CN-122020243-A - Vehicle diagnosis function construction method and device based on multi-source data and intelligent analysis

CN122020243ACN 122020243 ACN122020243 ACN 122020243ACN-122020243-A

Abstract

The application relates to the technical field of data processing and discloses a vehicle diagnosis function construction method and related equipment based on multi-source data and intelligent analysis, wherein the method comprises the steps of collecting original diagnosis data aiming at a target vehicle diagnosis function from at least two mutually independent compliance data sources; the method comprises the steps of carrying out standardized processing on original diagnosis data to generate a unified standardized diagnosis data description file, inputting the standardized diagnosis data description file into an intelligent analysis center for reasoning and verifying to obtain verified diagnosis function logic, and automatically generating a diagnosis function framework for integrated use of a diagnosis platform according to the verified diagnosis function logic. The application can realize the compliance and high-efficiency development of the vehicle diagnosis function and solve the problems of low development efficiency and insufficient legal compliance in the prior art.

Inventors

XIE YANCHUN
LI CHANGAN
JIANG YONGXIN

Assignees

南昌市星纬软件开发有限公司

Dates

Publication Date: 20260512
Application Date: 20260112

Claims (10)

1. A vehicle diagnosis function construction method based on multi-source data and intelligent analysis is characterized by comprising the following steps: collecting raw diagnostic data for a target vehicle diagnostic function from at least two mutually independent compliance data sources; Carrying out standardized processing on the original diagnostic data to generate a unified standardized diagnostic data description file; Inputting the standardized diagnosis data description file into an intelligent analysis center for reasoning and verification to obtain verified diagnosis function logic; And automatically generating a diagnosis function framework for integrated use of the diagnosis platform according to the verified diagnosis function logic.
2. The method for constructing a vehicle diagnostic function based on multi-source data and intelligent analysis according to claim 1, wherein the collecting raw diagnostic data for a target vehicle diagnostic function from at least two mutually independent compliant data sources comprises: Collecting first original data generated by executing a standard diagnosis process from original factory diagnosis equipment authorized by a host factory; collecting second original data generated by testing the diagnostic function of the same target vehicle from third-party diagnostic equipment subjected to market verification; and acquiring desensitization historical data related to the diagnosis function of the target vehicle from a legal authorized third party data service party as third initial data.
3. The vehicle diagnostic function construction method based on multi-source data and intelligent analysis according to claim 1, wherein the normalizing the raw diagnostic data to generate a unified normalized diagnostic data description file comprises: Performing data cleaning and analysis on all the collected original diagnostic data, and extracting to obtain diagnostic logic elements for forming a diagnostic logic function, wherein the diagnostic logic elements comprise diagnostic service sequences, diagnostic service parameters and diagnostic response information; and converting the format of the diagnosis logic element according to a preset automobile diagnosis data description standard to generate the standardized diagnosis data description file.
4. The method for constructing a vehicle diagnostic function based on multi-source data and intelligent analysis according to claim 1, wherein the step of inputting the standardized diagnostic data description file to an intelligent analysis center for reasoning and verification to obtain verified diagnostic function logic comprises the steps of: Calling a plurality of analysis models in the intelligent analysis center, and carrying out collaborative analysis on the standardized diagnosis data description file to obtain a preliminary reduced diagnosis function logic, wherein the analysis models sequentially analyze the standardized diagnosis data description file, and an analysis result of the former model is used as input of the latter model; Invoking a verification model in the intelligent analysis center, and performing cross verification and integrity check on the primarily restored diagnosis function logic by utilizing original diagnosis data or known diagnosis logic from different data sources to obtain a verification result; and according to the verification result, complementing and correcting logic missing or conflict existing in the primarily restored diagnosis function logic to form the verified diagnosis function logic.
5. The method for constructing a vehicle diagnostic function based on multi-source data and intelligent analysis according to claim 4, wherein the calling a plurality of analysis models in the intelligent analysis center, performing collaborative analysis on the standardized diagnostic data description file, and obtaining a preliminary restored diagnostic function logic, includes: A protocol identification model is called to analyze the standardized diagnostic data description file, and a communication protocol and a vehicle type platform which are followed are identified, so that a protocol identification result is obtained; Based on the protocol identification result, a state machine restoration model is called to analyze the diagnosis service sequence in the standardized diagnosis data description file, and state transition logic of a target vehicle diagnosis function is restored; and calling a semantic understanding model, and carrying out functional semantic annotation on the diagnosis identifier and the diagnosis service parameter in the state transition logic.
6. The method for constructing a vehicle diagnostic function based on multi-source data and intelligent analysis according to claim 5, wherein the calling a plurality of analysis models in the intelligent analysis center performs collaborative analysis on the standardized diagnostic data description file to obtain a preliminary reduced diagnostic function logic, and further comprising: And calling a knowledge migration model in the process of restoring and/or diagnosing identifiers of the state transition logic and semantic labeling of the diagnosis service parameters, and recommending a diagnosis logic template of a similar vehicle model platform for the current analysis target vehicle diagnosis function based on the pre-constructed diagnosis knowledge graph so as to assist in restoring the state transition logic and semantic labeling of the diagnosis service parameters.
7. The vehicle diagnostic function construction method based on multi-source data and intelligent analysis according to claim 1, wherein the automatically generating a diagnostic function framework for integrated use of a diagnostic platform according to the verified diagnostic function logic comprises: Generating a diagnosis data module conforming to a standard format according to the verified diagnosis function logic, wherein the diagnosis data module is used for being imported into a diagnosis platform to be directly executed; And generating corresponding Application Program Interface (API) codes according to the verified diagnosis function logic, wherein the API codes are used for secondary integration and calling by developers.
8. The vehicle diagnostic function construction method based on multi-source data and intelligent analysis according to claim 1, wherein the automatically generating a diagnostic function framework for integrated use of a diagnostic platform according to the verified diagnostic function logic further comprises: And generating a graphical configuration interface file according to the verified diagnosis function logic, wherein the graphical configuration interface file is used for a user to adjust control parameters and thresholds in the diagnosis function logic in a visual mode.
9. The method for constructing a vehicle diagnostic function based on multi-source data and intelligent analysis according to claim 1, wherein if the target vehicle diagnostic function is a battery management system BMS calibration function, the standardized diagnostic data description file is input to an intelligent analysis center for reasoning and verification, and the verified diagnostic function logic is further comprised of: after the validated diagnostic function logic is obtained, corresponding safety protection rules are injected into the validated diagnostic function logic based on historical data collected from the compliance data source and including typical failure modes.
10. A vehicle diagnostic function construction apparatus based on multi-source data and intelligent analysis, comprising: The data acquisition module is used for acquiring original diagnosis data aiming at a diagnosis function of the target vehicle from at least two mutually independent compliance data sources; The data processing module is used for carrying out standardized processing on the original diagnostic data and generating a unified standardized diagnostic data description file; the reasoning verification module is used for inputting the standardized diagnosis data description file into the intelligent analysis center for reasoning and verification to obtain verified diagnosis function logic; and the frame generation module is used for automatically generating a diagnosis function frame for the integrated use of the diagnosis platform according to the verified diagnosis function logic.

Description

Vehicle diagnosis function construction method and device based on multi-source data and intelligent analysis Technical Field The application relates to the technical field of data processing, in particular to a vehicle diagnosis function construction method and device based on multi-source data and intelligent analysis. Background Along with the improvement of the level of electronization and intellectualization of automobiles, the vehicle diagnosis function becomes a key for guaranteeing the safe operation of the vehicles, and the market has increasingly urgent requirements for the rapid development and the large-scale application of the cross-brand and multi-vehicle type diagnosis function. At present, the development of the existing vehicle diagnosis function mainly depends on reverse engineering, and communication protocols and diagnosis logics of ECUs (Electronic Control Unit, electronic control units) are presumed through monitoring vehicle bus messages and manually analyzing, but the method needs to develop reverse work one by one aiming at each model of vehicle and each ECU, so that the development period of a single diagnosis function is long, the efficiency is low, the requirements of quick market response and large-scale coverage are difficult to meet, meanwhile, the intellectual property rights of a host factory are easily violated by unauthorized protocol analysis behaviors, obvious legal and compliance risks exist, and the marketization popularization of diagnosis products is limited. In summary, the prior art has the problems of low development efficiency and insufficient legal compliance, and cannot adapt to the requirements of the rapid development of the automobile industry on diagnostic techniques. The foregoing description is provided for general background information and does not necessarily constitute prior art. Disclosure of Invention The embodiment of the application provides a vehicle diagnosis function construction method and device based on multi-source data and intelligent analysis, which can realize the compliance and high-efficiency development of the vehicle diagnosis function and solve the problems of low development efficiency and insufficient legal compliance in the prior art. In a first aspect, an embodiment of the present application provides a vehicle diagnostic function construction method based on multi-source data and intelligent analysis, including: collecting raw diagnostic data for a target vehicle diagnostic function from at least two mutually independent compliance data sources; Carrying out standardized processing on the original diagnostic data to generate a unified standardized diagnostic data description file; Inputting the standardized diagnosis data description file into an intelligent analysis center for reasoning and verification to obtain verified diagnosis function logic; And automatically generating a diagnosis function framework for integrated use of the diagnosis platform according to the verified diagnosis function logic. Further, in some embodiments of the present application, the collecting raw diagnostic data for a target vehicle diagnostic function from at least two mutually independent sources of compliance data includes: Collecting first original data generated by executing a standard diagnosis process from original factory diagnosis equipment authorized by a host factory; collecting second original data generated by testing the diagnostic function of the same target vehicle from third-party diagnostic equipment subjected to market verification; and acquiring desensitization historical data related to the diagnosis function of the target vehicle from a legal authorized third party data service party as third initial data. Further, in some embodiments of the present application, the normalizing the raw diagnostic data to generate a unified normalized diagnostic data description file includes: Performing data cleaning and analysis on all the collected original diagnostic data, and extracting to obtain diagnostic logic elements for forming a diagnostic logic function, wherein the diagnostic logic elements comprise diagnostic service sequences, diagnostic service parameters and diagnostic response information; and converting the format of the diagnosis logic element according to a preset automobile diagnosis data description standard to generate the standardized diagnosis data description file. Further, in some embodiments of the present application, the inputting the standardized diagnostic data description file into an intelligent analysis center for reasoning and verification, to obtain verified diagnostic function logic, includes: Calling a plurality of analysis models in the intelligent analysis center, and carrying out collaborative analysis on the standardized diagnosis data description file to obtain a preliminary reduced diagnosis function logic, wherein the analysis models sequentially analyze the standardized diagnosis data descriptio