Control method and system for chemical pipeline corridor inspection robot

Abstract

An embodiment of the present invention provides a control method and system for a chemical pipeline corridor inspection robot. This method obtains the operating data of a pipeline corridor inspection robot to be inspected, performs an anomaly analysis on this data using a method based on a robot operation anomaly analysis network, and generates a corresponding anomaly distribution vector. This anomaly analysis network uses combined knowledge learning, is trained using a template robot operation data sequence, and utilizes a target anomaly decision service in conjunction with the anomaly distribution vector of the pipeline corridor inspection robot operation data to determine an anomaly decision result for the robot operation data. This method can perform anomaly analysis and decision-making on the pipeline corridor inspection robot operation data, thereby accurately determining the robot's operational anomaly and providing a solution for repair. This method can improve the reliability and efficiency of the pipeline corridor inspection robot, reduce the impact of operational failures on inspection tasks, and enhance the overall effectiveness of pipeline corridor inspections.

Description

Technical Field

[0001] The present invention relates to the technical field of inspection robots, and in particular to a control method and system for a chemical pipeline corridor inspection robot. Background Art

[0002] Existing technologies present several challenges in analyzing and determining operational anomalies for chemical pipeline corridor inspection robots. Traditional methods often rely on manually set thresholds or rules to identify robot anomalies, and these methods struggle to adapt to complex and changing inspection environments. Furthermore, existing technologies can lack accuracy and reliability when handling different types of anomalies. Summary of the Invention

[0003] In view of this, the purpose of an embodiment of the present invention is to provide a control method and system for a chemical pipeline corridor inspection robot. By acquiring the operating data of the pipeline corridor inspection robot to be detected, the data is analyzed for abnormalities using a method based on a robot operation abnormality analysis network to generate corresponding abnormal distribution vectors. The abnormality analysis network uses combined knowledge learning, uses a template robot operation data sequence for training, and uses a target abnormality decision service combined with the abnormal distribution vector of the pipeline corridor inspection robot operation data to determine the abnormal decision result of the robot operation data. It can realize abnormal analysis and decision-making of the pipeline corridor inspection robot operation data, thereby accurately judging the abnormal operation of the robot and providing solutions for repair. This can improve the reliability and efficiency of the pipeline corridor inspection robot, reduce the impact of operational failures on inspection tasks, and improve the overall effect of pipeline corridor inspection.

[0004] According to one aspect of an embodiment of the present invention, a control method and system for a chemical pipeline corridor inspection robot are provided, the method comprising:

[0005] Obtain the operating data of the pipeline corridor inspection robot to be tested;

[0006] Based on the robot operation anomaly analysis network, an anomaly analysis is performed on the operation data of the pipeline corridor inspection robot to be detected, and an anomaly distribution vector of the operation data of the pipeline corridor inspection robot to be detected is generated; wherein, the robot operation anomaly analysis network is generated by combining knowledge learning of a first machine learning algorithm and a second machine learning algorithm based on a template robot operation data sequence, the template robot operation data sequence includes first template robot operation data, second template robot operation data and anomaly vector annotation data, the second template robot operation data is generated based on the first template robot operation data, and the combined knowledge learning includes to-be-labeled knowledge learning and comparative knowledge learning;

[0007] Determine an abnormality decision result of the operation data of the pipeline corridor inspection robot to be detected based on the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected and a target abnormality decision service, wherein the target abnormality decision service includes one or more of a sensor fault decision service, a motion system fault decision service, and a navigation fault decision service;

[0008] Based on the abnormal decision result of the operation data of the corridor inspection robot to be detected, a solution is issued to the control end corresponding to the corresponding target corridor inspection robot, so that the control end can repair the software and hardware of the target corridor inspection robot based on the issued solution, and then re-initiate the control task of the target corridor inspection robot.

[0009] In an alternative embodiment, the method further comprises:

[0010] Obtaining the first template robot operation data and corresponding abnormal vector annotation data;

[0011] Updating the parameters of the first machine learning algorithm based on the first template robot operation data and the corresponding abnormal vector annotation data to generate a first machine learning algorithm with updated parameters;

[0012] Generate second template robot operation data based on the first machine learning algorithm after the parameter update and the first template robot operation data;

[0013] Based on the first template robot operation data, the second template robot operation data and the corresponding abnormal vector annotation data, the second machine learning algorithm and the first machine learning algorithm after the parameter update are combined for knowledge learning to generate a robot operation abnormality analysis network, wherein the second machine learning algorithm is a parallel machine learning algorithm of the first machine learning algorithm.

[0014] In an alternative embodiment, based on the first template robot operation data, the second template robot operation data, and the corresponding abnormal vector annotation data, the second machine learning algorithm and the first machine learning algorithm after the parameter update are combined for knowledge learning to generate a robot operation abnormality analysis network, including:

[0015] Processing the first template robot operating data and the second template robot operating data based on the first machine learning algorithm after the parameter update, respectively, to generate abnormality estimation data and a first mapping vector for the first template robot operating data, and abnormality estimation data for the second template robot operating data, wherein the first mapping vector includes vectors of frequently responded data areas in the first template robot operating data;

[0016] Processing the second template robot operating data based on a second machine learning algorithm and a corresponding mapping function to generate a second mapping vector, where the second mapping vector includes a vector of an infrequent response data region in the second template robot operating data;

[0017] Determining a target learning error value based on the abnormality estimation data of the first template robot operation data, the abnormality estimation data of the second template robot operation data, the corresponding abnormality vector annotation data, the first mapping vector, and the second mapping vector;

[0018] Based on the target learning error value, the parameters of the second machine learning algorithm and the first machine learning algorithm after the parameter update are updated to generate a robot operation abnormality analysis network, which includes the first machine learning algorithm after the parameter update.

[0019] In an alternative embodiment, determining a target learning error value based on the anomaly estimation data of the first template robot operating data, the anomaly estimation data of the second template robot operating data, the corresponding anomaly vector annotation data, the first mapping vector, and the second mapping vector includes:

[0020] Determining a first learning error value based on the abnormality estimation data of the first template robot operation data and the corresponding abnormality vector annotation data;

[0021] Determining a second learning error value based on the abnormality estimation data of the second template robot operation data and the corresponding abnormality vector annotation data;

[0022] determining a comparative learning error value based on the first mapping vector and the second mapping vector;

[0023] A target learning error value is determined based on the first learning error value, the second learning error value, and the comparison learning error value.

[0024] In an alternative embodiment, generating the second template robot operating data based on the first machine learning algorithm after the parameter update and the first template robot operating data includes:

[0025] Processing the first template robot operation data based on the first machine learning algorithm after the parameter update to generate a target vector saliency map of the first template robot operation data;

[0026] Second template robot operation data is generated based on the first template robot operation data and the target vector saliency map.

[0027] In an alternative embodiment, generating the second template robot operation data based on the first template robot operation data and the target vector saliency map includes:

[0028] performing dimensionality reduction on the target vector saliency map based on the robot operation attention node of the first template robot operation data to generate a target vector saliency map after dimensionality reduction;

[0029] Performing regularization transformation on the target vector saliency map after dimensionality reduction; fusing the target vector saliency map after regularization transformation with the first template robot operation data to generate the second template robot operation data.

[0030] In an alternative embodiment, the operation data of the pipeline corridor inspection robot to be detected is carried in a service request for the operation data of the pipeline corridor inspection robot, and the operation data service request also includes an abnormal decision service ID of a target abnormal decision service; the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected and the target abnormal decision service are used to determine the abnormal decision result of the operation data of the pipeline corridor inspection robot to be detected, including:

[0031] Based on the abnormality decision service ID of the target abnormality decision service, determining the abnormality decision unit corresponding to the target abnormality decision service from at least one abnormality decision unit included in the robot operation abnormality analysis network;

[0032] Based on the abnormal decision unit corresponding to the target abnormal decision service, the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected is processed, and the corresponding abnormal decision result is generated and then sent.

[0033] According to another aspect of an embodiment of the present invention, a control method and system for a chemical pipeline corridor inspection robot are provided, the system comprising:

[0034] An acquisition unit, used to acquire the operating data of the pipeline corridor inspection robot to be detected;

[0035] A generating unit is configured to perform an anomaly analysis on the operation data of the pipeline corridor inspection robot to be detected based on a robot operation anomaly analysis network, and generate an anomaly distribution vector of the operation data of the pipeline corridor inspection robot to be detected; wherein the robot operation anomaly analysis network is generated by performing combined knowledge learning of a first machine learning algorithm and a second machine learning algorithm based on a template robot operation data sequence, the template robot operation data sequence including first template robot operation data, second template robot operation data and anomaly vector annotation data, the second template robot operation data is generated based on the first template robot operation data, and the combined knowledge learning includes to-be-labeled knowledge learning and comparative knowledge learning;

[0036] A determination unit is configured to determine an abnormality decision result of the operation data of the pipeline corridor inspection robot to be detected based on an abnormality distribution vector of the operation data of the pipeline corridor inspection robot to be detected and a target abnormality decision service, wherein the target abnormality decision service includes one or more of a sensor fault decision service, a motion system fault decision service, and a navigation fault decision service;

[0037] The sending unit is used to send a solution to the control end corresponding to the corresponding target corridor inspection robot based on the abnormal decision result of the operating data of the corridor inspection robot to be detected, so that the control end can repair the software and hardware of the target corridor inspection robot based on the sent solution and then re-initiate the control task of the target corridor inspection robot.

[0038] According to another aspect of an embodiment of the present invention, a server is provided, comprising: a memory and a processor, wherein a pipeline corridor inspection robot operation data processing program is stored on the memory, and when the pipeline corridor inspection robot operation data processing program is executed by the processor, the steps of the pipeline corridor inspection robot operation data processing method described in any one of the above items are implemented.

[0039] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for processing operation data of a pipeline corridor inspection robot described in any one of the above items are implemented.

[0040] Based on any of the above aspects, this application has the following technical effects:

[0041] First, the system acquires the operational data of the pipeline corridor inspection robot to be tested. Then, using a method based on a robot operation anomaly analysis network, it performs an anomaly analysis on this data and generates corresponding anomaly distribution vectors. This anomaly analysis network is trained using a template robot operation data sequence (including the first template robot operation data, the second template robot operation data, and anomaly vector annotation data) through combined knowledge learning. Next, a target anomaly decision service is used to combine the anomaly distribution vectors of the pipeline corridor inspection robot's operational data to determine an anomaly decision result for the robot's operational data. Target anomaly decision services include sensor fault decision services, motion system fault decision services, and navigation fault decision services. Finally, based on the anomaly decision result, a solution is issued to the control terminal corresponding to the target pipeline corridor inspection robot, enabling software and hardware repairs to be performed on the target robot and re-initiating control of the robot. This technology enables anomaly analysis and decision-making of pipeline corridor inspection robot operational data, accurately identifying robot operational anomalies and providing solutions for repair. This improves the reliability and efficiency of pipeline corridor inspection robots, reduces the impact of operational failures on inspection tasks, and enhances the overall effectiveness of pipeline corridor inspections.

[0042] In order to make the above-mentioned objects, features and advantages of the embodiments of the present invention more obvious and easy to understand, the embodiments will be described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 A schematic diagram showing components of a server provided by an embodiment of the present invention is shown;

[0045] Figure 2 A schematic flow chart of a control method for a chemical pipeline corridor inspection robot provided by an embodiment of the present invention is shown;

[0046] Figure 3 The figure shows a functional module block diagram of a chemical pipeline corridor inspection robot control system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To help students in this technical field better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0048] The terms "first," "second," "third," and the like (if any) in the description and claims of the present invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present invention described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions, e.g., a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0049] Figure 1 A schematic diagram of exemplary components of server 100 is shown. Server 100 may include one or more processors 104, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. Server 100 may also include any storage medium 106 for storing any type of information, such as code, settings, data, and the like. For example, and without limitation, storage medium 106 may include any one or more combinations of the following: any type of RAM, any type of ROM, a flash memory device, a hard disk, an optical disk, and the like. More generally, any storage medium may use any technology to store information. Furthermore, any storage medium may provide volatile or non-volatile retention of information. Furthermore, any storage medium may represent a fixed or removable component of server 100. In one embodiment, when processor 104 executes associated instructions stored in any storage medium or combination of storage media, server 100 may perform any operation of the associated instructions. Server 100 also includes one or more drive units 108, such as a hard disk drive unit, an optical disk drive unit, and the like, for interacting with any storage medium.

[0050] The server 100 also includes input / output 110 (I / O) for receiving various inputs (via input unit 112) and for providing various outputs (via output unit 114). One specific output mechanism may include a presentation device 116 and an associated graphical user interface (GUI) 118. The server 100 may also include one or more network interfaces 120 for exchanging data with other devices via one or more communication units 122. One or more communication buses 124 couple the components described above together.

[0051] The communication unit 122 can be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication unit 122 can include any combination of hardwired links, wireless links, routers, gateway functions, name servers 100, etc., governed by any protocol or combination of protocols.

[0052] Figure 2 The flow chart of the control method and system of the chemical pipeline corridor inspection robot provided by the embodiment of the present invention is shown. The control method and system of the chemical pipeline corridor inspection robot can be Figure 1 The server 100 shown in the figure is executed, and the detailed steps of the chemical pipeline corridor inspection robot control method are introduced as follows.

[0053] Step S110, obtaining the operation data of the pipeline corridor inspection robot to be inspected;

[0054] Step S120: performing an anomaly analysis on the operation data of the pipeline corridor inspection robot to be detected based on the robot operation anomaly analysis network to generate an anomaly distribution vector of the operation data of the pipeline corridor inspection robot to be detected; wherein the robot operation anomaly analysis network is generated by performing combined knowledge learning of a first machine learning algorithm and a second machine learning algorithm based on a template robot operation data sequence, the template robot operation data sequence includes first template robot operation data, second template robot operation data and anomaly vector annotation data, the second template robot operation data is generated based on the first template robot operation data, and the combined knowledge learning includes to-be-labeled knowledge learning and comparative knowledge learning;

[0055] Step S130: determining an abnormality decision result of the operation data of the pipeline corridor inspection robot to be detected based on the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected and a target abnormality decision service, wherein the target abnormality decision service includes one or more of a sensor fault decision service, a motion system fault decision service, and a navigation fault decision service;

[0056] Step S140: Based on the abnormal decision result of the operation data of the pipeline corridor inspection robot to be detected, a solution is issued to the control end corresponding to the corresponding target pipeline corridor inspection robot, so that the control end repairs the software and hardware of the target pipeline corridor inspection robot based on the issued solution, and then re-initiates the control task of the target pipeline corridor inspection robot.

[0057] Based on the above steps, this embodiment obtains the operating data of the pipeline corridor inspection robot to be tested, performs an anomaly analysis on this data using a method based on the robot operation anomaly analysis network, and generates a corresponding anomaly distribution vector. This anomaly analysis network uses combined knowledge learning, uses a template robot operation data sequence for training, and uses the target anomaly decision service combined with the anomaly distribution vector of the pipeline corridor inspection robot operation data to determine the abnormal decision result of the robot operation data. It can realize anomaly analysis and decision-making of the pipeline corridor inspection robot operation data, thereby accurately judging the robot's operational anomaly and providing solutions for repair. This can improve the reliability and efficiency of the pipeline corridor inspection robot, reduce the impact of operational failures on inspection tasks, and enhance the overall effectiveness of pipeline corridor inspection.

[0058] In an alternative embodiment, the method further comprises:

[0059] Obtaining the first template robot operation data and corresponding abnormal vector annotation data;

[0060] Updating the parameters of the first machine learning algorithm based on the first template robot operation data and the corresponding abnormal vector annotation data to generate a first machine learning algorithm with updated parameters;

[0061] Generate second template robot operation data based on the first machine learning algorithm after the parameter update and the first template robot operation data;

[0062] Based on the first template robot operation data, the second template robot operation data and the corresponding abnormal vector annotation data, the second machine learning algorithm and the first machine learning algorithm after the parameter update are combined for knowledge learning to generate a robot operation abnormality analysis network, wherein the second machine learning algorithm is a parallel machine learning algorithm of the first machine learning algorithm.

[0063] In an alternative embodiment, based on the first template robot operation data, the second template robot operation data, and the corresponding abnormal vector annotation data, the second machine learning algorithm and the first machine learning algorithm after the parameter update are combined for knowledge learning to generate a robot operation abnormality analysis network, including:

[0064] Processing the first template robot operating data and the second template robot operating data based on the first machine learning algorithm after the parameter update, respectively, to generate abnormality estimation data and a first mapping vector for the first template robot operating data, and abnormality estimation data for the second template robot operating data, wherein the first mapping vector includes vectors of frequently responded data areas in the first template robot operating data;

[0065] Processing the second template robot operating data based on a second machine learning algorithm and a corresponding mapping function to generate a second mapping vector, where the second mapping vector includes a vector of an infrequent response data region in the second template robot operating data;

[0066] Determining a target learning error value based on the abnormality estimation data of the first template robot operation data, the abnormality estimation data of the second template robot operation data, the corresponding abnormality vector annotation data, the first mapping vector, and the second mapping vector;

[0067] Based on the target learning error value, the parameters of the second machine learning algorithm and the first machine learning algorithm after the parameter update are updated to generate a robot operation abnormality analysis network, which includes the first machine learning algorithm after the parameter update.

[0068] In an alternative embodiment, determining a target learning error value based on the anomaly estimation data of the first template robot operating data, the anomaly estimation data of the second template robot operating data, the corresponding anomaly vector annotation data, the first mapping vector, and the second mapping vector includes:

[0069] Determining a first learning error value based on the abnormality estimation data of the first template robot operation data and the corresponding abnormality vector annotation data;

[0070] Determining a second learning error value based on the abnormality estimation data of the second template robot operation data and the corresponding abnormality vector annotation data;

[0071] determining a comparative learning error value based on the first mapping vector and the second mapping vector;

[0072] A target learning error value is determined based on the first learning error value, the second learning error value, and the comparison learning error value.

[0073] In an alternative embodiment, generating the second template robot operating data based on the first machine learning algorithm after the parameter update and the first template robot operating data includes:

[0074] Processing the first template robot operation data based on the first machine learning algorithm after the parameter update to generate a target vector saliency map of the first template robot operation data;

[0075] Second template robot operation data is generated based on the first template robot operation data and the target vector saliency map.

[0076] In an alternative embodiment, generating the second template robot operation data based on the first template robot operation data and the target vector saliency map includes:

[0077] performing dimensionality reduction on the target vector saliency map based on the robot operation attention node of the first template robot operation data to generate a target vector saliency map after dimensionality reduction;

[0078] Performing regularization transformation on the target vector saliency map after dimensionality reduction; fusing the target vector saliency map after regularization transformation with the first template robot operation data to generate the second template robot operation data.

[0079] In an alternative embodiment, the operation data of the pipeline corridor inspection robot to be detected is carried in a service request for the operation data of the pipeline corridor inspection robot, and the operation data service request also includes an abnormal decision service ID of a target abnormal decision service; the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected and the target abnormal decision service are used to determine the abnormal decision result of the operation data of the pipeline corridor inspection robot to be detected, including:

[0080] Based on the abnormality decision service ID of the target abnormality decision service, determining the abnormality decision unit corresponding to the target abnormality decision service from at least one abnormality decision unit included in the robot operation abnormality analysis network;

[0081] Based on the abnormal decision unit corresponding to the target abnormal decision service, the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected is processed, and the corresponding abnormal decision result is generated and then sent.

[0082] Figure 3 The functional module diagram of the chemical pipeline corridor inspection robot control system 200 provided by an embodiment of the present invention is shown. The functions implemented by the chemical pipeline corridor inspection robot control system 200 can correspond to the steps executed by the above method. The chemical pipeline corridor inspection robot control system 200 can be understood as the above server 100, or the processor of the server 100, or can also be understood as a component independent of the above server 100 or processor that implements the functions of the present invention under the control of the server 100, such as Figure 3As shown, the functions of each functional module of the chemical pipeline corridor inspection robot control system 200 are explained in detail below.

[0083] An acquisition unit 210 is used to acquire the operation data of the pipeline corridor inspection robot to be detected;

[0084] A generating unit 220 is configured to perform an anomaly analysis on the operation data of the pipeline corridor inspection robot to be detected based on a robot operation anomaly analysis network, and generate an anomaly distribution vector of the operation data of the pipeline corridor inspection robot to be detected; wherein the robot operation anomaly analysis network is generated by performing combined knowledge learning of a first machine learning algorithm and a second machine learning algorithm based on a template robot operation data sequence, the template robot operation data sequence including first template robot operation data, second template robot operation data, and anomaly vector annotation data, the second template robot operation data being generated based on the first template robot operation data, and the combined knowledge learning including to-be-labeled knowledge learning and comparative knowledge learning;

[0085] A determining unit 230 is configured to determine an abnormality decision result of the operation data of the pipeline corridor inspection robot to be detected based on the abnormality distribution vector of the operation data of the pipeline corridor inspection robot to be detected and a target abnormality decision service, wherein the target abnormality decision service includes one or more of a sensor fault decision service, a motion system fault decision service, and a navigation fault decision service;

[0086] The sending unit 240 is used to send a solution to the control end corresponding to the corresponding target corridor inspection robot based on the abnormal decision result of the operating data of the corridor inspection robot to be detected, so that the control end can repair the software and hardware of the target corridor inspection robot based on the sent solution and then re-initiate the control task of the target corridor inspection robot.

[0087] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0088] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations that come within the meaning and range of equivalents of the claims be embraced therein.

Claims

1. A control method for a chemical pipeline corridor inspection robot, characterized in that: The method comprises: Obtain the operating data of the pipeline corridor inspection robot to be tested; Based on the robot operation anomaly analysis network, an anomaly analysis is performed on the operation data of the pipeline corridor inspection robot to be detected, and an anomaly distribution vector of the operation data of the pipeline corridor inspection robot to be detected is generated; wherein, the robot operation anomaly analysis network is generated by combining knowledge learning of a first machine learning algorithm and a second machine learning algorithm based on a template robot operation data sequence, the template robot operation data sequence includes first template robot operation data, second template robot operation data and anomaly vector annotation data, the second template robot operation data is generated based on the first template robot operation data, and the combined knowledge learning includes to-be-labeled knowledge learning and comparative knowledge learning; Determine an abnormality decision result of the operation data of the pipeline corridor inspection robot to be detected based on the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected and a target abnormality decision service, wherein the target abnormality decision service includes one or more of a sensor fault decision service, a motion system fault decision service, and a navigation fault decision service; Based on the abnormal decision result of the operation data of the pipeline corridor inspection robot to be detected, a solution is issued to the control end corresponding to the corresponding target pipeline corridor inspection robot, so that the control end repairs the software and hardware of the target pipeline corridor inspection robot based on the issued solution and then re-initiates the control task of the target pipeline corridor inspection robot; The method further comprises: Obtaining the first template robot operation data and corresponding abnormal vector annotation data; Updating the parameters of the first machine learning algorithm based on the first template robot operation data and the corresponding abnormal vector annotation data to generate a first machine learning algorithm with updated parameters; Generate second template robot operation data based on the first machine learning algorithm after the parameter update and the first template robot operation data; Based on the first template robot operation data, the second template robot operation data and the corresponding abnormal vector annotation data, the second machine learning algorithm and the first machine learning algorithm after the parameter update are combined for knowledge learning to generate a robot operation abnormality analysis network, wherein the second machine learning algorithm is a parallel machine learning algorithm of the first machine learning algorithm.

2. The chemical pipeline corridor inspection robot control method according to claim 1 is characterized in that: The method of performing combined knowledge learning on the second machine learning algorithm and the first machine learning algorithm after the parameter update based on the first template robot operation data, the second template robot operation data, and the corresponding abnormal vector annotation data to generate a robot operation abnormality analysis network includes: Processing the first template robot operating data and the second template robot operating data based on the first machine learning algorithm after the parameter update, respectively, to generate abnormality estimation data and a first mapping vector for the first template robot operating data, and abnormality estimation data for the second template robot operating data, wherein the first mapping vector includes vectors of frequently responded data areas in the first template robot operating data; Processing the second template robot operating data based on a second machine learning algorithm and a corresponding mapping function to generate a second mapping vector, where the second mapping vector includes a vector of an infrequent response data region in the second template robot operating data; Determining a target learning error value based on the abnormality estimation data of the first template robot operation data, the abnormality estimation data of the second template robot operation data, the corresponding abnormality vector annotation data, the first mapping vector, and the second mapping vector; Based on the target learning error value, the parameters of the second machine learning algorithm and the first machine learning algorithm after the parameter update are updated to generate a robot operation abnormality analysis network, which includes the first machine learning algorithm after the parameter update.

3. The chemical pipeline corridor inspection robot control method according to claim 2 is characterized in that: The determining of a target learning error value based on the abnormality estimation data of the first template robot operation data, the abnormality estimation data of the second template robot operation data, the corresponding abnormality vector annotation data, the first mapping vector, and the second mapping vector includes: Determining a first learning error value based on the abnormality estimation data of the first template robot operation data and the corresponding abnormality vector annotation data; Determining a second learning error value based on the abnormality estimation data of the second template robot operation data and the corresponding abnormality vector annotation data; determining a comparative learning error value based on the first mapping vector and the second mapping vector; A target learning error value is determined based on the first learning error value, the second learning error value, and the comparison learning error value.

4. The chemical pipeline corridor inspection robot control method according to any one of claims 1 to 3, characterized in that: The generating second template robot operation data based on the first machine learning algorithm after the parameter update and the first template robot operation data includes: Processing the first template robot operation data based on the first machine learning algorithm after the parameter update to generate a target vector saliency map of the first template robot operation data; Second template robot operation data is generated based on the first template robot operation data and the target vector saliency map.

5. The chemical pipeline corridor inspection robot control method according to claim 4 is characterized in that: The generating of second template robot operation data based on the first template robot operation data and the target vector saliency map includes: performing dimensionality reduction on the target vector saliency map based on the robot operation attention node of the first template robot operation data to generate a target vector saliency map after dimensionality reduction; Performing regularization transformation on the target vector saliency map after dimensionality reduction; fusing the target vector saliency map after regularization transformation with the first template robot operation data to generate the second template robot operation data.

6. The chemical pipeline corridor inspection robot control method according to claim 1 is characterized in that: The operation data of the pipeline corridor inspection robot to be detected is carried in a service request for the operation data of the pipeline corridor inspection robot, and the operation data service request also includes an abnormal decision service ID of a target abnormal decision service; the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected and the target abnormal decision service are used to determine the abnormal decision result of the operation data of the pipeline corridor inspection robot to be detected, including: Based on the abnormality decision service ID of the target abnormality decision service, determining the abnormality decision unit corresponding to the target abnormality decision service from at least one abnormality decision unit included in the robot operation abnormality analysis network; Based on the abnormal decision unit corresponding to the target abnormal decision service, the abnormal distribution vector of the operation data of the pipeline corridor inspection robot to be detected is processed, and the corresponding abnormal decision result is generated and then sent.

7. A chemical pipeline corridor inspection robot control system, characterized in that: The chemical pipeline corridor inspection robot control system includes: An acquisition unit, used to acquire the operating data of the pipeline corridor inspection robot to be detected; A generating unit is configured to perform an anomaly analysis on the operation data of the pipeline corridor inspection robot to be detected based on a robot operation anomaly analysis network, and generate an anomaly distribution vector of the operation data of the pipeline corridor inspection robot to be detected; wherein the robot operation anomaly analysis network is generated by performing combined knowledge learning of a first machine learning algorithm and a second machine learning algorithm based on a template robot operation data sequence, the template robot operation data sequence including first template robot operation data, second template robot operation data and anomaly vector annotation data, the second template robot operation data is generated based on the first template robot operation data, and the combined knowledge learning includes to-be-labeled knowledge learning and comparative knowledge learning; A determination unit is configured to determine an abnormality decision result of the operation data of the pipeline corridor inspection robot to be detected based on an abnormality distribution vector of the operation data of the pipeline corridor inspection robot to be detected and a target abnormality decision service, wherein the target abnormality decision service includes one or more of a sensor fault decision service, a motion system fault decision service, and a navigation fault decision service; A sending unit is used to send a solution to the control terminal corresponding to the corresponding target corridor inspection robot based on the abnormal decision result of the operation data of the corridor inspection robot to be detected, so that the control terminal repairs the software and hardware of the target corridor inspection robot based on the sent solution and re-initiates the control task of the target corridor inspection robot; Among them, the generation methods of the robot operation abnormality analysis network include: Obtaining the first template robot operation data and corresponding abnormal vector annotation data; Updating the parameters of the first machine learning algorithm based on the first template robot operation data and the corresponding abnormal vector annotation data to generate a first machine learning algorithm with updated parameters; Generate second template robot operation data based on the first machine learning algorithm after the parameter update and the first template robot operation data; Based on the first template robot operation data, the second template robot operation data and the corresponding abnormal vector annotation data, the second machine learning algorithm and the first machine learning algorithm after the parameter update are combined for knowledge learning to generate a robot operation abnormality analysis network, wherein the second machine learning algorithm is a parallel machine learning algorithm of the first machine learning algorithm.

8. A server, characterized in that: The server includes: a memory and a processor, wherein the memory stores a pipeline corridor inspection robot operation data processing program, and when the pipeline corridor inspection robot operation data processing program is executed by the processor, the steps of the chemical pipeline corridor inspection robot control method as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the chemical pipeline corridor inspection robot control method according to any one of claims 1 to 6 are implemented.

Images