A mobile robot fault data troubleshooting method based on large language model

Abstract

The present invention relates to the field of mobile robot fault data troubleshooting and analysis, and in particular to a mobile robot fault data troubleshooting method based on a large language model, comprising: using control instruction data and real-time operating status data of the mobile robot to establish a comprehensive operating feature sequence of the mobile robot; establishing an occasional fault analysis model of the mobile robot based on the large language model according to the comprehensive operating feature sequence of the mobile robot; using the occasional fault analysis model of the mobile robot to perform fault recovery analysis and processing to obtain a mobile robot fault data troubleshooting result, and on the basis of integrating the large language model, extracting the input instructions and real-time status features of the mobile robot, and constructing independent analysis and screening models, which has the advantage of being able to more accurately identify and respond to diverse situations in complex environments.

Description

Technical Field

[0001] The present invention relates to the field of mobile robot fault data troubleshooting and analysis, and in particular to a mobile robot fault data troubleshooting method based on a large language model. Background Art

[0002] When faced with mobile robot failures, traditional troubleshooting solutions are often too general and fail to accurately locate the problem. Traditional solutions usually only provide a single process and lack flexibility, making them less applicable to different models of robots or different operating environments. When the implementation stability of the deployment environment cannot be guaranteed, or when the solution itself faces errors and needs to correct them, a stable, efficient and accurate method is needed. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention provides a mobile robot fault data troubleshooting method based on a large language model, which establishes a rapid analysis and troubleshooting model through the large language model, thereby achieving rapid detection and positioning of mobile robot faults.

[0004] To achieve the above object, the present invention provides a mobile robot fault data troubleshooting method based on a large language model, comprising:

[0005] S1. Establishing a comprehensive operation feature sequence of the mobile robot using the control instruction data and real-time operation status data of the mobile robot;

[0006] S2. Establishing an occasional fault analysis model of the mobile robot based on a large language model according to the comprehensive operation feature sequence of the mobile robot;

[0007] S3. Utilize the accidental fault analysis model of the mobile robot to perform fault recovery analysis and processing to obtain the mobile robot fault data troubleshooting result.

[0008] Preferably, the method of establishing a comprehensive operation feature sequence of the mobile robot using the control instruction data and real-time operation status data of the mobile robot comprises:

[0009] S1-1, respectively obtaining control instruction data and real-time operation status data of the mobile robot;

[0010] S1-2, according to the control instruction data of the mobile robot, obtaining the control instruction data type, control instruction data capacity and control instruction data time of the corresponding control instruction data as input data features of the mobile robot;

[0011] S1-3, acquiring posture adjustment data of the corresponding mobile robot according to the control instruction data type corresponding to the input data feature of the mobile robot;

[0012] S1-4, obtaining the data reception amount of the corresponding mobile robot according to the control instruction data capacity corresponding to the input data characteristics of the mobile robot;

[0013] S1-5, obtaining the current instruction time tag of the corresponding mobile robot according to the input data characteristics of the mobile robot corresponding to the control instruction data moment;

[0014] S1-6, determining whether the posture adjustment data of the mobile robot corresponds to the current instruction time tag, if so, executing S1-7, otherwise, updating the control instruction data of the mobile robot and returning to S1-1;

[0015] S1-7. Determine whether the data received by the mobile robot corresponds to the current instruction time label. If so, use the posture adjustment data, data received, current instruction time label and real-time operation status data of the mobile robot to obtain the comprehensive operation feature sequence of the mobile robot. Otherwise, update the control instruction data of the mobile robot and return to S1-1.

[0016] Furthermore, the comprehensive operation feature sequence of the mobile robot is obtained by using the posture adjustment data, data reception amount, current instruction time tag and real-time operation status data of the mobile robot, including:

[0017] S1-7-1, determine whether the posture adjustment data of the mobile robot corresponds to the data reception amount, if so, execute S1-7-2, otherwise, return to S1-3;

[0018] S1-7-2, establishing a time threshold of a running feature sequence according to the running time corresponding to the posture adjustment data;

[0019] S1-7-3, determine whether the current instruction time tag corresponds to the operation feature sequence time threshold, if so, execute S1-7-4, otherwise, return to S1-5;

[0020] S1-7-4, establishing a data sequence corresponding to the posture adjustment data of the mobile robot according to the time threshold of the operation feature sequence;

[0021] S1-7-5. Establish a data flow sequence corresponding to the data reception amount of the mobile robot according to the time threshold of the operation characteristic sequence;

[0022] S1-7-6. Use the data sequence corresponding to the posture adjustment data of the mobile robot and the data flow sequence corresponding to the data reception volume as the comprehensive operation feature sequence of the mobile robot.

[0023] Furthermore, establishing an occasional fault analysis model of the mobile robot based on a large language model according to the comprehensive operation feature sequence of the mobile robot includes:

[0024] S2-1, using the comprehensive operation feature sequence of the mobile robot to establish a fault location model of the mobile robot based on a large language model;

[0025] S2-2, establishing a fault analysis model of the mobile robot according to the fault location model of the mobile robot;

[0026] S2-3. Utilize the fault location model and fault analysis model of the mobile robot as an accidental fault analysis model of the mobile robot.

[0027] Furthermore, establishing a fault location model of the mobile robot based on a large language model using the comprehensive operation feature sequence of the mobile robot includes:

[0028] S2-1-1, obtaining a corresponding historical comprehensive operation feature sequence according to the comprehensive operation feature sequence of the mobile robot;

[0029] S2-1-2, respectively obtaining corresponding historical normal operation state data and historical abnormal operation state data according to the real-time operation state data corresponding to the historical comprehensive operation feature sequence;

[0030] S2-1-3, using the historical normal operation status data to divide into a positive training set;

[0031] S2-1-4, using the historical abnormal operation state data to divide into a negative training set;

[0032] S2-1-5. Using the positive training set and the negative training set as input, the historical comprehensive operation feature sequence corresponding to the positive training set and the historical comprehensive operation feature sequence corresponding to the negative training set as output, and training based on the large language model to establish a fault location model for the mobile robot.

[0033] Furthermore, establishing a fault analysis model of the mobile robot according to the fault location model of the mobile robot includes:

[0034] S2-2-1. Obtaining a corresponding historical comprehensive operation feature sequence according to the comprehensive operation feature sequence of the mobile robot;

[0035] S2-2-2, respectively obtaining corresponding historical normal operation state data and historical abnormal operation state data according to the real-time operation state data corresponding to the historical comprehensive operation feature sequence;

[0036] S2-2-3. Acquire corresponding input data features as a first data set according to the historical abnormal operation status data;

[0037] S2-2-4, using the historical abnormal operation status data as input and the first data set corresponding to the historical abnormal operation status data as output, training based on the large language model to establish a basic fault analysis model;

[0038] S2-2-5. Using the historical normal operation status data to input a basic fault analysis model to obtain a test model output of the basic fault analysis model;

[0039] S2-2-6. Determine whether the test model output of the basic fault analysis model corresponds to the first data set. If so, return to S2-2-4; otherwise, use the basic fault analysis model as the fault analysis model of the mobile robot.

[0040] Furthermore, the mobile robot fault data troubleshooting results obtained by using the mobile robot's occasional fault analysis model to perform fault recovery analysis processing include:

[0041] S3-1, using the accidental failure analysis model of the mobile robot to obtain an accidental failure analysis output result of the mobile robot;

[0042] S3-2. Perform fault recovery simulation processing based on the output results of the accidental fault analysis of the mobile robot to obtain the mobile robot fault data troubleshooting results.

[0043] Furthermore, the accidental failure analysis output results of the mobile robot obtained by using the accidental failure analysis model of the mobile robot include:

[0044] S3-1-1, using the mobile robot's occasional fault analysis model to correspond to the fault location model to obtain a fault location output result;

[0045] S3-1-2, using the occasional fault analysis model of the mobile robot to correspond to the fault analysis model to obtain a fault analysis output result;

[0046] S3-1-3, determine whether the fault location output result exists, if so, execute S3-1-4, otherwise, give up the process;

[0047] S3-1-4. Determine whether the fault location output result corresponds to the fault analysis output result. If so, use the fault location output result and the fault analysis output result as the occasional fault analysis output result of the mobile robot. Otherwise, return to S2-1-1.

[0048] Furthermore, the mobile robot fault data troubleshooting results obtained by performing fault recovery simulation processing according to the output results of the occasional fault analysis of the mobile robot include:

[0049] S3-2-1, obtaining a data sequence corresponding to the fault and a data reception amount corresponding to the fault according to the output result of the occasional fault analysis of the mobile robot and the output result of the fault location;

[0050] S3-2-2, according to the output result of the occasional fault analysis of the mobile robot, respectively obtain the corresponding fault control instruction data type, fault control instruction data capacity and fault control instruction data time corresponding to the fault analysis output result;

[0051] S3-2-3, obtaining corresponding historical verification input data features according to the data sequence with faults and the amount of data received with faults;

[0052] S3-2-4, determine whether the historical verification input data characteristics correspond to the fault control instruction data type. If so, output the faulty data sequence, the faulty data reception amount and the fault control instruction data type as the mobile robot fault data troubleshooting result. Otherwise, execute S3-2-5;

[0053] S3-2-5, determine whether the historical verification input data characteristics correspond to the fault control instruction data capacity, if so, output the data sequence with faults, the amount of data received with faults and the fault control instruction data capacity as the mobile robot fault data troubleshooting result, otherwise, execute S3-2-6;

[0054] S3-2-6, determine whether the historical verification input data characteristics correspond to the fault control instruction data time, if so, output the data sequence with faults, the amount of data received with faults and the fault control instruction data time as the mobile robot fault data troubleshooting result, otherwise, obtain the corresponding historical verification comprehensive operation characteristic sequence according to the fault control instruction data type, fault control instruction data capacity and fault control instruction data time;

[0055] S3-2-7, determine whether the historical verification comprehensive operation feature sequence corresponds to the data sequence with faults, if so, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data and the data sequence with faults as the mobile robot fault data troubleshooting result, otherwise, execute S3-2-8;

[0056] S3-2-8. Determine whether the historical verification comprehensive operation feature sequence corresponds to the data reception amount with faults. If so, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data and the data reception amount with faults as the mobile robot fault data troubleshooting result. Otherwise, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data time, the data sequence with faults and the data reception amount with faults as the mobile robot fault data troubleshooting result.

[0057] Compared with the closest prior art, the present invention has the following beneficial effects:

[0058] By using advanced algorithms and machine learning technology, combined with the robot's real-time data and historical operation records, the fault scope is gradually narrowed until the specific fault point is accurately located. In addition, based on the integration of large-scale language models, the input instructions and real-time status features of the mobile robot are extracted, and independent analysis and screening models are constructed. Its advantage lies in the ability to more accurately identify and respond to diverse situations in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of a mobile robot fault data troubleshooting method based on a large language model provided by the present invention. DETAILED DESCRIPTION

[0060] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0061] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] Embodiment 1:

[0063] The present invention provides a mobile robot fault data troubleshooting method based on a large language model, such as Figure 1 As shown, including:

[0064] S1. Establishing a comprehensive operation feature sequence of the mobile robot using the control instruction data and real-time operation status data of the mobile robot;

[0065] S2. Establishing an occasional fault analysis model of the mobile robot based on a large language model according to the comprehensive operation feature sequence of the mobile robot;

[0066] S3. Utilize the accidental fault analysis model of the mobile robot to perform fault recovery analysis and processing to obtain the mobile robot fault data troubleshooting result.

[0067] S1 specifically includes:

[0068] S1-1, respectively obtaining control instruction data and real-time operation status data of the mobile robot;

[0069] S1-2, according to the control instruction data of the mobile robot, obtaining the control instruction data type, control instruction data capacity and control instruction data time of the corresponding control instruction data as input data features of the mobile robot;

[0070] S1-3, acquiring posture adjustment data of the corresponding mobile robot according to the control instruction data type corresponding to the input data feature of the mobile robot;

[0071] S1-4, obtaining the data reception amount of the corresponding mobile robot according to the control instruction data capacity corresponding to the input data characteristics of the mobile robot;

[0072] S1-5, obtaining the current instruction time tag of the corresponding mobile robot according to the input data characteristics of the mobile robot corresponding to the control instruction data moment;

[0073] S1-6, determining whether the posture adjustment data of the mobile robot corresponds to the current instruction time tag, if so, executing S1-7, otherwise, updating the control instruction data of the mobile robot and returning to S1-1;

[0074] S1-7. Determine whether the data received by the mobile robot corresponds to the current instruction time label. If so, use the posture adjustment data, data received, current instruction time label and real-time operation status data of the mobile robot to obtain the comprehensive operation feature sequence of the mobile robot. Otherwise, update the control instruction data of the mobile robot and return to S1-1.

[0075] In this embodiment, a mobile robot fault data troubleshooting method based on a large language model is provided. In practical applications, the control instruction data type can be defined according to the instruction type received by the mobile robot, such as trajectory, action, adjustment parameters, etc.

[0076] S1-7 specifically includes:

[0077] S1-7-1, determine whether the posture adjustment data of the mobile robot corresponds to the data reception amount, if so, execute S1-7-2, otherwise, return to S1-3;

[0078] S1-7-2, establishing a time threshold of a running feature sequence according to the running time corresponding to the posture adjustment data;

[0079] S1-7-3, determine whether the current instruction time tag corresponds to the operation feature sequence time threshold, if so, execute S1-7-4, otherwise, return to S1-5;

[0080] S1-7-4, establishing a data sequence corresponding to the posture adjustment data of the mobile robot according to the time threshold of the operation feature sequence;

[0081] S1-7-5. Establish a data flow sequence corresponding to the data reception amount of the mobile robot according to the time threshold of the operation characteristic sequence;

[0082] S1-7-6. Use the data sequence corresponding to the posture adjustment data of the mobile robot and the data flow sequence corresponding to the data reception volume as the comprehensive operation feature sequence of the mobile robot.

[0083] S2 specifically includes:

[0084] S2-1, using the comprehensive operation feature sequence of the mobile robot to establish a fault location model of the mobile robot based on a large language model;

[0085] S2-2, establishing a fault analysis model of the mobile robot according to the fault location model of the mobile robot;

[0086] S2-3. Utilize the fault location model and fault analysis model of the mobile robot as an accidental fault analysis model of the mobile robot.

[0087] S2-1 specifically includes:

[0088] S2-1-1, obtaining a corresponding historical comprehensive operation feature sequence according to the comprehensive operation feature sequence of the mobile robot;

[0089] S2-1-2, respectively obtaining corresponding historical normal operation state data and historical abnormal operation state data according to the real-time operation state data corresponding to the historical comprehensive operation feature sequence;

[0090] S2-1-3, using the historical normal operation status data to divide into a positive training set;

[0091] S2-1-4, using the historical abnormal operation state data to divide into a negative training set;

[0092] S2-1-5. Using the positive training set and the negative training set as input, the historical comprehensive operation feature sequence corresponding to the positive training set and the historical comprehensive operation feature sequence corresponding to the negative training set as output, and training based on the large language model to establish a fault location model for the mobile robot.

[0093] S2-2 specifically includes:

[0094] S2-2-1. Obtaining a corresponding historical comprehensive operation feature sequence according to the comprehensive operation feature sequence of the mobile robot;

[0095] S2-2-2, respectively obtaining corresponding historical normal operation state data and historical abnormal operation state data according to the real-time operation state data corresponding to the historical comprehensive operation feature sequence;

[0096] S2-2-3. Acquire corresponding input data features as a first data set according to the historical abnormal operation status data;

[0097] S2-2-4, using the historical abnormal operation status data as input and the first data set corresponding to the historical abnormal operation status data as output, training based on the large language model to establish a basic fault analysis model;

[0098] S2-2-5. Using the historical normal operation status data to input a basic fault analysis model to obtain a test model output of the basic fault analysis model;

[0099] S2-2-6. Determine whether the test model output of the basic fault analysis model corresponds to the first data set. If so, return to S2-2-4; otherwise, use the basic fault analysis model as the fault analysis model of the mobile robot.

[0100] S3 specifically includes:

[0101] S3-1, using the accidental failure analysis model of the mobile robot to obtain an accidental failure analysis output result of the mobile robot;

[0102] S3-2. Perform fault recovery simulation processing based on the output results of the accidental fault analysis of the mobile robot to obtain the mobile robot fault data troubleshooting results.

[0103] S3-1 specifically includes:

[0104] S3-1-1, using the mobile robot's occasional fault analysis model to correspond to the fault location model to obtain a fault location output result;

[0105] S3-1-2, using the occasional fault analysis model of the mobile robot to correspond to the fault analysis model to obtain a fault analysis output result;

[0106] S3-1-3, determine whether the fault location output result exists, if so, execute S3-1-4, otherwise, give up the process;

[0107] S3-1-4. Determine whether the fault location output result corresponds to the fault analysis output result. If so, use the fault location output result and the fault analysis output result as the occasional fault analysis output result of the mobile robot. Otherwise, return to S2-1-1.

[0108] S3-2 specifically includes:

[0109] S3-2-1, obtaining a data sequence corresponding to the fault and a data reception amount corresponding to the fault according to the output result of the occasional fault analysis of the mobile robot and the output result of the fault location;

[0110] S3-2-2, according to the output result of the occasional fault analysis of the mobile robot, respectively obtain the corresponding fault control instruction data type, fault control instruction data capacity and fault control instruction data time corresponding to the fault analysis output result;

[0111] S3-2-3, obtaining corresponding historical verification input data features according to the data sequence with faults and the amount of data received with faults;

[0112] S3-2-4, determine whether the historical verification input data characteristics correspond to the fault control instruction data type. If so, output the faulty data sequence, the faulty data reception amount and the fault control instruction data type as the mobile robot fault data troubleshooting result. Otherwise, execute S3-2-5;

[0113] S3-2-5, determine whether the historical verification input data characteristics correspond to the fault control instruction data capacity, if so, output the data sequence with faults, the amount of data received with faults and the fault control instruction data capacity as the mobile robot fault data troubleshooting result, otherwise, execute S3-2-6;

[0114] S3-2-6, determine whether the historical verification input data characteristics correspond to the fault control instruction data time, if so, output the data sequence with faults, the amount of data received with faults and the fault control instruction data time as the mobile robot fault data troubleshooting result, otherwise, obtain the corresponding historical verification comprehensive operation characteristic sequence according to the fault control instruction data type, fault control instruction data capacity and fault control instruction data time;

[0115] S3-2-7, determine whether the historical verification comprehensive operation feature sequence corresponds to the data sequence with faults, if so, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data and the data sequence with faults as the mobile robot fault data troubleshooting result, otherwise, execute S3-2-8;

[0116] S3-2-8. Determine whether the historical verification comprehensive operation feature sequence corresponds to the data reception amount with faults. If so, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data and the data reception amount with faults as the mobile robot fault data troubleshooting result. Otherwise, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data time, the data sequence with faults and the data reception amount with faults as the mobile robot fault data troubleshooting result.

[0117] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0118] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0119] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A mobile robot fault data troubleshooting method based on a large language model, characterized in that: include: S1. Establishing a comprehensive operation feature sequence of the mobile robot using the control instruction data and real-time operation status data of the mobile robot; S1-1, respectively obtaining control instruction data and real-time operation status data of the mobile robot; S1-2, according to the control instruction data of the mobile robot, obtaining the control instruction data type, control instruction data capacity and control instruction data time of the corresponding control instruction data as input data features of the mobile robot; S1-3, acquiring posture adjustment data of the corresponding mobile robot according to the control instruction data type corresponding to the input data feature of the mobile robot; S1-4, obtaining the data reception amount of the corresponding mobile robot according to the control instruction data capacity corresponding to the input data characteristics of the mobile robot; S1-5, obtaining the current instruction time tag of the corresponding mobile robot according to the input data characteristics of the mobile robot corresponding to the control instruction data moment; S1-6, determining whether the posture adjustment data of the mobile robot corresponds to the current instruction time tag, if so, executing S1-7, otherwise, updating the control instruction data of the mobile robot and returning to S1-1; S1-7, judging whether the data received by the mobile robot corresponds to the current instruction time label, if so, obtaining the comprehensive operation feature sequence of the mobile robot by using the posture adjustment data, data received, current instruction time label and real-time operation status data of the mobile robot, otherwise, updating the control instruction data of the mobile robot and returning to S1-1; S1-7-1, determine whether the posture adjustment data of the mobile robot corresponds to the data reception amount, if so, execute S1-7-2, otherwise, return to S1-3; S1-7-2, establishing a time threshold of a running feature sequence according to the running time corresponding to the posture adjustment data; S1-7-3, determine whether the current instruction time tag corresponds to the operation feature sequence time threshold, if so, execute S1-7-4, otherwise, return to S1-5; S1-7-4, establishing a data sequence corresponding to the posture adjustment data of the mobile robot according to the time threshold of the operation feature sequence; S1-7-5. Establish a data flow sequence corresponding to the data reception amount of the mobile robot according to the time threshold of the operation characteristic sequence; S1-7-6, using the data sequence corresponding to the posture adjustment data of the mobile robot and the data flow sequence corresponding to the data reception amount as the comprehensive operation feature sequence of the mobile robot; S2. Establishing an occasional fault analysis model of the mobile robot based on a large language model according to the comprehensive operation feature sequence of the mobile robot; S2-1, using the comprehensive operation feature sequence of the mobile robot to establish a fault location model of the mobile robot based on a large language model; S2-2, establishing a fault analysis model of the mobile robot according to the fault location model of the mobile robot; S2-3, using the fault location model and fault analysis model of the mobile robot as an occasional fault analysis model of the mobile robot; S3, using the accidental fault analysis model of the mobile robot to perform fault recovery analysis and processing to obtain the mobile robot fault data troubleshooting result; S3-1, using the accidental failure analysis model of the mobile robot to obtain an accidental failure analysis output result of the mobile robot; S3-2. Perform fault recovery simulation processing based on the output results of the accidental fault analysis of the mobile robot to obtain the mobile robot fault data troubleshooting results.

2. A mobile robot fault data troubleshooting method based on a large language model as described in claim 1, characterized in that: Establishing a fault location model of the mobile robot based on a large language model using the comprehensive operation feature sequence of the mobile robot includes: S2-1-1, obtaining a corresponding historical comprehensive operation feature sequence according to the comprehensive operation feature sequence of the mobile robot; S2-1-2, respectively obtaining corresponding historical normal operation state data and historical abnormal operation state data according to the real-time operation state data corresponding to the historical comprehensive operation feature sequence; S2-1-3, using the historical normal operation status data to divide into a positive training set; S2-1-4, using the historical abnormal operation state data to divide into a negative training set; S2-1-5. Using the positive training set and the negative training set as input, the historical comprehensive operation feature sequence corresponding to the positive training set and the historical comprehensive operation feature sequence corresponding to the negative training set as output, and training based on the large language model to establish a fault location model for the mobile robot.

3. A mobile robot fault data troubleshooting method based on a large language model as described in claim 2, characterized in that: Establishing a fault analysis model of the mobile robot according to the fault location model of the mobile robot includes: S2-2-1. Obtaining a corresponding historical comprehensive operation feature sequence according to the comprehensive operation feature sequence of the mobile robot; S2-2-2, respectively obtaining corresponding historical normal operation state data and historical abnormal operation state data according to the real-time operation state data corresponding to the historical comprehensive operation feature sequence; S2-2-3. Acquire corresponding input data features as a first data set according to the historical abnormal operation status data; S2-2-4, using the historical abnormal operation status data as input and the first data set corresponding to the historical abnormal operation status data as output, training based on the large language model to establish a basic fault analysis model; S2-2-5. Using the historical normal operation status data to input a basic fault analysis model to obtain a test model output of the basic fault analysis model; S2-2-6. Determine whether the test model output of the basic fault analysis model corresponds to the first data set. If so, return to S2-2-4; otherwise, use the basic fault analysis model as the fault analysis model of the mobile robot.

4. A mobile robot fault data troubleshooting method based on a large language model as claimed in claim 1, characterized in that: The accidental failure analysis output results of the mobile robot obtained by using the accidental failure analysis model of the mobile robot include: S3-1-1, using the mobile robot's occasional fault analysis model to correspond to the fault location model to obtain a fault location output result; S3-1-2, using the occasional fault analysis model of the mobile robot to correspond to the fault analysis model to obtain a fault analysis output result; S3-1-3, determine whether the fault location output result exists, if so, execute S3-1-4, otherwise, give up the process; S3-1-4. Determine whether the fault location output result corresponds to the fault analysis output result. If so, use the fault location output result and the fault analysis output result as the occasional fault analysis output result of the mobile robot. Otherwise, return to S2-1-1.

5. A mobile robot fault data troubleshooting method based on a large language model as claimed in claim 4, characterized in that: According to the output results of the occasional fault analysis of the mobile robot, the fault recovery simulation processing is performed to obtain the mobile robot fault data troubleshooting results including: S3-2-1, obtaining a data sequence corresponding to the fault and a data reception amount corresponding to the fault according to the output result of the occasional fault analysis of the mobile robot and the output result of the fault location; S3-2-2, according to the output result of the occasional fault analysis of the mobile robot, respectively obtain the corresponding fault control instruction data type, fault control instruction data capacity and fault control instruction data time corresponding to the fault analysis output result; S3-2-3, obtaining corresponding historical verification input data features according to the data sequence with faults and the amount of data received with faults; S3-2-4, determine whether the historical verification input data characteristics correspond to the fault control instruction data type. If so, output the faulty data sequence, the faulty data reception amount and the fault control instruction data type as the mobile robot fault data troubleshooting result. Otherwise, execute S3-2-5; S3-2-5, determine whether the historical verification input data characteristics correspond to the fault control instruction data capacity, if so, output the data sequence with faults, the amount of data received with faults and the fault control instruction data capacity as the mobile robot fault data troubleshooting result, otherwise, execute S3-2-6; S3-2-6, determine whether the historical verification input data characteristics correspond to the fault control instruction data time, if so, output the data sequence with faults, the amount of data received with faults and the fault control instruction data time as the mobile robot fault data troubleshooting result, otherwise, obtain the corresponding historical verification comprehensive operation characteristic sequence according to the fault control instruction data type, fault control instruction data capacity and fault control instruction data time; S3-2-7, determine whether the historical verification comprehensive operation feature sequence corresponds to the data sequence with faults, if so, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data and the data sequence with faults as the mobile robot fault data troubleshooting result, otherwise, execute S3-2-8; S3-2-8. Determine whether the historical verification comprehensive operation feature sequence corresponds to the data reception amount with faults. If so, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data and the data reception amount with faults as the mobile robot fault data troubleshooting result. Otherwise, output the fault control instruction data type, fault control instruction data capacity, fault control instruction data time, the data sequence with faults and the data reception amount with faults as the mobile robot fault data troubleshooting result.

Images