CN-122022765-A - Vehicle fault detection method, device, equipment, storage medium and program product

CN122022765ACN 122022765 ACN122022765 ACN 122022765ACN-122022765-A

Abstract

The application discloses a vehicle fault checking method, a device, equipment, a storage medium and a program product, which relate to the technical field of fault diagnosis and comprise the steps of collecting real-time running state data of a target vehicle to be checked under the condition that a fault consultation request of a user is detected; and inputting the real-time running state data and the fault description information in the fault consultation request into a pre-trained generation type large model, generating a fault troubleshooting flow through the generation type large model, and executing fault troubleshooting on the target vehicle based on the fault troubleshooting flow. The real-time running state data of the vehicle and the fault checking flow are deeply fused through the generated large model, the running state of the vehicle is used as the basis for generating the fault checking flow, the fault checking flow can be customized according to the real-time state of the vehicle and used for checking the fault of the target vehicle, redundant checking steps can be prevented from being executed under the condition that the state of the vehicle is clear, and therefore checking efficiency is remarkably improved.

Inventors

ZHANG QIYU
LIN ZHIGUI
FU GUANG
ZHANG TAO
XIAO LEI

Assignees

上汽通用五菱汽车股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260113

Claims (10)

1. The vehicle fault checking method is characterized by comprising the following steps of: Under the condition that a fault consultation request of a user is detected, acquiring real-time running state data of a target vehicle to be checked; Inputting the real-time running state data and the fault description information in the fault consultation request into a pre-trained generation type large model, and generating a fault troubleshooting flow through the generation type large model; and performing troubleshooting on the target vehicle based on the troubleshooting process.
2. The method according to claim 1, wherein the troubleshooting process includes at least one troubleshooting step, expected verification conditions corresponding to each of the troubleshooting steps, and a jump rule between steps triggered based on an execution result of the troubleshooting step, and the step of performing troubleshooting of the target vehicle based on the troubleshooting process includes: based on the fault checking flow, a fault checking step of the target vehicle is executed one by one; For a currently executed target investigation step, acquiring verification data for representing an execution effect of the target investigation step; Comparing the verification data with expected verification conditions corresponding to the target investigation step to obtain an execution result of the target investigation step; and carrying out decision on the execution result based on the jump rule, and jumping to the next fault checking step of the target checking step according to the decision result and executing until the target checking step is the last fault checking step of the fault checking flow.
3. The vehicle troubleshooting method of claim 1, wherein said inputting the real-time operation state data and the fault description information in the fault consultation request into a pre-trained generative large model, generating a troubleshooting flow through the generative large model, includes: inputting the real-time running state data and the fault description information in the fault consultation request into a pre-trained generation type large model, and matching a target troubleshooting template from a plurality of preset basic troubleshooting templates through the generation type large model based on the fault description information; and dynamically adjusting the flow nodes of the target troubleshooting template by using the generated large model based on the real-time running state data to generate a troubleshooting flow.
4. The vehicle troubleshooting method as set forth in claim 3, wherein said step of matching a target troubleshooting template from among a preset plurality of basic troubleshooting templates by said generated large model based on said fault description information further includes: Classifying and summarizing the historical fault cases based on the fault types, the fault phenomena, the troubleshooting step sequences and the fault solutions of the historical fault cases to form a plurality of fault categories; for each fault category, extracting a core node shared by fault investigation under the fault category by adopting a fault tree analysis method, and determining the dependency relationship and the execution sequence among the core nodes; And constructing a basic investigation template corresponding to each fault category based on the core node, the dependency relationship and the execution sequence.
5. The vehicle troubleshooting method as set forth in any one of claims 1 to 4, wherein said inputting the real-time operation state data and the fault description information in the fault consultation request into a pre-trained generation-type large model, before the step of generating a troubleshooting stream by the generation-type large model, further includes: Constructing a pairing corpus based on flow branches in a fault solution of a historical fault case and historical data of a vehicle running state corresponding to the historical fault case; constructing a sample data set based on the matching corpus and the fault description information of the historical fault cases; and performing field fine adjustment on the generated basic large model by using the sample data set to obtain a pre-trained generated large model.
6. The vehicle trouble shooting method according to any one of claims 1 to 4, characterized in that the step of inputting the real-time running state data and the trouble shooting information in the trouble shooting request into a pre-trained generative large model, generating a trouble shooting flow by the generative large model, further comprises: filtering and cleaning the real-time running state data to obtain effective data; Packaging the structured data in the effective data into structured data objects according to a preset key value pair format; extracting fault keywords from unstructured data in the effective data and the fault description information; and generating a structured fault label based on the fault keyword.
7. A vehicle fault checking device is characterized in that, the vehicle trouble shooting device includes: The data acquisition module is used for acquiring real-time running state data of the target vehicle to be checked under the condition that the fault consultation request of the user is detected; The flow generating module is used for inputting the real-time running state data and the fault description information in the fault consultation request into a pre-trained generation type large model, and generating a fault troubleshooting flow through the generation type large model; and the flow execution module is used for executing the fault investigation of the target vehicle based on the fault investigation flow.
8. A vehicle troubleshooting device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the vehicle troubleshooting method of any one of claims 1 to 6.
9. A storage medium, characterized in that the storage medium is a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, realizes the steps of the vehicle troubleshooting method according to any one of claims 1 to 6.
10. A computer program product, characterized in that it comprises a computer program which, when executed by a processor, implements the steps of the vehicle troubleshooting method according to any one of claims 1 to 6.

Description

Vehicle fault detection method, device, equipment, storage medium and program product Technical Field The present application relates to the field of fault diagnosis technologies, and in particular, to a vehicle fault detection method, device, apparatus, storage medium, and program product. Background With the continuous improvement of the functional complexity of the intelligent driving system, tens of sensors and hundreds of control modules including ultrasonic radars, millimeter wave radars and laser radars are integrated, so that the after-sale fault investigation scene of the intelligent network-connected automobile is changed from the traditional single hardware fault to the complex software-hardware cooperative fault. However, existing troubleshooting procedures and decision-making mechanisms are static and predefined and cannot adapt to the dynamically changing real-time operating conditions of the vehicle. Specifically, the fault checking flow is disjointed with the real-time state of the vehicle, the existing mode relies on fixed linear flow or static history case matching, and when the checking flow is generated, the real-time running state data (such as sensor shielding condition, network flow, software version and the like) of the vehicle is not used as decision basis. This results in lack of pertinence to the troubleshooting process of the vehicle, and all the preset steps must be executed regardless of the actual state of the vehicle, resulting in a large number of redundant operations, which seriously affect the troubleshooting efficiency. The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present application and is not intended to represent an admission that the foregoing is prior art. Disclosure of Invention The application mainly aims to provide a vehicle fault checking method, device, equipment, storage medium and program product, and aims to solve the technical problems that the existing fault checking flow and decision mechanism are static and predefined and cannot be matched with the real-time running state of the dynamic change of a vehicle, so that the fault checking efficiency is low. In order to achieve the above object, the present application provides a vehicle fault detection method, which includes: Under the condition that a fault consultation request of a user is detected, acquiring real-time running state data of a target vehicle to be checked; Inputting the real-time running state data and the fault description information in the fault consultation request into a pre-trained generation type large model, and generating a fault troubleshooting flow through the generation type large model; and performing troubleshooting on the target vehicle based on the troubleshooting process. In one embodiment, the troubleshooting process includes at least one troubleshooting step, expected verification conditions corresponding to each troubleshooting step, and a jump rule between steps triggered based on an execution result of the troubleshooting step, and the step of performing troubleshooting on the target vehicle based on the troubleshooting process includes: based on the fault checking flow, a fault checking step of the target vehicle is executed one by one; For a currently executed target investigation step, acquiring verification data for representing an execution effect of the target investigation step; Comparing the verification data with expected verification conditions corresponding to the target investigation step to obtain an execution result of the target investigation step; and carrying out decision on the execution result based on the jump rule, and jumping to the next fault checking step of the target checking step according to the decision result and executing until the target checking step is the last fault checking step of the fault checking flow. In an embodiment, the step of inputting the real-time running state data and the fault description information in the fault consultation request into a pre-trained generation type large model, and generating a fault troubleshooting flow through the generation type large model includes: inputting the real-time running state data and the fault description information in the fault consultation request into a pre-trained generation type large model, and matching a target troubleshooting template from a plurality of preset basic troubleshooting templates through the generation type large model based on the fault description information; and dynamically adjusting the flow nodes of the target troubleshooting template by using the generated large model based on the real-time running state data to generate a troubleshooting flow. In an embodiment, before the step of matching the target troubleshooting template from the preset plurality of basic troubleshooting templates by the generated big model based on the fault description information, the method further includes: Classif