An industrial robot trajectory intelligent detection system

Abstract

The application discloses an industrial robot track intelligent detection system, and relates to the field of intelligent detection.The system comprises a collection array layout module, a multi-source data collection module, a pose track analysis module, a precision calibration analysis module and a precision compensation analysis module.The collection array layout module is used for obtaining industrial robot information to arrange sensors.The multi-source data collection module is used for collecting corresponding control monitoring information and target monitoring information according to the sensor arrangement result and binocular vision technology.The pose track analysis module is used for analyzing and processing the obtained monitoring information to construct a kinematic model.The precision calibration analysis module is used for identifying and optimizing the structure parameters in the constructed kinematic model.The precision compensation analysis module is used for setting corresponding sampling points according to the load state of the end of the industrial robot, and precisely compensating the corresponding kinematic model according to the sampling points.The application improves the accuracy of track detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent detection, and in particular to an industrial robot trajectory intelligent detection system. BACKGROUND

[0002] In modern industrial production, with the advent of the era of Industry 4.0 and the development of intelligent manufacturing, more and more emphasis is placed on digitization, intelligentization and automation in the production process, and industrial robots are widely used in many fields. With the development of manufacturing towards high-end and intelligentization, the requirement for the operation precision of industrial robots is increasingly improved, and intelligent production requires industrial robots to have real-time monitoring and self-adaptive adjustment capabilities, and trajectory intelligent detection can provide accurate information to realize intelligent operation. Therefore, intelligent detection of the track position of an industrial robot during operation is a problem that needs to be solved.

[0003] After searching, the invention patent with Chinese patent number CN118977276A discloses a robot motion trajectory monitoring method based on machine vision, which relates to the field of data monitoring and solves the technical problem of robot motion trajectory monitoring. The robot motion trajectory data information can be obtained, and the obtained data information is transmitted through a narrowband Internet of Things (NB-IoT) module. Multiple cameras are set up to cross-collect robot motion images 3D and 2D optical markers through mutual projection, and the imaging parallax is calculated to quickly realize the robot motion image acquisition process. The NB-IoT module obtains the angular displacement, angular velocity, angular acceleration, force and torque data information of the robot joint through the serial port. If the data is abnormal, the local directly alarms, and the robot motion image is identified through binocular stereo machine vision technology. Through the three-dimensional coordinates of the robot optical marker in the field of view of multiple cameras, the edge monitoring of the motion trajectory is realized. The present application can greatly improve the robot motion trajectory monitoring capability.

[0004] Compared with the prior art, the invention patent with Chinese patent number CN118977276A can mark the robot motion images obtained by multiple cameras through the narrowband Internet of Things module, obtain the corresponding sampling robot running image information, and detect the details of the motion trajectory feature points, thereby improving the monitoring capability of the robot motion trajectory.

[0005] But the above method in the actual use process, through the corresponding industrial robot motion trajectory monitoring processing, does not consider the corresponding industrial robot in the execution corresponding industrial operation process the trajectory movement accuracy of the corresponding execution end, and the corresponding execution end in the industrial production process is involved in the load state, resistance state, connection state and other state variety, therefore, how to improve the corresponding industrial robot in the process of corresponding motion trajectory detection accuracy in different states is the problem we need to solve, for this purpose, the present application provides an industrial robot trajectory intelligent detection system. SUMMARY

[0006] The purpose of the present application is to solve the problem of lack of accuracy in the prior art, and propose an industrial robot trajectory intelligent detection system.

[0007] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0008] An industrial robot trajectory intelligent detection system, comprising:

[0009] The acquisition array layout module is used to obtain the corresponding industrial robot software information and industrial robot hardware information, to set up the sensor array layout information according to the obtained information.

[0010] The multi-source data acquisition module is used to collect the corresponding control monitoring information and the target monitoring information corresponding to each target monitoring point of the corresponding industrial robot according to the sensor array layout information and the binocular vision technology, and to generate a multi-source heterogeneous data set from the obtained monitoring information.

[0011] The pose trajectory analysis module is used to analyze and process the control monitoring information and the target monitoring information in the multi-source heterogeneous data set, to obtain the corresponding instruction data and pose data, to construct a kinematic model according to the instruction data and the pose data, to fit and compare the kinematic model, and to optimize the fitting and comparison results.

[0012] The precision calibration analysis module is used to establish a linear-nonlinear parameter separation robot MD-H model to identify the structure parameters of the corresponding kinematic model of the corresponding industrial robot, to obtain the optimal value of the pose parameters corresponding to the corresponding structure and the execution end in the industrial robot.

[0013] The precision compensation analysis module is used to set the corresponding sampling points according to the optimal value of the pose parameters at the corresponding position of the industrial robot, to establish the corresponding candidate pose pool of the corresponding industrial robot for the corresponding sampling points, to select the corresponding sampling points, to perform parameter identification and precision compensation processing on the kinematic model analysis processing of the corresponding execution end according to the selected sampling points, and to obtain the trajectory intelligent detection data of the corresponding industrial robot.

[0014] The technical solution further comprises that the array layout collection module comprises:

[0015] a hardware collection unit configured to collect corresponding industrial robot hardware information, the industrial robot hardware information comprising robot component information and robot mechanical structure information;

[0016] a software collection unit configured to collect corresponding industrial robot software information, the industrial robot software information comprising robot power drive information and robot communication control information;

[0017] a layout management unit configured to plan the erection and arrangement of the corresponding flexible composite sensor of the industrial robot according to the industrial robot hardware information and the industrial robot software information, and set sensor array layout information.

[0018] Further, the multi-source data collection module comprises:

[0019] a sensor collection unit configured to collect control monitoring information corresponding to the industrial robot according to the sensor array layout information, the control monitoring information comprising external mechanical control monitoring information, internal electrical control monitoring information, and drive transmission control monitoring information;

[0020] a vision collection unit configured to monitor and process a plurality of target monitoring points in an execution end corresponding to the industrial robot based on binocular vision technology, and obtain target monitoring information corresponding to each target monitoring point;

[0021] a collection and integration unit configured to integrate the obtained control monitoring information and target monitoring information to generate a multi-source heterogeneous data set.

[0022] Further, the pose trajectory analysis module comprises:

[0023] obtaining a multi-source heterogeneous data set corresponding to the industrial robot;

[0024] a pose analysis unit configured to perform imaging processing on the target monitoring information obtained by each target monitoring point corresponding to the corresponding execution end in the industrial robot, establish a three-dimensional space image corresponding to the corresponding execution end, sequentially perform feature extraction on the target monitoring data obtained by the corresponding target monitoring points in the three-dimensional space image according to the order of points, lines, and surfaces, establish an end attitude plane corresponding to the industrial robot, and obtain corresponding pose data according to the end attitude plane;

[0025] The instruction analysis unit is configured to perform feature extraction on the electrical control monitoring information and the drive transmission control monitoring information in the control monitoring information corresponding to the industrial robot, obtain control monitoring information associated with an execution end, and extract instruction data associated with the execution end of the industrial robot.

[0026] Further, the pose trajectory analysis module comprises:

[0027] The comparative fitting unit is configured to set actual observation values according to the pose data and the instruction data corresponding to the corresponding industrial robot, set a historical fitting data set with the obtained pose data and instruction data, analyze and process the historical fitting data set, and establish a kinematic model corresponding to the industrial robot.

[0028] The corresponding pose data and instruction data are input into the corresponding kinematic model for analysis and processing, a predicted model value is obtained, the obtained predicted model value is compared and analyzed with the actual observation value based on the least square method, the kinematic model is optimized according to the comparison and analysis result, and the kinematic model after optimization is output.

[0029] Further, the precision calibration analysis module comprises:

[0030] The running classification unit is configured to classify the instruction data and the pose data corresponding to the kinematic model, obtain a load state of the pose data corresponding to the instruction data, the load state comprising an empty load state and a load state, and mark the corresponding kinematic model according to the corresponding load state.

[0031] The precision calibration unit is configured to analyze and process the corresponding mark processing result at the corresponding position in the kinematic model, obtain a structure parameter of the corresponding kinematic model in the corresponding load state, and identify the corresponding load state.

[0032] The corresponding load state is identified based on the robot MD-H model, linear-nonlinear parameter separation is performed on the corresponding to-be-identified parameter in the corresponding kinematic model, linear parameters, nonlinear parameters, and a structure parameter identification original problem are obtained.

[0033] The structure parameter identification original problem is converted into a separable nonlinear least square problem, the obtained nonlinear least square problem, linear parameters, and nonlinear parameters are solved, and an optimal value of a pose parameter corresponding to the corresponding to-be-identified parameter is obtained.

[0034] Further, the precision compensation analysis module comprises:

[0035] The sampling management unit is used for acquiring instruction data corresponding to the industrial robot, classifying the corresponding instruction data according to the corresponding work space and operation range, acquiring pose data corresponding to the corresponding instruction data in the work space and operation range, and setting the obtained pose data as a candidate pose pool in the corresponding work space and operation range.

[0036] The initial poses of the industrial robot corresponding to different load states in the corresponding candidate pose pool are acquired based on a random sampling method, and the observability index and the distance index are set according to the corresponding load states.

[0037] The corresponding sampling points in the work space and operation range of the industrial robot are respectively comprehensively processed according to the observability index and the distance index, the comprehensive processing results are sorted, and the corresponding sampling point set is acquired according to the sorting results.

[0038] Further, the precision compensation analysis module comprises:

[0039] The precision compensation unit is used for acquiring the sampling point set obtained by the corresponding pose data in the work space and operation range of the industrial robot, analyzing and processing each sampling point in the sampling point set based on the corresponding kinematic model, acquiring the corresponding predicted pose data, and acquiring the actual pose data corresponding to each sampling point.

[0040] The predicted pose data and the actual pose data corresponding to each sampling point are compared and analyzed to acquire error data, the error data corresponding to each sampling point is statistically analyzed to acquire error rule data corresponding to each sampling point, and precision compensation processing is performed according to the error rule data.

[0041] The precision compensation processing results of the sampling points corresponding to the analysis results of the kinematic model to which the corresponding pose data in the corresponding work space and operation range belong are comprehensively marked.

[0042] The corresponding instruction data and pose data are acquired, the obtained data are input into the kinematic model corresponding to the corresponding industrial robot, online monitoring results are acquired, precision compensation processing is performed according to the corresponding sampling points, and trajectory intelligent detection data corresponding to the corresponding industrial robot are acquired.

[0043] The present application has the following advantages:

[0044] 1、 The application discloses an industrial robot trajectory intelligent detection system and a method thereof.

[0045] 2、 The application discloses an industrial robot trajectory intelligent detection system and a method thereof. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 FIG. 1 is a structural schematic diagram of the industrial robot trajectory intelligent detection system. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments of the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0048] Embodiment one

[0049] As shown in FIG. 1, the industrial robot trajectory intelligent detection system comprises a collection array layout module, a multi-source data collection module, a pose trajectory analysis module, a precision calibration analysis module and a precision compensation analysis module. Figure 1 The collection array layout module is used for acquiring industrial robot software information and industrial robot hardware information, arranging sensors according to the acquired information, and setting sensor array layout information.

[0050] The multi-source data collection module is used for collecting control monitoring information of the industrial robot and target monitoring information of each target monitoring point according to the sensor array layout information and binocular vision technology, and generating a multi-source heterogeneous data set from the acquired monitoring information.

[0051]

[0052] ​The pose trajectory analysis module is configured to analyze and process the control monitoring information and the target monitoring information in the multi-source heterogeneous data set, obtain corresponding instruction data and pose data, construct a kinematic model according to the instruction data and the pose data, perform fitting comparison on the kinematic model, and perform optimization processing according to the fitting comparison result.

[0053] The precision calibration analysis module is configured to identify the structure parameters of the kinematic model corresponding to the industrial robot by using the robot MD-H model with linear-nonlinear parameter separation, and obtain the optimal value of the pose parameters corresponding to the corresponding structure and the execution end in the industrial robot.

[0054] The precision compensation analysis module is configured to set corresponding sampling points according to the optimal value of the pose parameters at the corresponding position of the industrial robot, establish a candidate pose pool corresponding to the corresponding industrial robot for the corresponding sampling points, select the corresponding sampling points, perform parameter identification and precision compensation processing on the kinematic model analysis processing corresponding to the corresponding execution end according to the selected sampling points, and obtain the trajectory intelligent detection data corresponding to the corresponding industrial robot.

[0055] It should be further explained that, in the specific implementation process, the internal electrical control system, the external mechanical control system and the drive transmission control system corresponding to the industrial robot are subjected to multi-source performance data collection through the sensor array layout information, thereby improving the reliability and reliability of the corresponding trajectory detection process of the industrial robot. In addition, the three optimization processing processes of the set kinematic model based on instruction data and pose data optimization processing, the corresponding mechanical structure and the corresponding execution model optimization processing, and the setting of the sampling points under different load states for compensation optimization processing are analyzed and processed, thereby greatly improving the accuracy of the kinematic model corresponding to the corresponding industrial robot in the trajectory detection analysis process.

[0056] In the specific implementation process, the collection array layout module comprises:

[0057] The hardware collection unit is configured to collect industrial robot hardware information corresponding to the corresponding industrial robot, and the industrial robot hardware information comprises robot component information and robot mechanical structure information.

[0058] The software collection unit is configured to collect industrial robot software information corresponding to the corresponding industrial robot, and the industrial robot software information comprises key component failure mechanism information, robot power driving information and robot communication control information.

[0059] The layout management unit is configured to analyze and process the industrial robot hardware information and the industrial robot software information, plan the sensor erection and arrangement corresponding to the industrial robot according to the industrial robot hardware information and the industrial robot software information, integrate the sensor erection and arrangement result, and set the sensor array layout information according to the integration result.

[0060] It should be further explained that, in the process of setting up and arranging the sensors corresponding to the industrial robot, firstly, a three-dimensional mechanical structure model corresponding to the industrial robot is set up according to the industrial robot hardware information and the industrial robot software information; a flexible composite sensor layout is performed on the corresponding space structure in the obtained three-dimensional mechanical structure model based on a global optimization algorithm, to generate a sensor layout three-dimensional model; the obtained sensor layout three-dimensional model is iteratively optimized in sequence until the corresponding sensor layout three-dimensional model meets the optimal solution of the fitness value, and is output as the sensor array layout information corresponding to the corresponding three-dimensional mechanical structure model.

[0061] In addition, the flexible composite sensors involved in the sensor array layout information are respectively used to acquire control monitoring information (current, pose, torque, oil temperature, etc.) inside the industrial robot and control monitoring information (vibration, noise, acoustic emission, etc.) of the mechanical and transmission systems outside the industrial robot, and involve multiple sensor devices.

[0062] In the specific implementation process, the multi-source data acquisition module includes:

[0063] The sensor acquisition unit is configured to acquire the corresponding sensor array layout information, acquire the control monitoring information corresponding to the position information of the corresponding flexible composite sensor according to the corresponding sensor array layout information, the control monitoring information including external mechanical control monitoring information, internal electrical control monitoring information, and drive transmission control monitoring information, and mark the corresponding control monitoring information;

[0064] The visual acquisition unit is configured to monitor the execution end of the corresponding industrial robot based on a binocular vision camera three-dimensional recognition technology, set multiple target monitoring points on the corresponding execution end according to the three-dimensional mechanical structure model corresponding to the corresponding industrial robot, acquire the target monitoring information corresponding to the multiple target monitoring points respectively, and mark the corresponding target monitoring information;

[0065] The acquisition integration unit is configured to integrate the marking results of the control monitoring information obtained by the sensor acquisition unit, the target monitoring information obtained by the visual acquisition unit, and the instruction monitoring information obtained by the signal acquisition unit respectively, to acquire the multi-source heterogeneous data set corresponding to the corresponding position in the three-dimensional mechanical structure model corresponding to the corresponding industrial robot.

[0066] It needs to be further explained that the corresponding multi-source heterogeneous data set is marked according to the three-dimensional mechanical structure model corresponding to the corresponding industrial robot, and the corresponding execution end is obtained according to the three-dimensional mechanical structure model. The obtained execution end is respectively associated and analyzed according to the corresponding external mechanical control monitoring information, internal electrical control monitoring information and drive transmission control monitoring information, the mechanical part structure and transmission control structure involved in the corresponding execution end are obtained, the multi-source heterogeneous data set corresponding to the corresponding execution end is set according to the information involved, and the corresponding industrial robot is marked in the three-dimensional mechanical structure model.

[0067] In the specific implementation process, the pose trajectory analysis module comprises:

[0068] Obtaining the multi-source heterogeneous data set obtained at the corresponding position in the three-dimensional mechanical structure model corresponding to the corresponding industrial robot in the multi-source data acquisition module;

[0069] The pose analysis unit is used for imaging processing of the target monitoring information obtained by each target monitoring point corresponding to the corresponding execution end in the industrial robot, establishing a three-dimensional space image corresponding to the corresponding execution end, and sequentially extracting features of the target monitoring data obtained by the corresponding target monitoring point in the three-dimensional space image according to the order of points, lines and surfaces. The end attitude plane corresponding to the corresponding industrial robot is obtained, and the corresponding pose data is obtained according to the end attitude plane;

[0070] It needs to be further explained that when setting the end attitude platform, the position distribution characteristics corresponding to the corresponding target monitoring point are preferentially obtained, the mean and variance are obtained, the stability and dispersion degree are obtained, and the corresponding target monitoring point is fed back according to the stability and dispersion degree; secondly, the coordinate information is obtained according to the feedback result of the corresponding target monitoring point, and each coordinate information corresponding to the adjacent target monitoring point is connected into a line segment according to the time sequence, the geometric characteristics of the corresponding line segment are calculated, and the straight line fitting processing is carried out according to the corresponding geometric line segment; when there are three or more than three target monitoring points not in a line, the corresponding plane information is constructed, the geometric characteristics corresponding to the line segment information are mapped into the corresponding plane information, the reference plane of the corresponding plane information is determined, and the attitude parameters such as rotation angle, translation distance, etc. are calculated. The corresponding pose data is obtained according to the corresponding attitude parameter, and the corresponding platform information is set as the end attitude platform.

[0071] The instruction analysis unit is configured to perform feature extraction on the electrical control monitoring information and the drive transmission control monitoring information in the control monitoring information corresponding to the industrial robot, obtain control monitoring information associated with an execution end of the industrial robot, and extract instruction data associated with the execution end of the industrial robot.

[0072] It should be further explained that the corresponding instruction data is obtained from the programming instructions involved in the internal electrical control monitoring information and the drive transmission control monitoring information corresponding to the corresponding industrial robot, and the corresponding component information and power drive information of the industrial robot involved in the corresponding programming instructions are uniformly marked to obtain the corresponding instruction data.

[0073] The comparison and fitting unit is configured to set actual observation values according to the pose data and the instruction data corresponding to the corresponding industrial robot, set the obtained pose data and instruction data as a historical fitting data set, analyze and process the historical fitting data set, and establish a kinematic model corresponding to the industrial robot.

[0074] The corresponding pose data and instruction data are input into the corresponding kinematic model for analysis and processing to obtain corresponding predicted model values. The obtained predicted model values are compared and analyzed with the actual observation values based on the least squares method. The kinematic model is optimized according to the comparison and analysis results, and the kinematic model after optimization is output.

[0075] In the specific implementation process, the precision calibration analysis module includes:

[0076] The running classification unit is configured to classify the instruction data and the pose data corresponding to the kinematic model, obtain a load state of the pose data corresponding to the corresponding instruction data, the load state including an empty load state and a load state, and mark the corresponding kinematic model according to the corresponding load state.

[0077] The precision calibration unit is configured to analyze and process the corresponding mark processing results at the corresponding positions in the kinematic model, identify the structure parameters of the corresponding kinematic model under the corresponding load state, and obtain the to-be-identified parameters.

[0078] The robot MD-H model is used to separate the linear and nonlinear parameters of the to-be-identified parameters in the corresponding kinematic model, obtain the linear parameters, the nonlinear parameters, and the structure parameter identification original problem.

[0079] The structure parameter identification original problem is converted into a separable nonlinear least squares problem. The obtained nonlinear least squares problem, linear parameters, and nonlinear parameters are solved to obtain the optimal value of the pose parameter corresponding to the to-be-identified parameters.

[0080] It should be further explained that, in the process of obtaining the optimal solution for the configuration parameters, the corresponding kinematic model is represented as follows: ,in, Let A be the corresponding actual pose data and the linear parameter. The relevant coefficient matrix, It is a nonlinear function. It is a nonlinear parameter vector;

[0081] For the linear parameter part, by minimizing the error function Where m is the corresponding sample size, the estimated value of the corresponding linear parameter is obtained. ;

[0082] For nonlinear parameter deployment, minimize the error function. The corresponding estimated values ​​are iteratively processed until the corresponding convergence conditions are met.

[0083] Based on the analysis results, obtain the optimal values ​​of the corresponding configuration parameters.

[0084] In practical implementation, the accuracy compensation analysis module includes:

[0085] The sampling management unit is used to acquire the instruction data corresponding to the industrial robot, classify the instruction data according to the corresponding workspace and operating range, acquire the pose data corresponding to the instruction data within the workspace and operating range, and set the acquired pose data as a candidate pose pool within the corresponding workspace and operating range.

[0086] The initial poses of industrial robots in the pool under different load states are obtained based on the random sampling method, and observability index and distance index are set according to the corresponding load states.

[0087] Among them, the observability index is obtained based on the Fisher information matrix; the distance index is obtained based on the spatial distance information of the components.

[0088] Based on observability and distance indicators, the corresponding sampling points in the workspace and operating range of the industrial robot are comprehensively processed, the comprehensive processing results are sorted, and the corresponding sampling point set is obtained based on the sorting results.

[0089] It should be further explained that the weighting coefficients for the observability metric OG and the distance metric OD are set according to the corresponding load conditions. and Based on the weighting coefficients of the corresponding indicator data, the comprehensive data S of the sampling points corresponding to the corresponding load state is obtained, where: The final sampling point set is obtained by selecting a certain number of poses from high to low according to the sorting results of the comprehensive data corresponding to each sampling point in the respective workspace and operating range.

[0090] The precision compensation unit is configured to obtain the sampling point set obtained by the industrial robot from the corresponding pose data in the corresponding workspace and operating range, analyze and process each sampling point in the sampling point set based on the corresponding kinematic model, obtain the corresponding predicted pose data, and obtain the actual pose data corresponding to each sampling point.

[0091] The error data corresponding to each sampling point is obtained by comparing and analyzing the predicted pose data and the actual pose data corresponding to each sampling point, the error rule data corresponding to each sampling point is obtained by statistically analyzing the error data corresponding to each sampling point, the error rule data is the statistical result of the error angle value and the error key position value corresponding to the respective sampling point, and the precision compensation processing is performed according to the error rule data.

[0092] The precision compensation processing results of the sampling points corresponding to the kinematic model analysis results of the corresponding pose data in the respective workspace and operating range are comprehensively marked.

[0093] The corresponding instruction data and pose data are obtained, the obtained data are input into the kinematic model corresponding to the respective industrial robot, the online monitoring result is obtained, the precision compensation processing is performed according to the respective sampling point, and the trajectory intelligent detection data corresponding to the respective industrial robot is obtained.

[0094] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and deformations can be made to the embodiments without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. An industrial robot trajectory intelligent detection system, characterized in that, Comprise: The acquisition array layout module is used for obtaining corresponding industrial robot software information and industrial robot hardware information, arranging sensors according to the obtained information, and setting sensor array layout information; The multi-source data acquisition module is used for acquiring corresponding control monitoring information and target monitoring information corresponding to each target monitoring point of the industrial robot according to the sensor array layout information and binocular vision technology, and generating a multi-source heterogeneous data set from the obtained monitoring information; The pose trajectory analysis module is used for analyzing and processing the control monitoring information and target monitoring information in the multi-source heterogeneous data set, obtaining corresponding instruction data and pose data, constructing a kinematic model according to the instruction data and pose data, fitting and comparing the kinematic model, and optimizing according to the fitting and comparison result; The precision calibration analysis module is used for identifying the structure parameters of the corresponding kinematic model of the industrial robot by establishing a linear-nonlinear parameter separation robot MD-H model, and obtaining the optimal value of the pose parameters corresponding to the corresponding structure and execution end in the industrial robot; The precision compensation analysis module is used for setting corresponding sampling points according to the optimal value of the pose parameters corresponding to the position of the industrial robot, establishing a corresponding candidate pose pool corresponding to the industrial robot for the corresponding sampling points, selecting corresponding sampling points, and performing parameter identification and precision compensation processing on the kinematic model analysis processing corresponding to the corresponding execution end according to the selected sampling points, to obtain the trajectory intelligent detection data corresponding to the corresponding industrial robot.

2. The industrial robot trajectory intelligent detection system according to claim 1, characterized in that, The acquisition array layout module comprises: A hardware acquisition unit is used for acquiring corresponding industrial robot hardware information, the industrial robot hardware information comprising robot component information and robot mechanical structure information; A software acquisition unit is used for acquiring corresponding industrial robot software information, the industrial robot software information comprising robot power drive information and robot communication control information; A layout management unit is used for planning the erection and arrangement of the corresponding flexible composite sensor of the industrial robot according to the industrial robot hardware information and the industrial robot software information, and setting the sensor array layout information.

3. The industrial robot trajectory intelligent detection system of claim 2, wherein, The multi-source data acquisition module comprises: A sensor acquisition unit is used for acquiring control monitoring information corresponding to the industrial robot according to the sensor array layout information, the control monitoring information comprising external mechanical control monitoring information, internal electrical control monitoring information, and drive transmission control monitoring information; A vision acquisition unit is used for monitoring a plurality of target monitoring points corresponding to the execution end of the corresponding industrial robot based on binocular vision technology, and acquiring target monitoring information corresponding to each target monitoring point; An acquisition integration unit is used for integrating the obtained control monitoring information and target monitoring information to generate a multi-source heterogeneous data set.

4. The industrial robot trajectory intelligent detection system according to claim 3, characterized in that, The pose trajectory analysis module comprises: Acquiring a multi-source heterogeneous data set corresponding to the corresponding industrial robot; The pose analysis unit is configured to perform imaging processing on target monitoring information obtained by corresponding target monitoring points corresponding to the execution end of the industrial robot, establish a three-dimensional space image corresponding to the execution end, perform feature extraction on target monitoring data obtained by the corresponding target monitoring points in the three-dimensional space image in sequence according to points, lines and planes, establish an end pose plane corresponding to the corresponding industrial robot, and obtain corresponding pose data according to the end pose plane. The instruction analysis unit is configured to perform feature extraction on corresponding electrical control monitoring information and drive transmission control monitoring information in control monitoring information corresponding to the industrial robot, obtain control monitoring information associated with the execution end, and extract instruction data associated with the execution end of the industrial robot.

5. The industrial robot trajectory intelligent detection system according to claim 4, wherein, The pose trajectory analysis module comprises: The contrast fitting unit is configured to set actual observation values according to the pose data and the instruction data corresponding to the corresponding industrial robot, set the obtained pose data and instruction data as a historical fitting data set, perform analysis processing on the historical fitting data set, and establish a kinematic model corresponding to the industrial robot. The pose data and the instruction data are input into the corresponding kinematic model for analysis processing, a predicted model value is obtained, the obtained predicted model value is compared and analyzed with the actual observation value based on the least square method, the kinematic model is optimized according to the comparison and analysis result, and the kinematic model after optimization processing is output.

6. The industrial robot trajectory intelligent detection system according to claim 5, wherein, The precision calibration analysis module comprises: The running classification unit is configured to perform classification processing on the instruction data and the pose data corresponding to the kinematic model, obtain a load state of the pose data corresponding to the corresponding instruction data, the load state comprises an empty load state and a load state, and mark the corresponding kinematic model according to the corresponding load state. The precision calibration unit is configured to perform analysis processing on the corresponding mark processing result at the corresponding position in the kinematic model, identify the structure parameters of the corresponding kinematic model in the corresponding load state, and obtain the to-be-identified parameters. The robot MD-H model is used to separate linear and nonlinear parameters of the to-be-identified parameters in the corresponding kinematic model, obtain linear parameters, nonlinear parameters and structure parameter identification original problems of the corresponding to-be-identified parameters, and convert the structure parameter identification original problem into a separable nonlinear least square problem. The obtained nonlinear least square problem, linear parameter and nonlinear parameter are solved respectively, and an optimal value of the pose parameter corresponding to the to-be-identified parameter is obtained.

7. The industrial robot trajectory intelligent detection system according to claim 6, wherein, The precision compensation analysis module comprises: The sampling management unit is configured to obtain instruction data corresponding to the corresponding industrial robot, perform classification processing on the corresponding instruction data according to corresponding workspaces and running ranges, obtain pose data corresponding to the corresponding instruction data in the workspaces and the running ranges, and set the obtained pose data as a candidate pose pool in the workspaces and the running ranges. The random sampling method is used to obtain initial poses of the industrial robot in different load states in the corresponding candidate pose pool, and an observability index and a distance index are set according to the corresponding load states. According to the observability index and the distance index, corresponding sampling points in the corresponding workspace and the corresponding operating range of the industrial robot are respectively comprehensively processed, the comprehensive processing results are sorted, and corresponding sampling point sets are obtained according to the sorting results.

8. The industrial robot trajectory intelligent detection system according to claim 7, characterized in that, The precision compensation analysis module comprises: A precision compensation unit is configured to obtain a sampling point set obtained from corresponding pose data in a corresponding workspace and operating range of an industrial robot, analyze each sampling point in the sampling point set based on a corresponding kinematic model, obtain corresponding predicted pose data, and obtain actual pose data corresponding to each sampling point; According to the predicted pose data and the actual pose data corresponding to each sampling point, comparative analysis is performed to obtain corresponding error data, statistical analysis is performed on the error data corresponding to each sampling point to obtain error rule data corresponding to each sampling point, and precision compensation processing is performed according to the error rule data; The precision compensation processing results of the sampling points corresponding to the kinematic model analysis results of the corresponding pose data in the corresponding workspace and operating range are comprehensively marked; Corresponding instruction data and pose data are obtained, the obtained data are input into a kinematic model corresponding to the corresponding industrial robot, online monitoring results are obtained, precision compensation processing is performed according to the corresponding sampling points, and trajectory intelligent detection data corresponding to the corresponding industrial robot are obtained.

Images