CN-121996697-A - Intelligent analysis method, system, equipment and medium for intelligent networking vehicle test data

CN121996697ACN 121996697 ACN121996697 ACN 121996697ACN-121996697-A

Abstract

The invention belongs to the technical field of vehicle testing, and particularly relates to an intelligent analysis method, an intelligent analysis system, intelligent analysis equipment and an intelligent analysis medium for intelligent network vehicle testing data, wherein the intelligent analysis method, the intelligent analysis system, the intelligent analysis equipment and the intelligent network vehicle testing medium are used for receiving initial testing data and marking test item identifiers based on testing events; generating a candidate test item option set according to the test requirement of a user, aggregating related data to form a target test item, generating a test strategy comprising a test scene, steps and judgment indexes by combining a knowledge base, analyzing the data required by the strategy, outputting an execution scheme, monitoring the environment and the data flow in real time in execution, automatically repairing and recording when abnormal, and if the test case fails, returning to the related abnormal record, and retrying after repairing if the abnormal record is confirmed to be caused by the abnormal condition. The invention realizes the intellectualization and automation of the test scheme formulation, improves the multiplexing rate of the historical test data, and improves the reliability of test execution and the accuracy of results through environmental monitoring and failure analysis.

Inventors

LI WEIBING
WU SHENGGANG
Qi Geshan
CHEN HAIYAN
HAN WEI
CHEN GAOHUA

Assignees

北京科技职业大学
北京丰华源码信息技术有限公司

Dates

Publication Date: 20260508
Application Date: 20251230

Claims (10)

1. An intelligent analysis method for intelligent network-connected vehicle test data is characterized by comprising the following steps: S101, receiving vehicle test initial data, identifying a test event corresponding to the data, associating the vehicle test initial data according to the test event, adding a test item identifier, and storing the marked vehicle test initial data into a database; S102, receiving test requirement information of a user, analyzing the test requirement information, and generating a test item option set containing a plurality of candidate test item identifiers; S103, receiving a selection instruction triggered by the selection of a user from the test item option set, and determining a target test item identifier selected by the user based on the selection instruction; s104, retrieving and retrieving initial vehicle test data of all the associated identifiers from the database according to the target test item identifiers, and aggregating to form a target test item data set; S105, analyzing by combining a preset automobile test knowledge base based on the target test item identification and the test requirement information to generate a test strategy aiming at the target test item, wherein the test strategy comprises a test scene, a test step and a judgment index; S106, based on the target test item data set, analyzing the data sufficiency and the characteristic coincidence of the test scene and the judgment index defined in the test strategy, and generating a test execution scheme comprising test parameter suggestions and test case priority ordering according to the analysis result; S107, in the process of executing the test according to the test execution scheme, monitoring the test environment and the test data stream in real time, executing a preset repair operation when abnormality is monitored, and recording an abnormality event and an associated target test item identifier to the database; S108, when any test case in the test execution scheme fails to be executed, inquiring whether an associated unresolved abnormal event record exists in the database according to a target test item identifier associated with the test case, if so, judging that the failure is caused by an abnormality, and re-executing the test case after the abnormality is repaired.
2. The intelligent analysis method for intelligent network-oriented vehicle test data according to claim 1, wherein S101 comprises the steps of: Receiving the original data streams of the vehicle-mounted controller, the vehicle sensor and the test bench, and analyzing and converting the original data streams into standardized data frames with uniform time stamp sequences; The method comprises the steps of configuring an event recognition rule base, inputting standardized data frames into the event recognition rule base for frame-by-frame matching and state machine judgment, and judging and outputting event type labels of test events when the content of the data frames continuously meets specific trigger conditions; establishing a matching relation between event type labels and test item identifiers; And writing the acquired test item identification, the corresponding standardized data frame and the timestamp thereof into a designated data table of the database in the form of an association record.
3. The intelligent analysis method for intelligent network-oriented vehicle test data according to claim 1, wherein S102 comprises the steps of: Receiving test requirement description text input by a user through a human-computer interaction interface; Calling to perform word segmentation and keyword extraction operation on the test requirement description text, and inputting the extracted keyword set into a pre-constructed test item identification database to perform fuzzy matching and semantic association query; and according to the matching query result, all the test item identifiers with the association degree higher than a preset threshold value are retrieved from the test item identifier database, and are sequenced according to the order of the association degree from high to low, and a candidate test item identifier list is generated to be used as the test item option set.
4. The intelligent analysis method for intelligent network-oriented vehicle test data according to claim 1, wherein S104 comprises the steps of: constructing a query statement according to the target test item identification, wherein the query statement takes the target test item identification as a query condition and points to a data table for storing marked vehicle test initial data in a database; extracting all records meeting the conditions from the data table, and packaging the vehicle test initial data entity and the metadata thereof in each record into an intermediate data object; And performing timestamp alignment and data field merging operation on the acquired plurality of intermediate data objects, removing repeated time segment data, and integrating the time segment data according to time sequences to generate a data set as a target test item data set.
5. The intelligent analysis method for intelligent network-oriented vehicle test data according to claim 1, wherein S105 comprises the steps of: the extracted key words of the requirements are screened to obtain key scene elements based on environment, road and performance, and the key scene elements are combined with target test item identifiers to generate query vectors; The elements in the query vector are taken as entry nodes, and map traversal is started in a preset automobile test knowledge base, the knowledge base is stored in a map form, the nodes represent test scene elements, vehicle parts or judging indexes, and the edges represent logic, subordinate or causal association among the nodes; the traversal process searches and collects all relevant nodes directly connected with the entry node and associated with the specified hop count and attributes thereof based on a preset heuristic rule to form a strategy knowledge segment; Based on a preset strategy template, strategy know a little identifies segments to be ordered, screened and assembled in a filling way, and according to the types and attribute weights of the knowledge segments, the corresponding positions of the templates are filled in, so that a test strategy document is finally generated.
6. The intelligent analysis method for intelligent network-oriented vehicle test data according to claim 1, wherein S106 comprises the steps of: According to a preset signal matching dictionary, each scene parameter and each judgment index are matched to one or more specific data signal channels corresponding to a target test item data set to form a matching relation table; Loading a predefined data sufficiency checking rule base, and carrying out scanning analysis on the data signal channels corresponding to the target test item data set based on the matching relation table; And comparing the actual characteristic value with a pass/fail threshold value specified in the strategy document, and recording the coincidence state of each index.
7. The intelligent analysis method for intelligent network-oriented vehicle test data according to claim 1, wherein S107 comprises the steps of: Loading an environment monitoring probe set and a data flow checker set according to a target test item identifier defined in a test execution scheme; the environment monitoring probe set comprises a detection program for testing the power supply of the rack, network delay and power supply state of the sensor, and the data flow inspector set comprises a verification program for verifying the data format, frame frequency and numerical physical rationality; The environment monitoring probe set sends a detection request to each subsystem of the test environment according to a fixed period and analyzes the response, and the data flow inspector set performs frame-by-frame or batch-by-batch verification on the test data flow acquired in real time; Generating a normalized abnormal alarm event when the output of either probe or inspector exceeds its predefined normal range threshold; inquiring a preset repair operation library according to an abnormal type code carried by the alarm event, and matching and generating a repair instruction; And the executing mechanism executes actions, and records the abnormal alarm event, the triggered repairing instruction, the execution time stamp and the associated target test item identifier as abnormal events.
8. An intelligent analysis system for intelligent network-oriented vehicle test data, which is characterized in that the system is used for realizing the intelligent analysis method for intelligent network-oriented vehicle test data according to any one of claims 1 to 7, and the system comprises: The receiving association module is used for receiving vehicle test initial data, identifying a test event corresponding to the data, associating the vehicle test initial data according to the test event, adding a test item identifier, and storing the marked vehicle test initial data into the database; The analysis generation module is used for receiving the test requirement information of the user, analyzing the test requirement information and generating a test item option set containing a plurality of candidate test item identifiers; the item selection determining module is used for receiving a selection instruction triggered by the selection of a user from the test item option set and determining a target test item identification selected by the user based on the selection instruction; The searching and aggregating module is used for searching and retrieving the vehicle testing initial data of all the associated identifiers from the database according to the target testing item identifiers, and aggregating to form a target testing item data set; The strategy generation module is used for analyzing the target test item identification and the test requirement information by combining a preset automobile test knowledge base to generate a test strategy aiming at the target test item, wherein the test strategy comprises a test scene, a test step and a judgment index; The scheme generating module is used for analyzing the data sufficiency and the characteristic consistency of the test scene and the judgment index defined in the test strategy based on the target test item data set, and generating a test execution scheme comprising test parameter suggestions and test case priority ordering according to the analysis result; The exception handling module is used for monitoring a test environment and a test data stream in real time in the process of executing the test according to the test execution scheme, executing a preset repairing operation when the exception is monitored, and recording an exception event and an associated target test item identifier thereof to the database; the test case rechecking module is used for inquiring whether the associated unresolved abnormal event record exists in the database according to the target test item identifier associated with the test case when any test case in the test execution scheme fails to execute, if so, judging that the failure is caused by the abnormality, and re-executing the test case after the abnormality is repaired.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the intelligent analysis method for intelligent networked vehicle test data according to any one of claims 1 to 7 when the program is executed.
10. A storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the intelligent analysis method for intelligent networked vehicle test data according to any of claims 1 to 7.

Description

Intelligent analysis method, system, equipment and medium for intelligent networking vehicle test data Technical Field The invention belongs to the technical field of vehicle testing, and particularly relates to an intelligent analysis method, system, equipment and medium for intelligent networking vehicle testing data. Background Along with the propulsion of the automatic driving level, the vehicles need to realize the coordination of perception, decision and control under the conditions of urban congestion, high-speed lane change and extreme weather, so that effective test and verification on the networked vehicles are needed. In the related art, test data are acquired and processed according to time and equipment types, and data generated by different test items are used in a mixed mode. If a certain test item is analyzed, when corresponding data is called, a large amount of irrelevant data is traversed, the time is long, interference data is mixed in the analysis due to data mixing, and the accuracy of an analysis result is affected. The related technology finds out the abnormality in the monitoring of the test environment and the data flow, the repairing mode is to restart the equipment, the resetting system and the like, the abnormality cannot be prompted, the restarting equipment and the resetting system need to rearrange the test process, the time consumption of the abnormality solving is long, the test flow is frequently interrupted, and the continuity of the test is affected. After the test case fails to execute in the related technology, contents such as vehicle functions, test equipment states, environment parameters, data stream transmission and the like need to be checked. Failure caused by possible failure reasons cannot be locked, failures caused by environment or data flow abnormality are easily misjudged as vehicle function defects, invalid investigation workload is increased, and deviation of test conclusion can be caused. Disclosure of Invention The intelligent analysis method for the intelligent network-connected vehicle test data provided by the invention realizes the intellectualization of the test scheme formulation, improves the multiplexing rate and the value of the historical test data, and improves the reliability of test execution and the accuracy of results through active environment monitoring and closed loop failure analysis. The method comprises the following steps: S101, receiving vehicle test initial data, identifying a test event corresponding to the data, associating the vehicle test initial data according to the test event, adding a test item identifier, and storing the marked vehicle test initial data into a database; S102, receiving test requirement information of a user, analyzing the test requirement information, and generating a test item option set containing a plurality of candidate test item identifiers; S103, receiving a selection instruction triggered by the selection of a user from the test item option set, and determining a target test item identifier selected by the user based on the selection instruction; s104, retrieving and retrieving initial vehicle test data of all the associated identifiers from the database according to the target test item identifiers, and aggregating to form a target test item data set; S105, analyzing by combining a preset automobile test knowledge base based on the target test item identification and the test requirement information to generate a test strategy aiming at the target test item, wherein the test strategy comprises a test scene, a test step and a judgment index; S106, based on the target test item data set, analyzing the data sufficiency and the characteristic coincidence of the test scene and the judgment index defined in the test strategy, and generating a test execution scheme comprising test parameter suggestions and test case priority ordering according to the analysis result; S107, in the process of executing the test according to the test execution scheme, monitoring the test environment and the test data stream in real time, executing a preset repair operation when abnormality is monitored, and recording an abnormality event and an associated target test item identifier to the database; S108, when any test case in the test execution scheme fails to be executed, inquiring whether an associated unresolved abnormal event record exists in the database according to a target test item identifier associated with the test case, if so, judging that the failure is caused by an abnormality, and re-executing the test case after the abnormality is repaired. Preferably, S101 includes the steps of: Receiving the original data streams of the vehicle-mounted controller, the vehicle sensor and the test bench, and analyzing and converting the original data streams into standardized data frames with uniform time stamp sequences; The method comprises the steps of configuring an event recognition rule base, inputting standardized d