A robot equipment health status assessment method and system

Abstract

The present invention relates to the technical field of equipment health status assessment, and specifically to a method and system for assessing the health status of a robot device. First, the robot device is monitored during operation to obtain a first data set; the operation trajectory is obtained and the first data set is divided to obtain a first data subset; the motor is monitored during operation to obtain a second data set; the second data set is split according to the operation trajectory to obtain a second data subset. Secondly, a working health status recognition model is constructed to identify the first data subset to obtain working health status data; a motor health status recognition model is constructed to identify the second data subset to obtain motor health status data. Finally, the equipment health status data set is obtained through the working health status data and the motor health status data; a status assessment model is constructed to identify the equipment health status data set to obtain a health status assessment coefficient, and the robot equipment health status is assessed.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment health status assessment, and in particular to a method and system for assessing the health status of a robot equipment. Background Art

[0002] In modern industrial production, the development of automation technology has improved production efficiency, among which fully automatic palletizing robots play an important role in the field of material handling and warehouse management. By efficiently completing the tasks of stacking and handling goods, the intensity of manual labor is reduced and the safety of operations is improved. However, with the increase in work intensity and the extension of usage time, the health status and stability of robots have gradually become the focus of enterprises.

[0003] Existing robot health assessment methods mainly judge the working status of the robot based on parameters such as the robot's running trajectory, running time and speed. These methods provide a certain scientific basis for the health status of robot equipment, but pay less attention to key parameters such as motor load, which is directly related to the operating pressure and wear of key components of the robot. Especially in the application scenario of fully automatic palletizing, the robot's workload will change significantly under different operating conditions, such as cargo type, stacking height and movement speed, which makes traditional health assessment methods insufficient. Therefore, it is difficult to fully reflect the health status of the equipment by relying solely on parameters such as running trajectory or speed. The motor load will be affected by many factors such as the robot's operating environment, movement mode, task load, etc. The changes in parameters such as current, torque, vibration and temperature caused by these factors are also important indicators for evaluating the status of robot equipment.

[0004] Therefore, a robot equipment health status assessment method and system are proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for evaluating the health status of a robot device. The present invention relates to the technical field of equipment health status evaluation, specifically a method and system for evaluating the health status of a robot device. First, the robot device is monitored during operation to obtain a first data set; the operation trajectory is obtained and the first data set is divided to obtain a first data subset; the motor is monitored during operation to obtain a second data set; the second data set is split according to the operation trajectory to obtain a second data subset. Secondly, a working health status recognition model is constructed to identify the first data subset to obtain working health status data; a motor health status recognition model is constructed to identify the second data subset to obtain motor health status data. Finally, the equipment health status data set is obtained through the working health status data and the motor health status data; a status evaluation model is constructed to identify the equipment health status data set to obtain a health status evaluation coefficient, and the robot device health status is evaluated.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method and system for evaluating the health status of a robot device, comprising:

[0008] The fully automatic palletizing robot is monitored during operation to obtain a first data set; the motor is monitored during operation to obtain a second data set;

[0009] Acquire the running trajectory of the fully automatic palletizing robot, divide the first data set and the second data set according to the running trajectory, and obtain a first data subset and a second data subset; the first data subset includes a work running picture subset, an item parameter subset, an equipment running parameter subset, an operation task parameter subset and an environment parameter subset; the second data subset includes a physical data subset and an operation status data subset;

[0010] Constructing a work health status recognition model to recognize the first data subset, and obtaining work health status data through the work operation picture subset, the operation task parameter subset and the environment parameter subset;

[0011] Constructing a motor health status recognition model to recognize the second data subset, and obtaining motor health status data by analyzing the physical data subset and the operating status data subset;

[0012] A state assessment model is constructed to identify the equipment health state data set obtained from the working health state data and the motor health state data to obtain a health state assessment coefficient; the specific calculation formula of the health state assessment coefficient is:

[0013] ;

[0014] in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient;

[0015] The health status of the robot device is evaluated by using the health status evaluation coefficient.

[0016] Preferably, the first data set includes working operation image video, item parameters, equipment operation parameters, operation task parameters and environmental parameters; the second data set includes operation physical data and operation status data;

[0017] The working operation image video includes video pictures of grabbing goods, video pictures of arm lifting and lowering, and video pictures of the main body rotating; the item parameters include item material, item hardness, item size, and item weight; the equipment operation parameters include arm length, joint angle, maximum load weight, and maximum load height; the operation task parameters include operation mode, operation time, operation frequency, maximum operation weight, and maximum operation height; the environmental parameters include ambient temperature, ambient humidity, and working environment flatness;

[0018] The operation physical data includes operation current data and operation voltage data; the operation state data includes operation torque data, operation temperature data and operation vibration data.

[0019] Preferably, the first data subset includes a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset and an environment parameter subset; the second data subset includes a physical data subset and an operation status data subset;

[0020] The operation trajectory of the fully automatic palletizing robot is obtained by analyzing the working operation image video, and the first data set and the second data set are divided according to the operation trajectory to obtain a first data subset and a second data subset;

[0021] The specific process of dividing the first data set is as follows:

[0022] Analyze and identify the motion behavior of the periodic motion of the fully automatic palletizing robot in the working operation image video, and take one periodic motion behavior as an identification unit;

[0023] All recognition units of the full-automatic palletizing robot in the working operation image video are obtained, and the time range corresponding to the recognition unit is obtained; the first data set is divided according to the time range corresponding to the recognition unit to obtain a first data subset.

[0024] Preferably, the working health status identification model includes a first parameter input layer, a first feature extraction layer, an output prediction layer and a health status identification layer;

[0025] The first parameter output layer obtains a first standard data subset by preprocessing a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset, and an environment parameter subset in the first data subset;

[0026] The first feature extraction layer includes an image feature extraction sublayer, an object and environment feature extraction sublayer, and a device and operation task feature extraction sublayer;

[0027] The image feature extraction sublayer obtains the motion features of the robot device by analyzing the standard subset of working operation pictures in the first standard data subset; the object and environment feature extraction sublayer obtains the influence features of the external factors on the device operation by identifying the standard subset of object parameters and the standard subset of environment parameters in the first standard data subset; the equipment and operation task feature extraction sublayer obtains the influence features of the load on the motion state by analyzing the standard subset of equipment operation parameters and the standard subset of operation task parameters in the first standard data subset;

[0028] The output prediction layer generates a prediction output data set by analyzing the equipment motion characteristics, the influence characteristics of external factors on equipment operation, and the influence characteristics of load on motion state; the prediction output data set includes equipment motion health data, the influence data of external factors on equipment operation, and the influence data of load on motion state;

[0029] The health status recognition layer generates working health status data by comprehensively analyzing the equipment motion health data, the impact data of external factors on equipment operation, and the impact data of load on motion status.

[0030] Preferably, the motor health status identification model includes a second parameter input layer, a second feature extraction layer, a physical health status identification layer, an operation health status identification layer and a motor health status identification layer;

[0031] The second parameter output layer obtains a second standard data subset by preprocessing the operating physical data and the operating status data in the second data subset;

[0032] The second feature extraction layer includes a physical data feature extraction sublayer and a state data feature extraction sublayer;

[0033] The physical data feature extraction sublayer obtains the physical features of the motor by analyzing the physical data standard subset in the second standard data subset; the state data feature extraction sublayer obtains the motor running state features by analyzing the running state data standard subset in the second standard data subset;

[0034] The physical health status recognition layer obtains the motor physical health status data by analyzing the motor physical characteristics;

[0035] The running health status identification layer analyzes the motor running status characteristics to obtain the motor running health status data;

[0036] The motor health status identification layer obtains the motor health status data by comprehensively analyzing the motor physical health status data and the motor operation health status data.

[0037] Preferably, the state assessment model includes a state output prediction layer and an equipment health state data identification layer;

[0038] The state output prediction layer obtains health state prediction output data by analyzing the equipment health state data set;

[0039] The equipment health status data identification layer obtains the health status assessment coefficient by analyzing the health status prediction output data.

[0040] Preferably, the specific calculation formula of the health status assessment coefficient is:

[0041] ;

[0042] in, is the motor health data coefficient, is the working health status data coefficient;

[0043] in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the motor physical health status data coefficient weight, is the motor physical health status data coefficient, is the motor running status health status data coefficient weight, is the motor running health status data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient, is the coefficient weight of the equipment sports health data, is the equipment motion health data coefficient, is the coefficient weight of the data on the impact of external factors on equipment operation, is the coefficient of influence of external factors on equipment operation, is the coefficient weight of the load's influence on the motion state, is the coefficient of influence of load on motion state;

[0044] The health status assessment coefficient The value range is ,when The closer it is to 1, the better the health of the automatic palletizing robot.

[0045] A robot equipment health status assessment system, comprising:

[0046] A data acquisition unit is used to monitor the fully automatic palletizing robot during operation to obtain a first data set; the first data set includes working operation image video, item parameters, equipment operation parameters, operation task parameters and environmental parameters; monitor the working motor to obtain a second data set, the second data set includes operation physical data and operation status data;

[0047] A data division unit, configured to obtain an operation trajectory of the fully automatic palletizing robot by analyzing the work operation image video, and divide the first data set according to the operation trajectory to obtain a plurality of data subsets; the first data subset includes a work operation image subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset, and an environment parameter subset; the second data set is split according to the operation trajectory to obtain a plurality of second data subsets; the second data subset includes a physical data subset and an operation status data subset;

[0048] A work health data acquisition unit, used for building a work health status recognition model to recognize the first data subset, and obtaining work health status data through the work operation picture subset, the operation task parameter subset and the environment parameter subset;

[0049] A motor health data acquisition unit, used for constructing a motor health status recognition model to recognize the second data subset, and obtaining motor health status data by analyzing the physical data subset and the operating status data subset;

[0050] The health status evaluation unit is used to construct a status evaluation model to identify the working health status data and the motor health status data to obtain a health status evaluation coefficient; the specific calculation formula of the health status evaluation coefficient is:

[0051] ;

[0052] in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient;

[0053] The health status of the robot device is evaluated by using the health status evaluation coefficient.

[0054] Preferably, the state assessment model includes a state output prediction layer and an equipment health state data identification layer;

[0055] The state output prediction layer obtains health state prediction output data by analyzing the equipment health state data set;

[0056] The equipment health status data identification layer obtains the health status assessment coefficient by analyzing the health status prediction output data.

[0057] Preferably, the specific calculation formula of the health status assessment coefficient is:

[0058] ;

[0059] in, is the motor health data coefficient, is the working health status data coefficient;

[0060] in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the motor physical health status data coefficient weight, is the motor physical health status data coefficient, is the motor running status health status data coefficient weight, is the motor running health status data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient, is the coefficient weight of the equipment sports health data, is the equipment motion health data coefficient, is the coefficient weight of the data on the impact of external factors on equipment operation, is the coefficient of influence of external factors on equipment operation, is the coefficient weight of the load's influence on the motion state, is the coefficient of influence of load on motion state;

[0061] The health status assessment coefficient The value range is ,when The closer it is to 1, the better the health of the automatic palletizing robot.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] 1. The present invention constructs a working health status recognition model to recognize a first data subset, and preprocesses and extracts features through a work operation picture subset, an operation task parameter subset, and an environmental parameter subset in the first data subset to obtain equipment motion features, characteristics of the impact of external factors on equipment operation, and characteristics of the impact of load on motion state; according to the output prediction layer, the equipment motion features, the impact of external factors on equipment operation, and the impact of load on motion state are analyzed to obtain a predicted output data set; according to the health status recognition layer, the predicted output data set is comprehensively analyzed to generate working health status data, thereby accurately evaluating the health status of the fully automatic palletizing robot during operation.

[0064] 2. The present invention constructs a motor health status recognition model to recognize the second data subset, and obtains the motor physical characteristics and the motor operating status characteristics by preprocessing and extracting features of the operating physical data and the operating status data in the second data subset; analyzes the motor physical characteristics and the motor operating status characteristics through the physical health status recognition layer and the operating health status recognition layer to obtain the motor physical health status data and the motor operating status health status data; and then comprehensively analyzes the motor physical health status data and the motor operating status health status data according to the motor health status recognition layer to obtain the motor health status data, thereby accurately evaluating the health status of the motor when the fully-automatic palletizing robot is running.

[0065] 3. The present invention constructs a status assessment model to analyze the equipment health status data set; obtains the equipment health status data set by integrating the working health status data and the motor health status data; predicts and analyzes the equipment health status data set through the status output prediction layer to obtain health status prediction output data; then analyzes the health status prediction output data according to the equipment health status data identification layer to obtain a health status assessment coefficient; evaluates the health status of the robot equipment through the health status assessment coefficient, thereby accurately making a comprehensive assessment of the health status of the fully automatic palletizing robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A schematic diagram of a process of evaluating the health status of a robot device according to the present invention;

[0067] Figure 2 It is a structural schematic diagram of a robot equipment health status assessment system of the present invention;

[0068] Figure 3 It is a structural schematic diagram of the working health status identification model of the present invention;

[0069] Figure 4 It is a structural schematic diagram of the motor health status identification model of the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] The present invention provides a method for evaluating the health status of a robot device. The method is applied to a robot device health status evaluation system. The flowchart of the specific method and system is shown in FIG. Figure 1 and Figure 2 .

[0072] Embodiment 1

[0073] As an embodiment of the present invention, refer to Figure 1 S10 in the figure is applied to a data acquisition unit of a robot equipment health status assessment system, and the data acquisition unit is used to monitor the fully automatic palletizing robot during operation to obtain a first data set; the first data set includes working operation image video, item parameters, equipment operation parameters, operation task parameters and environmental parameters; the working motor is monitored to obtain a second data set, and the second data set includes operation physical data and operation status data.

[0074] Furthermore, the working operation image video includes video pictures of grabbing goods, video pictures of arm lifting and lowering, and video pictures of main body rotation; the item parameters include item material, item hardness, item size, and item weight; the equipment operation parameters include arm length, joint angle, maximum load weight, and maximum load height; the operation task parameters include operation mode, operation time, operation frequency, maximum operation weight, and maximum operation height; the environmental parameters include ambient temperature, ambient humidity, and working environment flatness;

[0075] The operation physical data includes operation current data and operation voltage data; the operation state data includes operation torque data, operation temperature data and operation vibration data.

[0076] In this embodiment, by acquiring physical data such as current, voltage, torque, temperature and vibration, external environmental parameters such as temperature, humidity and flatness, as well as specific operation tasks and item parameters, the evaluation of equipment operation is made more accurate. Through multi-dimensional parameter monitoring and analysis, a comprehensive evaluation of the equipment status of the fully automatic palletizing robot is achieved.

[0077] Reference Figure 1S20 in the data partitioning unit is applied to a robot equipment health status assessment system, and the data partitioning unit is used to obtain the operation trajectory of the full-automatic palletizing robot by analyzing the work operation image video, and divide the first data set according to the operation trajectory to obtain multiple data subsets; the first data subset includes a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset and an environment parameter subset; the second data set is split according to the operation trajectory to obtain multiple second data subsets; the second data subset includes a physical data subset and an operation status data subset.

[0078] Furthermore, the specific process of dividing the first data set is as follows:

[0079] Analyze and identify the motion behavior of the periodic motion of the fully automatic palletizing robot in the working operation image video, and take one periodic motion behavior as an identification unit;

[0080] All recognition units of the full-automatic palletizing robot in the working operation image video are obtained, and the time range corresponding to the recognition unit is obtained; the first data set is divided according to the time range corresponding to the recognition unit to obtain a first data subset.

[0081] The specific process of dividing the second data set is as follows:

[0082] Analyze and identify the motion behavior of the periodic motion of the fully automatic palletizing robot in the working operation image video, and take one periodic motion behavior as an identification unit;

[0083] All recognition units of the full-automatic palletizing robot in the working operation image video are obtained, and the time range corresponding to the recognition unit is obtained; the second data set is divided according to the time range corresponding to the recognition unit to obtain a second data subset.

[0084] In this embodiment, the recognition and segmentation of periodic motion behaviors are achieved through analysis and recognition of working operation images and videos. Each periodic action of the fully automatic palletizing robot is used as an identification unit. The first data set and the second data set are divided into multiple data subsets, and the data are divided according to the time range, thereby improving the accuracy of the robot health status assessment.

[0085] Reference Figure 1 S30 in the figure is applied to a working health data acquisition unit of a robot equipment health status assessment system, and the working health data acquisition unit is used to construct a working health status recognition model to identify the first data subset, and obtain the working health status data through the working operation picture subset, operation task parameter subset and environmental parameter subset.

[0086] Further, the working health status identification model includes a first parameter input layer, a first feature extraction layer, an output prediction layer and a health status identification layer;

[0087] The first parameter output layer obtains a first standard data subset by preprocessing a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset, and an environment parameter subset in the first data subset;

[0088] The first feature extraction layer includes an image feature extraction sublayer, an object and environment feature extraction sublayer, and a device and operation task feature extraction sublayer;

[0089] The image feature extraction sublayer obtains the motion features of the robot device by analyzing the standard subset of the working operation pictures in the first standard data subset; the motion features include the motion trajectory, speed change and shape change under different working tasks; the object and environment feature extraction sublayer obtains the influence features of the external factors on the equipment operation by identifying the standard subset of the object parameters and the standard subset of the environment parameters in the first standard data subset; the equipment and operation task feature extraction sublayer obtains the influence features of the load on the motion state by analyzing the standard subset of the equipment operation parameters and the standard subset of the operation task parameters in the first standard data subset;

[0090] The output prediction layer generates a prediction output data set by analyzing the equipment motion characteristics, the influence characteristics of external factors on equipment operation, and the influence characteristics of load on motion state; the prediction output data set includes equipment motion health data, the influence data of external factors on equipment operation, and the influence data of load on motion state;

[0091] The health status recognition layer generates working health status data by comprehensively analyzing the equipment motion health data, the impact data of external factors on equipment operation, and the impact data of load on motion status.

[0092] In this embodiment, the first data subset is analyzed by constructing a working health status recognition model, and the health status of the robot device is evaluated more comprehensively and accurately by combining preprocessing and feature extraction and analysis. The working health status recognition model includes a first parameter input layer, a feature extraction layer, an output prediction layer, and a health status recognition layer, so that the model can effectively cope with complex robot working environments, thereby improving the accuracy and real-time performance of health status recognition.

[0093] Reference Figure 1S40 in the figure is applied to a motor health data acquisition unit of a robot equipment health status assessment system, and the motor health data acquisition unit is used to construct a motor health status identification model to identify the second data subset, and obtain the motor health status data by analyzing the physical data subset and the operating status data subset.

[0094] Further, the motor health status identification model includes a second parameter input layer, a second feature extraction layer, a physical health status identification layer, an operation health status identification layer and a motor health status identification layer;

[0095] The second parameter output layer obtains a second standard data subset by preprocessing the operating physical data and the operating status data in the second data subset;

[0096] The second feature extraction layer includes a physical data feature extraction sublayer and a state data feature extraction sublayer;

[0097] The physical data feature extraction sublayer obtains the physical features of the motor by analyzing the physical data standard subset in the second standard data subset; the state data feature extraction sublayer obtains the motor running state features by analyzing the running state data standard subset in the second standard data subset;

[0098] The physical health status recognition layer obtains the motor physical health status data by analyzing the motor physical characteristics;

[0099] The running health status identification layer analyzes the motor running status characteristics to obtain the motor running health status data;

[0100] The motor health status identification layer obtains the motor health status data by comprehensively analyzing the motor physical health status data and the motor operation health status data.

[0101] In this embodiment, the second data subset is analyzed by constructing a motor health status recognition model, and a comprehensive analysis and recognition is performed from two dimensions: physical data and operating status; the health status of the motor structure and performance is monitored separately to achieve higher precision detection. In addition, the above data is comprehensively analyzed through the motor health status recognition layer, and finally the motor health status data is generated, which evaluates the health status of the motor as a whole, further improving the accuracy of motor health status monitoring.

[0102] Reference Figure 1S50 in the figure is applied to a health status assessment unit of a robot device health status assessment system, wherein the health status assessment unit is used to obtain a device health status data set by integrating working health status data and motor health status data; construct a status assessment model, identify the device health status data set through the status assessment model, and obtain a health status assessment coefficient; and perform robot device health status assessment through the health status assessment coefficient.

[0103] Further, the state assessment model includes a state output prediction layer and an equipment health state data identification layer;

[0104] The state output prediction layer obtains health state prediction output data by analyzing the equipment health state data set;

[0105] The equipment health status data identification layer obtains the health status assessment coefficient by analyzing the health status prediction output data.

[0106] The specific calculation formula of the health status assessment coefficient is:

[0107] ;

[0108] in, is the motor health data coefficient, is the working health status data coefficient;

[0109] in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the motor physical health status data coefficient weight, is the motor physical health status data coefficient, is the motor running status health status data coefficient weight, is the motor running health status data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient, is the coefficient weight of the equipment sports health data, is the equipment motion health data coefficient, is the coefficient weight of the data on the impact of external factors on equipment operation, is the coefficient of influence of external factors on equipment operation, is the coefficient weight of the load's influence on the motion state, is the coefficient of influence of load on motion state; , and .

[0110] The health status assessment coefficient The value range is ,when The closer it is to 1, the better the health of the automatic palletizing robot. The smaller the value, the more abnormal analysis is needed for different systems to obtain the specific factors affecting the health of the equipment; refer to Table 1 for details;

[0111] Table 1 Health status assessment form

[0112]

[0113] In Table 1, the motion health data coefficient is the equipment motion health data coefficient, the equipment operation data coefficient is the data coefficient of the impact of external factors on equipment operation, the motion state data coefficient is the data coefficient of the impact of load on the motion state, the motor physical health data coefficient is the motor physical health state data coefficient, and the health state assessment coefficient is the motor operation state health state data coefficient.

[0114] The specific calculation formula of the equipment sports health data coefficient is:

[0115] ;

[0116] in, is the weight of the running trajectory, is the absolute value of the deviation between the actual trajectory and the standard running trajectory, is the standard running track. is the weight of the running speed, is the absolute value of the deviation between the actual speed and the standard operating speed, is the standard operating speed; .

[0117] The specific calculation formula of the influence coefficient of the external factors on the equipment operation is:

[0118] ;

[0119] in, is the weight of the ambient temperature, is the actual ambient temperature, is the standard ambient temperature, is the weight of ambient humidity, is the actual ambient humidity, is the standard ambient humidity, is the weight of the operating environment flatness, is the absolute value of the deviation between the actual environment flatness and the standard environment flatness, It is the standard environment flatness; .

[0120] The specific calculation formula of the load's influence coefficient on the motion state is:

[0121] ;

[0122] in, is the weight of the load on the motion state, is the actual load weight, is the standard load weight, is the weight of the influence of load height on the motion state, is the actual load height, is the standard load height; .

[0123] The specific calculation formula of the motor physical health status data coefficient is:

[0124] ;

[0125] in, is the motor current health status weight, is the actual current, is the standard current under normal health conditions, is the motor voltage health status weight, is the actual voltage, It is the standard voltage under normal healthy condition; .

[0126] The specific calculation formula of the motor running state health state data coefficient is:

[0127] ;

[0128] in, is the weight of the impact of the torque on the health status when the motor is running, is the actual torque, It is the standard torque under normal healthy condition. is the weight of the impact of motor running temperature on the health status, is the actual temperature, It is the standard temperature under normal health conditions. is the weight of the impact of motor vibration on the health status during operation, is the motor vibration data; .

[0129] In this embodiment, the weights are obtained by training the machine learning model based on the historical data of cargo handling by the palletizing robot. The health status assessment unit comprehensively analyzes the working health status data and the motor health status data to generate a device health status data set, and constructs a status assessment model to assess the overall health status of the equipment. At the same time, by introducing the calculation formula of the health status assessment coefficient, targeted weight allocation is performed on data such as the physical and operating status of the motor, the motion characteristics of the equipment, and the impact of the external environment, thereby improving the scientific nature of the health status assessment.

[0130] In this embodiment, a health status assessment model is constructed by comprehensively collecting and analyzing the working status data, motor health status data and external environmental factors of the fully automatic palletizing robot equipment, so as to achieve accurate assessment of the health status of the equipment from multiple dimensions. The weights of various parameters are obtained by training historical data using machine learning, so that the assessment model can be dynamically adjusted according to different influencing factors. Through the calculation formula of the health status assessment coefficient, the weights of the equipment motion characteristics, motor operating parameters, load conditions and environmental impacts are scientifically allocated, so as to more accurately judge the health status of the equipment and effectively improve the reliability of equipment operation and maintenance efficiency.

[0131] Embodiment 2

[0132] The fully automatic palletizing robot is a highly intelligent industrial equipment specially designed for fast and accurate handling and stacking of goods. It is widely used in warehousing, logistics and manufacturing. Its main functions include cargo grabbing, handling, stacking and unloading, usually through a robotic arm, multi-joint structure and multi-axis motion to achieve efficient operation. As the equipment is in a state of high-intensity operation for a long time, it is particularly important to accurately evaluate its health status to ensure stable working performance and avoid unexpected downtime.

[0133] To achieve this goal, this system monitors and evaluates the health status of the fully automatic palletizing robot through multiple modules. Figure 1 S10 in the data acquisition unit of a robot equipment health status assessment system, see Figure 2 The data acquisition unit is used to monitor the fully automatic palletizing robot during operation to obtain a first data set; the first data set includes working operation image video, item parameters, equipment operation parameters, operation task parameters and environmental parameters; the working motor is monitored to obtain a second data set, and the second data set includes operation physical data and operation status data.

[0134] Furthermore, the working operation image video includes video pictures of grabbing goods, video pictures of arm lifting and lowering, and video pictures of main body rotation; the item parameters include item material, item hardness, item size and item weight; the equipment operation parameters include arm length, joint angle, maximum load weight and maximum load height; the operation task parameters include operation mode, operation time, operation frequency, maximum operation weight and maximum operation height; the environmental parameters include ambient temperature, ambient humidity and flatness of the working environment; the operation physical data includes operation current data and operation voltage data; the operation status data includes operation torque data, operation temperature data and operation vibration data.

[0135] In the working operation video, the image monitoring of the fully automatic palletizing robot grabbing goods, arm lifting, and body rotation can be used to identify the health status of the palletizing operation trajectory during the working process, determine the trajectory, posture, and accuracy of the robot in each working step, and abnormal motion trajectory may indicate potential problems in the motion system. The material, hardness, size, and weight of the object will directly affect the load state of the robot during the palletizing process. At the same time, the change in weight will affect the motor load and the torque of the joint. The material and hardness will affect the grasping method and strength of the robot hand, while the arm length and joint angle will affect the range of motion of the robot arm. The maximum load weight and the highest load height determine the stability of the robot when subjected to extreme load conditions. These parameters help to determine whether the robot equipment is overloaded during the task. Frequent operations, high loads, and high-frequency operations will cause equipment wear and heat accumulation, thereby affecting the health status of the robot equipment. Environmental factors such as temperature, humidity, and ground flatness will not only affect the operating stability of the equipment and the life of electrical components, but may also cause motor temperature rise and electrical system instability. The current and voltage data reflect the instantaneous load status of the motor. When the current or voltage fluctuates greatly, it may mean that the equipment load is abnormal or the motor is overloaded. Long-term excessive current may cause excessive temperature rise in the motor and shorten the life of the equipment.

[0136] Reference Figure 1 S20 in the data partitioning unit of a robot equipment health status assessment system, see Figure 2 The data division unit is used to obtain the operation trajectory of the full-automatic palletizing robot by analyzing the work operation image video, and divide the first data set according to the operation trajectory to obtain multiple data subsets; the first data subset includes a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset and an environment parameter subset; the second data set is split according to the operation trajectory to obtain multiple second data subsets; the second data subset includes a physical data subset and an operation status data subset.

[0137] Furthermore, the specific process of dividing the first data set is as follows:

[0138] Analyze and identify the motion behavior of the periodic motion of the fully automatic palletizing robot in the working operation image video, and take one periodic motion behavior as an identification unit;

[0139] All recognition units of the full-automatic palletizing robot in the working operation image video are obtained, and the time range corresponding to the recognition unit is obtained; the first data set is divided according to the time range corresponding to the recognition unit to obtain a first data subset.

[0140] The specific process of dividing the second data set is as follows:

[0141] Analyze and identify the motion behavior of the periodic motion of the fully automatic palletizing robot in the working operation image video, and take one periodic motion behavior as an identification unit;

[0142] All recognition units of the full-automatic palletizing robot in the working operation image video are obtained, and the time range corresponding to the recognition unit is obtained; the second data set is divided according to the time range corresponding to the recognition unit to obtain a second data subset.

[0143] Reference Figure 1 S30 in which S30 is applied to a work health data acquisition unit of a robot equipment health status assessment system, see Figure 2 The work health data acquisition unit is used to construct a work health status recognition model to identify the first data subset, and obtain the work health status data through the work operation picture subset, operation task parameter subset and environment parameter subset.

[0144] Furthermore, the working health status recognition model includes a first parameter input layer, a first feature extraction layer, an output prediction layer and a health status recognition layer, see Figure 3 ;

[0145] The first parameter output layer obtains a first standard data subset by preprocessing a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset, and an environment parameter subset in the first data subset;

[0146] The first feature extraction layer includes an image feature extraction sublayer, an object and environment feature extraction sublayer, and a device and operation task feature extraction sublayer;

[0147] The image feature extraction sublayer obtains the motion features of the robot device by analyzing the standard subset of the working operation pictures in the first standard data subset; the motion features include the motion trajectory, speed change and shape change under different working tasks; the object and environment feature extraction sublayer obtains the influence features of the external factors on the equipment operation by identifying the standard subset of the object parameters and the standard subset of the environment parameters in the first standard data subset; the equipment and operation task feature extraction sublayer obtains the influence features of the load on the motion state by analyzing the standard subset of the equipment operation parameters and the standard subset of the operation task parameters in the first standard data subset;

[0148] The output prediction layer generates a prediction output data set by analyzing the equipment motion characteristics, the influence characteristics of external factors on equipment operation, and the influence characteristics of load on motion state; the prediction output data set includes equipment motion health data, the influence data of external factors on equipment operation, and the influence data of load on motion state;

[0149] The health status recognition layer generates working health status data by comprehensively analyzing the equipment motion health data, the impact data of external factors on equipment operation, and the impact data of load on motion status.

[0150] Reference Figure 1 S40 in the embodiment of the present invention is applied to a motor health data acquisition unit of a robot equipment health status assessment system, see Figure 2 The motor health data acquisition unit is used to construct a motor health status identification model to identify the second data subset, and obtain the motor health status data by analyzing the physical data subset and the operating status data subset.

[0151] Furthermore, the motor health state recognition model includes a second parameter input layer, a second feature extraction layer, a physical health state recognition layer, an operation health state recognition layer and a motor health state recognition layer, see Figure 4 ;

[0152] The second parameter output layer obtains a second standard data subset by preprocessing the operating physical data and the operating status data in the second data subset;

[0153] The second feature extraction layer includes a physical data feature extraction sublayer and a state data feature extraction sublayer;

[0154] The physical data feature extraction sublayer obtains the physical features of the motor by analyzing the physical data standard subset in the second standard data subset; the state data feature extraction sublayer obtains the motor running state features by analyzing the running state data standard subset in the second standard data subset;

[0155] The physical health status recognition layer obtains the motor physical health status data by analyzing the motor physical characteristics;

[0156] The running health status identification layer analyzes the motor running status characteristics to obtain the motor running health status data;

[0157] The motor health status identification layer obtains the motor health status data by comprehensively analyzing the motor physical health status data and the motor operation health status data.

[0158] Reference Figure 1 S50 in the figure is applied to a health status assessment unit of a robot device health status assessment system, wherein the health status assessment unit is used to obtain a device health status data set by integrating working health status data and motor health status data; construct a status assessment model, identify the device health status data set through the status assessment model, and obtain a health status assessment coefficient; and perform robot device health status assessment through the health status assessment coefficient.

[0159] Further, the state assessment model includes a state output prediction layer and an equipment health state data identification layer;

[0160] The state output prediction layer obtains health state prediction output data by analyzing the equipment health state data set;

[0161] The equipment health status data recognition layer obtains the health status evaluation coefficient by analyzing the health status prediction output data. The specific calculation formula of the health status evaluation coefficient is:

[0162] ;

[0163] in, is the motor health data coefficient, is the working health status data coefficient;

[0164] in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the motor physical health status data coefficient weight, is the motor physical health status data coefficient, is the motor running status health status data coefficient weight, is the motor running health status data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient, is the coefficient weight of the equipment sports health data, is the equipment motion health data coefficient, is the coefficient weight of the data on the impact of external factors on equipment operation, is the coefficient of influence of external factors on equipment operation, is the coefficient weight of the load's influence on the motion state, is the coefficient of influence of load on motion state; , and .

[0165] The health status assessment coefficient The value range is ,when The closer it is to 1, the better the health of the automatic palletizing robot.

[0166] The present invention makes an accurate assessment of the health status of the robot equipment by comprehensively considering the health status of the fully automatic palletizing robot during operation and the health status of the motor; the present invention constructs Model 1, Model 2, Model 3 and Model 4, and through comparative analysis, it shows the accuracy of the health status assessment of the present invention;

[0167] Among them, Model 1 is the coefficient obtained by the health status evaluation of the present invention, and the health status is evaluated by considering the working operation image video, item parameters, equipment operation parameters, operation task parameters, environmental parameters, operation physical data and operation status data of the full-automatic palletizing robot; Model 2 is based on Model 1, without considering the operation trajectory of the full-automatic palletizing robot during operation, and the health status is evaluated; Model 3 is based on Model 1, without considering the load weight of the full-automatic palletizing robot during operation, and the health status is evaluated; Model 4 is based on Model 1, without considering the motor vibration data of the full-automatic palletizing robot during operation, and the health status is evaluated.

[0168] The same data set was used to verify Model 1, Model 2, Model 3 and Model 4, and the health status assessment coefficients were obtained.

[0169] The data of each model data dimension combination and health status assessment coefficient finally obtained are shown in Table 2.

[0170] Table 2 Health status assessment form

[0171]

[0172] From the data in Table 2, it can be seen that the health status assessment accuracy of model 1 is the highest.

[0173] The health status assessment system for a fully automatic palletizing robot proposed in the present invention first obtains multi-dimensional data in the operation process of the robot equipment in real time through a data acquisition unit, including equipment operating parameters, environmental factors, task operation parameters, motor physical data, etc., to ensure the comprehensiveness and real-time nature of the assessment data. Secondly, by intelligently analyzing the collected data, efficient monitoring of the motion state of the equipment and the health state of the motor is achieved, so that the system can accurately identify and feedback the health state of the robot. The state assessment model and health status assessment coefficient are used again to optimize the weight distribution of multi-dimensional feature data, making the calculation and assessment of the health status more scientific and reasonable. Finally, the weight model is trained through machine learning to improve the adaptability of the system in complex industrial environments, providing a scientific basis for achieving long-term, efficient and stable operation of robot equipment.

[0174] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the health status of a robot device, characterized in that: include: Monitor the fully automatic palletizing robot while it is working to obtain a first data set; Monitoring the working motor to obtain a second data set; Acquire the running trajectory of the fully automatic palletizing robot, divide the first data set and the second data set according to the running trajectory, and obtain a first data subset and a second data subset; the first data subset includes a work running picture subset, an item parameter subset, an equipment running parameter subset, an operation task parameter subset and an environment parameter subset; the second data subset includes a physical data subset and an operation status data subset; Constructing a work health status recognition model to recognize the first data subset, and obtaining work health status data through the work operation picture subset, the operation task parameter subset and the environment parameter subset; The working health status recognition model includes a first parameter input layer, a first feature extraction layer, an output prediction layer and a health status recognition layer; the first parameter output layer obtains a first standard data subset by preprocessing a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset and an environment parameter subset in a first data subset; the first feature extraction layer includes an image feature extraction sublayer, an item and environment feature extraction sublayer and an equipment and operation task feature extraction sublayer; the image feature extraction sublayer obtains the robot equipment motion characteristics by analyzing the standard subset of work operation pictures in the first standard data subset; the item and environment feature extraction sublayer obtains the robot equipment motion characteristics by identifying the standard subset of item parameters and the standard subset of environment parameters in the first standard data subset. The equipment and operation task feature extraction sublayer obtains the characteristics of the influence of external factors on the equipment operation by analyzing the equipment operation parameter standard subset and the operation task parameter standard subset in the first standard data subset; the output prediction layer generates a predicted output data set by analyzing the equipment motion characteristics, the characteristics of the influence of external factors on the equipment operation and the characteristics of the influence of load on the motion state; the predicted output data set includes equipment motion health data, the influence of external factors on the equipment operation and the influence of load on the motion state; the health status recognition layer generates working health status data by comprehensively analyzing the equipment motion health data, the influence of external factors on the equipment operation and the influence of load on the motion state; Constructing a motor health status recognition model to recognize the second data subset, and obtaining motor health status data by analyzing the physical data subset and the operating status data subset; The motor health status identification model includes a second parameter input layer, a second feature extraction layer, a physical health status identification layer, an operation health status identification layer and a motor health status identification layer; the second parameter output layer obtains a second standard data subset by preprocessing the operation physical data and the operation status data in the second data subset; the second feature extraction layer includes a physical data feature extraction sublayer and a state data feature extraction sublayer; the physical data feature extraction sublayer obtains the motor physical characteristics by analyzing the physical data standard subset in the second standard data subset; the state data feature extraction sublayer obtains the motor operation status characteristics by analyzing the operation status data standard subset in the second standard data subset; the physical health status identification layer obtains the motor physical health status data by analyzing the motor physical characteristics; the operation health status identification layer obtains the motor operation status health status data by analyzing the motor operation status characteristics; the motor health status identification layer obtains the motor health status data by comprehensively analyzing the motor physical health status data and the motor operation status health status data; Constructing a status assessment model to identify the equipment health status data set obtained from the working health status data and the motor health status data, and obtaining a health status assessment coefficient; The state assessment model includes a state output prediction layer and an equipment health state data identification layer; the state output prediction layer obtains health state prediction output data by analyzing the equipment health state data set; The equipment health status data identification layer obtains a health status assessment coefficient by analyzing the health status prediction output data; The specific calculation formula of the health status assessment coefficient is: in, is the motor health data coefficient, is the working health status data coefficient; in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the motor physical health status data coefficient weight, is the motor physical health status data coefficient, is the motor running status health status data coefficient weight, is the motor running health status data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient, is the coefficient weight of the equipment sports health data, is the equipment motion health data coefficient, is the coefficient weight of the data on the impact of external factors on equipment operation, is the coefficient of influence of external factors on equipment operation, is the data coefficient weight of the load's influence on the motion state, is the coefficient of influence of load on motion state; The health status assessment coefficient The value range is when The closer it is to 1, the better the health of the automatic palletizing robot; The health status of the robot device is evaluated by using the health status evaluation coefficient.

2. A method for evaluating the health status of a robot device according to claim 1, characterized in that: The first data set includes work operation image video, item parameters, equipment operation parameters, operation task parameters and environmental parameters; the second data set includes operation physical data and operation status data; The working operation image video includes video pictures of grabbing goods, video pictures of arm lifting and lowering, and video pictures of the main body rotating; the item parameters include item material, item hardness, item size, and item weight; the equipment operation parameters include arm length, joint angle, maximum load weight, and maximum load height; the operation task parameters include operation mode, operation time, operation frequency, maximum operation weight, and maximum operation height; the environmental parameters include ambient temperature, ambient humidity, and working environment flatness; The operation physical data includes operation current data and operation voltage data; the operation state data includes operation torque data, operation temperature data and operation vibration data.

3. A method for evaluating the health status of a robot device according to claim 1, characterized in that: The first data subset includes a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset and an environment parameter subset; the second data subset includes a physical data subset and an operation status data subset; The operation trajectory of the fully automatic palletizing robot is obtained by analyzing the working operation image video, and the first data set and the second data set are divided according to the operation trajectory to obtain a first data subset and a second data subset; The specific process of dividing the first data set is as follows: Analyze and identify the motion behavior of the periodic motion of the fully automatic palletizing robot in the working operation image video, and take one periodic motion behavior as an identification unit; All recognition units of the full-automatic palletizing robot in the working operation image video are obtained, and the time range corresponding to the recognition unit is obtained; the first data set is divided according to the time range corresponding to the recognition unit to obtain a first data subset.

4. A method for evaluating the health status of a robot device according to claim 1, characterized in that: The specific calculation formula of the health status assessment coefficient is: in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient.

5. A robot equipment health status assessment system, characterized in that: include: A data acquisition unit is used to monitor the fully automatic palletizing robot during operation to obtain a first data set; the first data set includes working operation image video, item parameters, equipment operation parameters, operation task parameters and environmental parameters; monitor the working motor to obtain a second data set, the second data set includes operation physical data and operation status data; a data division unit, configured to obtain an operation trajectory of the fully automatic palletizing robot by analyzing the work operation image video, and to divide the first data set according to the operation trajectory to obtain a plurality of data subsets; the first data subset includes a work operation image subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset, and an environment parameter subset; The second data set is split according to the operation trajectory to obtain a plurality of second data subsets; the second data subsets include a physical data subset and an operation status data subset; A work health data acquisition unit, used for building a work health status recognition model to recognize the first data subset, and obtaining work health status data through the work operation picture subset, the operation task parameter subset and the environment parameter subset; The working health status recognition model includes a first parameter input layer, a first feature extraction layer, an output prediction layer and a health status recognition layer; the first parameter output layer obtains a first standard data subset by preprocessing a work operation picture subset, an item parameter subset, an equipment operation parameter subset, an operation task parameter subset and an environment parameter subset in a first data subset; the first feature extraction layer includes an image feature extraction sublayer, an item and environment feature extraction sublayer and an equipment and operation task feature extraction sublayer; the image feature extraction sublayer obtains the robot equipment motion characteristics by analyzing the standard subset of working operation pictures in the first standard data subset; the item and environment feature extraction sublayer obtains the robot equipment motion characteristics by preprocessing the standard subset of item parameters and the standard subset of environment parameters in the first standard data subset Identify and obtain the impact characteristics of external factors on equipment operation; the equipment and operation task feature extraction sublayer obtains the impact characteristics of load on motion state by analyzing the equipment operation parameter standard subset and the operation task parameter standard subset in the first standard data subset; the output prediction layer generates a predicted output data set by analyzing the equipment motion characteristics, the impact characteristics of external factors on equipment operation and the impact characteristics of load on motion state; the predicted output data set includes equipment motion health data, the impact data of external factors on equipment operation and the impact data of load on motion state; the health state recognition layer generates working health state data by comprehensively analyzing the equipment motion health data, the impact data of external factors on equipment operation and the impact data of load on motion state; A motor health data acquisition unit, used for constructing a motor health status recognition model to recognize the second data subset, and obtaining motor health status data by analyzing the physical data subset and the operating status data subset; The motor health status identification model includes a second parameter input layer, a second feature extraction layer, a physical health status identification layer, an operation health status identification layer and a motor health status identification layer; the second parameter output layer obtains a second standard data subset by preprocessing the operation physical data and the operation status data in the second data subset; the second feature extraction layer includes a physical data feature extraction sublayer and a state data feature extraction sublayer; the physical data feature extraction sublayer obtains the motor physical characteristics by analyzing the physical data standard subset in the second standard data subset; the state data feature extraction sublayer obtains the motor operation status characteristics by analyzing the operation status data standard subset in the second standard data subset; the physical health status identification layer obtains the motor physical health status data by analyzing the motor physical characteristics; the operation health status identification layer obtains the motor operation status health status data by analyzing the motor operation status characteristics; the motor health status identification layer obtains the motor health status data by comprehensively analyzing the motor physical health status data and the motor operation status health status data; A health status assessment unit, used for constructing a status assessment model to identify the working health status data and the motor health status data, and obtain a health status assessment coefficient; The state assessment model includes a state output prediction layer and an equipment health state data identification layer; the state output prediction layer obtains health state prediction output data by analyzing the equipment health state data set; The equipment health status data identification layer obtains a health status assessment coefficient by analyzing the health status prediction output data; The specific calculation formula of the health status assessment coefficient is: in, is the motor health data coefficient, is the working health status data coefficient; in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the motor physical health status data coefficient weight, is the motor physical health status data coefficient, is the motor running status health status data coefficient weight, is the motor running health status data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient, is the coefficient weight of the equipment sports health data, is the equipment sports health data coefficient, is the coefficient weight of the data on the impact of external factors on equipment operation, is the coefficient of influence of external factors on equipment operation, is the coefficient weight of the load's influence on the motion state, is the coefficient of influence of load on motion state; The health status assessment coefficient The value range is when The closer it is to 1, the better the health of the automatic palletizing robot; The health status of the robot device is evaluated by using the health status evaluation coefficient.

6. A robot equipment health status assessment system according to claim 5, characterized in that: The specific calculation formula of the health status assessment coefficient is: in, is the health status assessment coefficient, is the motor health status data coefficient weight, is the motor health data coefficient, is the coefficient weight of the working health status data, is the working health status data coefficient.

Images