task (weight, volume, quantity, etc.), and the start and end points of the task.

[0060] In addition, preprocessing includes data cleaning and data standardization. Specifically, data cleaning involves handling missing values ​​(e.g., imputing them using methods such as mean or interpolation), outliers (removing data that does not conform to the pattern using statistical methods or algorithms), and duplicate values ​​(deleting duplicate records) to ensure the accuracy and consistency of the data. Secondly, data standardization (e.g., mapping data to a preset interval to eliminate differences in units), normalization (making the data distribution more consistent with the requirements of the analysis or model), and aggregation (summarizing data according to certain rules, such as calculating statistics such as mean and sum) are performed to make the data more comparable and applicable.

[0061] It should be noted that, based on the geographical coordinates (such as GPS positioning) and environmental characteristics (such as warehouse) in the historical operation data, the historical operation data is divided into subdivided scenarios such as warehouse racking area and loading and unloading area to ensure that the pattern of environmental interference on logistics robots is consistent.

[0062] Step S11: Then, based on the task load in the historical task data under different operating scenarios, the historical operating data is divided into historical operating data, and the historical operating data is clustered and historical samples are filtered through the DBSCAN algorithm to obtain the real-time operating data of the logistics robot. The deviation between the real-time operating data and the historical samples is calculated to obtain the aging degree of the logistics robot. The real-time operating data includes the real-time performance data, real-time environmental data and real-time task data of the logistics robot.

[0063] Specifically, the historical running data is further divided according to the task load in the historical task data under different running scenarios, and the historical running data is clustered and historical samples are filtered by the DBSCAN algorithm. The task load includes light-load tasks, medium-load tasks and heavy-load tasks.

[0064] Historical samples can provide a key benchmark for logistics robots in the initial operation phase of a logistics system. For example, historical operating data from the first 30 days can accurately characterize the core performance indicators of a logistics robot in its unaged state, so as to determine the degree of robot aging later.

[0065] Real-time operational data of the logistics robot is acquired, including real-time performance data, real-time environmental data, and real-time task data. Based on the task load corresponding to the real-time task data, a task load correction factor is calculated using the following formula:

[0066] ,

[0067] in, for Task load correction factor at any time, for The basic weight of the task load corresponding to real-time task data. For parameters, for The task load in real-time task data. for The base load corresponding to the real-time task data;

[0068] Among them, heavy-load tasks have a more significant impact on the degree of aging than light-load tasks (for example, the battery cycle life decays faster under heavy load). By quantifying this difference through correction factors, the assessment can better reflect the actual stress state of logistics robots.

[0069] Based on the task load correction factor, the deviation between real-time performance data and historical samples is calculated to obtain the aging degree of the logistics robot, specifically:

[0070] Based on the task load correction factor, the deviation between real-time performance data and historical samples is calculated using the following formula:

[0071] ,

[0072] in, for Deviation in time, for Real-time performance data at any given moment. This is the mean vector of the historical samples. The covariance matrix of the historical samples;

[0073] Based on the aforementioned deviation, the aging degree of the logistics robot is calculated using the following formula:

[0074] ,

[0075] in, for The aging level of logistics robots at all times. This is the failure threshold under the task load in this operating scenario.

[0076] It should be noted that this embodiment not only considers the absolute deviation of performance parameters, but also incorporates the influence of load to accurately reflect the actual aging degree of the logistics robot.

[0077] Step S12: Calculate the external fault factors of the logistics robot based on real-time environmental data under different operating scenarios, and then calculate the degree of abnormality of the logistics robot based on the external fault factors and the real-time performance data.

[0078] Specifically, based on historical environmental data and historical operational data under different operating scenarios, a gradient boosting decision tree model is used to establish the correlation between environmental parameters and fault types, and to calculate and output the contribution weight of each environmental parameter to the fault.

[0079] The external failure factor is calculated by weighted summation based on real-time environmental data and contribution weights under different operating scenarios.

[0080] Among them, by combining real-time environmental data and contribution weights obtained based on historical environmental data, the calculated external failure factor can reflect the potential risk of logistics robot failure under the current environmental conditions in real time.

[0081] Next, dynamic density clustering algorithm is used in historical performance data and historical environment data to mine the core dense regions of normal features and output the core clusters of normal features.

[0082] Obtain external fault factors within a preset time period, calculate time-series statistical characteristics, and calculate time-domain and frequency-domain characteristics of real-time performance data;

[0083] The time-series statistical features, time-domain features, and frequency-domain features are mapped to the anomaly space to obtain an anomaly feature vector;

[0084] The outlier degree between the abnormal feature vector and the core cluster is calculated to obtain the degree of abnormality.

[0085] It should be noted that by calculating time-series statistical features, time-domain features, and frequency-domain features, features are extracted from external fault factors and real-time performance data from different dimensions (overall time-series statistics, local time-domain changes, and frequency-domain component distribution). This comprehensively captures the operational status information of the logistics robot contained in the data, providing rich feature inputs for subsequent anomaly detection. That is, different features reflect the operational status of the logistics robot from different perspectives. Time-series statistical features reflect the overall trend of fault risk, time-domain features reflect the dynamic changes of performance in the time domain, and frequency-domain features reflect the frequency-domain characteristics of performance, which can effectively improve the accuracy of anomaly detection.

[0086] Step S13: Based on the collaborative relationship between various logistics robots, construct the association weight matrix of the logistics robots, and then comprehensively evaluate the health of the logistics robots based on the association weight matrix, the aging degree, and the abnormality degree.

[0087] Specifically, data on collaboration anomalies among logistics robots are obtained from historical performance data, and task dependencies among logistics robots are obtained from historical task data. Based on the system anomaly data and task dependencies, the association weights of the logistics robots are calculated, and the formula is as follows:

[0088] ,

[0089] in, For logistics robots For logistics robots Association weights For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies To eliminate logistics robots External logistics robots For logistics robots With logistics robots Task dependency between them This represents the maximum value of the task dependency among all logistics robots. For associated parameters;

[0090] Based on the association weights between various logistics robots, an association weight matrix for logistics robots is constructed.

[0091] Next, based on the correlation weight matrix, the degree of aging, and the degree of anomaly, the health of the logistics robot is comprehensively evaluated, and the calculation formula is as follows:

[0092] ,

[0093] in, For logistics robots health To eliminate logistics robots All other logistics robots, , They are respectively , In logistics robots With logistics robots The allocation coefficients on the correlation dimension between them , They are respectively Time Logistics Robot Logistics robots The degree of aging, , They are respectively Time Logistics Robot Logistics robots The degree of abnormality, , They are respectively , In logistics robots The distribution coefficient on, , This is the health weighting coefficient.

[0094] Among them, the health status not only reflects the aging and abnormality of the logistics robot itself, but also incorporates the status of all related logistics robots, so as to avoid system failure caused by a single logistics robot being normal but related logistics robots being abnormal.

[0095] The method further includes:

[0096] By combining health status with score ranges and categorizing them into different levels, alarm and maintenance strategies can be formulated. For example, this could be:

[0097] The health score is divided into five levels: Excellent, Good, Medium, Poor, and Critical. Excellent level: Alarms are set to green (normal), and operational data is only recorded in the background system without real-time alarm push notifications. Maintenance employs routine inspections plus preventative maintenance, extending the inspection cycle and focusing on cleaning and lubricating key components of the logistics robot to maintain optimal condition.

[0098] Good Level: Triggers a blue alert. The operations and maintenance system displays a pop-up indicating sub-health status and sends a background notification to maintenance personnel. Targeted inspections are initiated, focusing on subsystems with slightly lower health levels to proactively identify potential problems and prevent further deterioration.

[0099] Intermediate: Upgraded to a yellow alert; a system pop-up and SMS notification will be sent to the responsible person, indicating a potential fault risk. Preventative maintenance will be initiated to analyze the root causes of the health decline and optimize operating parameters.

[0100] Poor Level: Triggers an orange alarm, including audible and visual alarms and an emergency email notification, indicating a significant fault. Repair requires a rapid response and emergency repairs. Non-core tasks should be suspended. The fault point should be located by disassembling the device based on its health status. After replacing the faulty component, the health status should be reassessed to ensure it is restored to a good level or above.

[0101] Critical Level: Activate the red emergency alert, forcibly suspend operations, and notify the maintenance team through multiple channels including phone and app push notifications. Repairs will be emergency repairs, prioritizing the allocation of spare parts and professional personnel for a comprehensive overhaul of the core system; if necessary, replace the faulty module entirely. After repairs, a full-process test must be conducted, and operations can only resume after the system's health has been restored to a good level or above.

[0102] By using tiered alarms and differentiated maintenance strategies, we can avoid the waste of resources caused by over-maintenance, promptly curb the escalation of faults, and ensure the reliable operation of the robot throughout its entire lifecycle.

[0103] Compared with existing technologies, the intelligent warehousing and logistics robot health assessment method shown in this embodiment obtains and preprocesses historical operating data of logistics robots in multiple scenarios, and uses the DBSCAN algorithm to cluster and screen samples based on task load. This can accurately determine the aging degree of the equipment, providing a reliable historical benchmark for subsequent assessments. By mining the correlation between the environment and faults using historical operating data, external fault factors are calculated, and the impact of the environment on logistics robot faults is quantified, improving the accuracy of fault prediction. Based on the collaborative relationship of logistics robots, a correlation weight matrix is ​​constructed, and the health is comprehensively assessed by integrating correlation weights, aging degree, and anomaly degree. This method breaks through the limitation of isolated assessment of a single logistics robot, realizes system-level health assessment, accurately reflects the chain effect between logistics robots, and improves the accuracy of health assessment. Thus, it solves the technical problem of low accuracy in assessing health with a single operating parameter in existing technologies.

[0104] Example 2

[0105] Please see Figure 2 The figure shows a health assessment system for intelligent warehousing and logistics robots according to a second embodiment of the present invention. The system includes:

[0106] The data acquisition module 100 is used to acquire historical operating data of the logistics robot under different operating scenarios. The historical operating data includes historical performance data, historical environmental data and historical task data of the logistics robot. The historical operating data is preprocessed and divided according to different operating scenarios.

[0107] The aging degree calculation module 200 is used to further divide the historical operation data according to the task load in the historical task data under different operating scenarios, and to cluster and filter the historical operation data through the DBSCAN algorithm to obtain the real-time operation data of the logistics robot, calculate the deviation between the real-time operation data and the historical samples, and obtain the aging degree of the logistics robot. The real-time operation data includes the real-time performance data, real-time environmental data and real-time task data of the logistics robot.

[0108] The anomaly degree calculation module 300 is used to calculate the external fault factors of the logistics robot based on real-time environmental data under different operating scenarios, and then calculate the anomaly degree of the logistics robot based on the external fault factors and the real-time performance data.

[0109] The health assessment module 400 is used to construct a correlation weight matrix of logistics robots based on the collaborative relationship between each logistics robot, and then comprehensively assess the health of the logistics robots based on the correlation weight matrix, the degree of aging, and the degree of abnormality.

[0110] In summary, the intelligent warehousing and logistics robot health assessment system in the above embodiments of the present invention can accurately obtain the aging degree of equipment by using the DBSCAN algorithm of the aging degree calculation module to cluster and screen samples, providing a reliable historical benchmark for subsequent assessment. By using the anomaly degree calculation module to mine the correlation between the environment and faults, calculate external fault factors, quantify the impact of the environment on logistics robot faults, and improve the accuracy of fault prediction, the system constructs a correlation weight matrix through the health assessment module, integrates correlation weights, aging degree, and anomaly degree to comprehensively assess health. This system breaks through the limitations of isolated assessment of a single logistics robot, realizes system-level health assessment, accurately reflects the chain effect between logistics robots, and improves the accuracy of health assessment. Thus, it solves the technical problems of low flexibility of full physical simulation and low accuracy of full virtual simulation.

[0111] Example 3

[0112] A third embodiment of the present invention provides a storage medium storing computer instructions that, when executed by a processor, implement the steps of the method described in the first embodiment.

[0113] Example 4

[0114] A fourth embodiment of the present invention provides an apparatus including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method described in the first embodiment.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] Those skilled in the art will understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable instructions for implementing logical functions, and can be embodied in any computer-readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable storage medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0117] More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable storage media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0118] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0119] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0120] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for assessing the health of intelligent warehousing and logistics robots, characterized in that, The method includes: Historical operating data of logistics robots under different operating scenarios are obtained. The historical operating data includes historical performance data, historical environmental data and historical task data of logistics robots. The historical operating data is preprocessed and divided according to different operating scenarios. Then, based on the task load in the historical task data under different operating scenarios, historical operating data is divided, and the historical operating data is clustered and filtered using the DBSCAN algorithm to obtain real-time operating data of the logistics robot. The deviation between the real-time operating data and the historical samples is calculated to obtain the aging degree of the logistics robot. The real-time operating data includes the real-time performance data, real-time environmental data, and real-time task data of the logistics robot, including: The task load includes light-load tasks, medium-load tasks, and heavy-load tasks. Real-time operational data of the logistics robot is acquired, including real-time performance data, real-time environmental data, and real-time task data. Based on the task load corresponding to the real-time task data, a task load correction factor is calculated using the following formula: , in, for Task load correction factor at any time for The basic weight of the task load corresponding to real-time task data. For parameters, for The task load in real-time task data. for The base load corresponding to the real-time task data. Based on the task load correction factor, the deviation between real-time performance data and historical samples is calculated using the following formula: ， in, for Deviation in time, for Real-time performance data at any given moment. This is the mean vector of the historical samples. The covariance matrix of the historical samples, Based on the aforementioned deviation, the degree of aging of the logistics robot is calculated using the following formula: ， in, for The aging level of logistics robots at all times. This is the failure threshold under the task load in this operating scenario; Based on real-time environmental data under different operating scenarios, the external fault factors of the logistics robot are calculated, and then based on the external fault factors and the real-time performance data, the degree of abnormality of the logistics robot is calculated. Based on the collaborative relationships between various logistics robots, a correlation weight matrix for the logistics robots is constructed. Then, based on the correlation weight matrix, the aging degree, and the anomaly degree, the health of the logistics robots is comprehensively evaluated, including: Data on collaboration anomalies among logistics robots is obtained from historical performance data, and task dependencies among logistics robots are obtained from historical task data. Based on the collaboration anomaly data and task dependencies, the association weights of the logistics robots are calculated, and the formula is as follows: ， in, For logistics robots For logistics robots Association weights For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies To eliminate logistics robots External logistics robots For logistics robots With logistics robots Task dependency between them This represents the maximum value of the task dependency among all logistics robots. For associated parameters, Based on the correlation weights between various logistics robots, a correlation weight matrix for logistics robots is constructed. Based on the correlation weight matrix, the degree of aging, and the degree of anomaly, the health of the logistics robot is comprehensively evaluated using the following formula: ， in, For logistics robots health To eliminate logistics robots All other logistics robots , They are respectively , In logistics robots With logistics robots The allocation coefficients on the correlation dimension between them , They are respectively Time Logistics Robot Logistics robots The degree of aging, , They are respectively Time Logistics Robot Logistics robots The degree of abnormality, , They are respectively , In logistics robots The distribution coefficient on, , This is the health weighting coefficient.

2. The method for assessing the health of intelligent warehousing and logistics robots according to claim 1, characterized in that, The steps for calculating the external failure factors of logistics robots based on real-time environmental data under different operating scenarios include: Based on historical environmental data and historical operational data under different operating scenarios, a gradient boosting decision tree model is used to establish the correlation between environmental parameters and fault types, and to calculate and output the contribution weight of each environmental parameter to the fault. The external failure factor is calculated by weighted summation based on real-time environmental data and contribution weights under different operating scenarios.

3. The health assessment method for intelligent warehousing and logistics robots according to claim 2, characterized in that, The step of calculating the degree of abnormality of the logistics robot based on the external failure factors and the real-time performance data specifically includes: Using dynamic density clustering algorithms in historical performance data and historical environment data, core dense regions of normal features are mined, and core clusters of normal features are output. Obtain external fault factors within a preset time period, calculate time-series statistical characteristics, and calculate time-domain and frequency-domain characteristics of real-time performance data; The time-series statistical features, time-domain features, and frequency-domain features are mapped to the anomaly space to obtain an anomaly feature vector; The outlier degree between the abnormal feature vector and the core cluster is calculated to obtain the degree of abnormality.

4. A health assessment system for intelligent warehousing and logistics robots, characterized in that, The system is used to execute the health assessment method for intelligent warehousing and logistics robots according to any one of claims 1 to 3, and the system includes: The data acquisition module is used to acquire historical operating data of logistics robots under different operating scenarios. The historical operating data includes historical performance data, historical environmental data, and historical task data of logistics robots. The historical operating data is preprocessed and divided according to different operating scenarios. The aging degree calculation module is used to further divide historical operation data according to the task load in historical task data under different operating scenarios, and to cluster and filter historical samples using the DBSCAN algorithm to obtain real-time operation data of the logistics robot. It then calculates the deviation between the real-time operation data and historical samples to obtain the aging degree of the logistics robot. The real-time operation data includes real-time performance data, real-time environmental data, and real-time task data of the logistics robot, including: The task load includes light-load tasks, medium-load tasks, and heavy-load tasks. Real-time operational data of the logistics robot is acquired, including real-time performance data, real-time environmental data, and real-time task data. Based on the task load corresponding to the real-time task data, a task load correction factor is calculated using the following formula: , in, for Task load correction factor at any time for The basic weight of the task load corresponding to real-time task data. For parameters, for The task load in real-time task data. for The base load corresponding to the real-time task data. Based on the task load correction factor, the deviation between real-time performance data and historical samples is calculated using the following formula: ， in, for Deviation in time, for Real-time performance data at any given moment. This is the mean vector of the historical samples. The covariance matrix of the historical samples, Based on the aforementioned deviation, the degree of aging of the logistics robot is calculated using the following formula: ， in, for The aging level of logistics robots at all times. This is the failure threshold under the task load in this operating scenario; The anomaly degree calculation module is used to calculate the external fault factors of the logistics robot based on real-time environmental data under different operating scenarios, and then calculate the anomaly degree of the logistics robot based on the external fault factors and the real-time performance data. The health assessment module is used to construct a correlation weight matrix for logistics robots based on the collaborative relationships between them. Then, based on the correlation weight matrix, the degree of aging, and the degree of abnormality, it comprehensively assesses the health of the logistics robots, including: Data on collaboration anomalies among logistics robots is obtained from historical performance data, and task dependencies among logistics robots are obtained from historical task data. Based on the collaboration anomaly data and task dependencies, the association weights of the logistics robots are calculated, and the formula is as follows: ， in, For logistics robots For logistics robots Association weights For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies For logistics robots When an anomaly occurs, it causes the logistics robot The frequency of anomalies To eliminate logistics robots External logistics robots For logistics robots With logistics robots Task dependency between them This represents the maximum value of the task dependency among all logistics robots. For associated parameters, Based on the correlation weights between various logistics robots, a correlation weight matrix for logistics robots is constructed. Based on the correlation weight matrix, the degree of aging, and the degree of anomaly, the health of the logistics robot is comprehensively evaluated using the following formula: ， in, For logistics robots health To eliminate logistics robots All other logistics robots , They are respectively , In logistics robots With logistics robots The allocation coefficients on the correlation dimension between them , They are respectively Time Logistics Robot Logistics robots The degree of aging, , They are respectively Time Logistics Robot Logistics robots The degree of abnormality, , They are respectively , In logistics robots The distribution coefficient on, , This is the health weighting coefficient.

5. A readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the health assessment method for intelligent warehousing and logistics robots as described in any one of claims 1 to 3.

6. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the health assessment method for intelligent warehousing and logistics robots as described in any one of claims 1 to 3.

Images