Intelligent analysis method, system and equipment for intelligent network connection vehicle test data, and medium

Abstract

The invention belongs to the technical field of vehicle testing, and particularly relates to an intelligent analysis method, system and device for intelligent network connection vehicle test data and a medium, initial test data is received, and a test item identifier is printed based on a test event; generating a candidate test item option set according to a user test demand, and aggregating related data to form a target test item; generating a test strategy including test scenes, steps and evaluation indexes in combination with the knowledge base; analyzing data required by the strategy, and outputting an execution scheme; monitoring the environment and the data flow in real time during execution, and automatically repairing and recording when the environment and the data flow are abnormal; and if the test case fails, re-checking the associated exception record, and retrying after repairing if it is confirmed that the test case is caused by the exception. According to the invention, the intelligent and automatic formulation of the test scheme is realized, the reuse rate of historical test data is improved, and the reliability of test execution and the result accuracy are improved through environment monitoring and failure analysis.

Description

Technical Field

[0001] This invention belongs to the field of vehicle testing technology, specifically relating to an intelligent analysis method, system, equipment, and medium for intelligent connected vehicle test data. Background Technology

[0002] As autonomous driving levels advance, vehicles need to achieve coordinated perception, decision-making, and control in situations such as urban congestion, highway lane changes, and extreme weather, which requires effective testing and verification of connected vehicles.

[0003] In related technologies, test data is acquired and processed according to time and device type, and data from different test items are used interchangeably. For example, when analyzing a specific test item, retrieving the corresponding data involves traversing a large amount of irrelevant data, which is time-consuming. Furthermore, the mixing of data can introduce interfering data into the analysis, affecting the accuracy of the analysis results.

[0004] The relevant technologies monitor the test environment and data flow and detect anomalies. The repair methods include restarting the equipment and resetting the system. However, these methods do not effectively identify the anomalies or provide any prompts. Furthermore, restarting the equipment and resetting the system requires reorganizing the test process, which makes anomaly resolution time-consuming, frequently interrupts the test process, and affects the continuity of the test.

[0005] In related technologies, when test cases fail to execute, it is necessary to investigate vehicle functionality, test equipment status, environmental parameters, data stream transmission, and other aspects. Failure to pinpoint the possible causes of failure can easily lead to misjudging failures caused by environmental or data stream anomalies as vehicle functional defects, increasing the workload of ineffective investigation and potentially causing biased test conclusions. Summary of the Invention

[0006] This invention provides an intelligent analysis method for test data of intelligent connected vehicles, which realizes the intelligent formulation of test plans, improves the reuse rate and value of historical test data, and enhances the reliability and accuracy of test execution through proactive environmental monitoring and closed-loop failure analysis.

[0007] The methods include: S101. Receive initial vehicle test data, identify the test events corresponding to the data, associate the initial vehicle test data with test events and add test item identifiers, and store the labeled initial vehicle test data in the database. S102. Receive the user's test requirement information, parse the test requirement information, and generate a test item option set containing several candidate test item identifiers; S103. Receive a selection instruction triggered by the user's selection from the test item option set, and determine the target test item identifier selected by the user based on the selection instruction; S104. Based on the target test item identifier, retrieve and call up the initial vehicle test data of all associated identifiers from the database, aggregate them, and form the target test item dataset; S105. Based on the target test item identifier and the test requirement information, and combined with a pre-set automotive test knowledge base, an analysis is performed to generate a test strategy for the target test item; the test strategy includes test scenarios, test steps, and evaluation indicators. S106. Based on the target test project dataset, perform data sufficiency and feature conformity analysis on the test scenarios and evaluation indicators defined in the test strategy; based on the analysis results, generate a test execution plan that includes test parameter suggestions and test case priority ranking; S107. During the test execution process according to the test execution plan, the test environment and test data flow are monitored in real time. When an anomaly is detected, a preset repair operation is performed, and the abnormal event and its associated target test item identifier are recorded in the database. S108. When any test case in the test execution plan fails, query the database to see if there is an associated unresolved exception event record based on the target test project identifier associated with the test case; if so, determine that the failure was caused by an exception, and re-execute the test case after the exception is fixed.

[0008] Preferably, S101 includes the following steps: It receives raw data streams from the vehicle controller, vehicle sensors, and test bench, and parses and converts the raw data streams into standardized data frames with a unified timestamp sequence. Configure an event recognition rule base, which pre-stores trigger conditions corresponding to various test items; input standardized data frames into the event recognition rule base for frame-by-frame matching and state machine judgment; when the content of the data frame continuously meets specific trigger conditions, the event type label of the test event is determined and output. Establish a matching relationship between event type labels and test item identifiers; Based on the output event type label, query the matching relationship to obtain one or more corresponding test item identifiers; write the obtained test item identifiers, corresponding standardized data frames and their timestamps into the specified data table of the database in the form of associated records.

[0009] Preferably, S102 includes the following steps: The test requirement description text is received from the user through a human-computer interaction interface. The system performs word segmentation and keyword extraction on the test requirement description text, and inputs the extracted keyword set into a pre-built test project identifier database for fuzzy matching and semantic association query. Based on the matching query results, all test item identifiers with a relevance higher than a preset threshold are retrieved from the test item identifier database and sorted in descending order of relevance to generate a candidate test item identifier list as the test item option set.

[0010] Preferably, S104 includes the following steps: Based on the target test item identifier, construct a query statement; the query statement uses the target test item identifier as the query condition and points to the data table in the database that stores the initial test data of the labeled vehicles. Extract all records that meet the conditions from the data table, and encapsulate the vehicle test initial data entity and its metadata in each record into an intermediate data object; The acquired intermediate data objects are timestamped and their data fields are merged to remove duplicate time segments. The data is then integrated according to the time series to generate a dataset for the target test project.

[0011] Preferably, S105 includes the following steps: From the extracted keyword requirements, key scenario elements based on environment, road, and performance are selected and combined with the target test project identifier to generate a query vector. The strategy generates a query vector that includes the main test project type, relevant environmental variables, road type, and performance dimensions. Using the elements in the aforementioned query vector as entry nodes, a graph traversal is initiated in the pre-set automotive test knowledge base. The knowledge base is stored in the form of a graph, where nodes represent test scenario elements, vehicle parts, or evaluation indicators, and edges represent logical, subordinate, or causal relationships between them. The traversal process is based on preset heuristic rules, which retrieve and collect all relevant nodes and their attributes that are directly connected to the entry node and associated within a specified number of hops, forming a strategy knowledge fragment; Based on a pre-defined strategy template, strategy knowledge fragments are sorted, filtered, and assembled using a fill-in-the-blank method. According to the type and attribute weight of the knowledge fragments, they are filled into the corresponding positions in the template, and finally, a test strategy document is generated.

[0012] Preferably, S106 includes the following steps: Parse the test strategy document to extract the defined test scenario parameters and evaluation index thresholds; based on the preset signal matching dictionary, match each scenario parameter and evaluation index to one or more specific data signal channels in the target test project dataset to form a matching relationship table; Load a predefined data sufficiency check rule base, and scan and analyze the corresponding data signal channels in the target test project dataset based on the matching relationship table; For each evaluation metric pointed to in the matching relationship table, its actual feature value is calculated based on the statistical method of the maximum and average values ​​defined in the strategy definition; the actual feature value is compared with the pass / fail threshold specified in the strategy document, and the compliance status of each metric is recorded.

[0013] Preferably, S107 includes the following steps: Based on the target test item identifier defined in the test execution plan, load the environment monitoring probe set and data flow inspector set; The environmental monitoring probe set includes probes for test bench power, network latency, and sensor power status; the data stream checker set includes verification procedures for validating data format, frame rate, and the physical validity of numerical data. Parallel execution of probe detection and inspector analysis, triggering repair instructions: The environmental monitoring probe set sends detection requests to each subsystem of the test environment at fixed intervals and analyzes the responses; the data stream inspector set verifies the real-time acquired test data stream frame by frame or batch by batch. When the output of any probe or inspector exceeds its predefined normal range threshold, a standardized abnormal alarm event is generated; based on the abnormal type code carried by the alarm event, a pre-set repair operation library is queried, and a repair instruction is generated. The generated repair instructions are sent to the corresponding execution mechanism; the execution mechanism performs the action and records the abnormal alarm event, the triggered repair instructions, the execution timestamp, and the associated target test project identifier as abnormal events.

[0014] This invention also provides an intelligent analysis system for test data of intelligent connected vehicles, the system comprising: The receiving and association module is used to receive initial vehicle test data, identify the test events corresponding to the data, associate the initial vehicle test data with test events and add test item identifiers, and store the labeled initial vehicle test data in the database. The parsing and generation module is used to receive the user's test requirement information, parse the test requirement information, and generate a test item option set containing several candidate test item identifiers; The project selection and determination module is used to receive a selection instruction triggered by the user's selection from the test project option set, and determine the target test project identifier selected by the user based on the selection instruction; The retrieval and aggregation module is used to retrieve and call up the initial vehicle test data of all associated identifiers from the database according to the target test project identifier, and aggregate them to form the target test project dataset; The strategy generation module analyzes the target test item identifier and the test requirement information in conjunction with a pre-built automotive test knowledge base to generate a test strategy for the target test item; the test strategy includes test scenarios, test steps, and evaluation indicators. The solution generation module performs data sufficiency and feature conformity analysis on the test scenarios and evaluation indicators defined in the test strategy based on the target test project dataset; and generates a test execution plan that includes test parameter suggestions and test case priority ranking based on the analysis results. The exception handling module is used to monitor the test environment and test data flow in real time during the test execution according to the test execution plan. When an exception is detected, a preset repair operation is performed, and the exception event and its associated target test item identifier are recorded in the database. The test case review module is used to query the database for any associated unresolved exception event records when any test case in the test execution plan fails, based on the target test project identifier associated with the test case; if such records exist, the failure is determined to be caused by an exception, and the test case is re-executed after the exception is fixed.

[0015] According to another embodiment of this application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the intelligent analysis method for intelligent connected vehicle test data.

[0016] According to another embodiment of this application, a storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the intelligent analysis method for intelligent connected vehicle test data.

[0017] As can be seen from the above technical solutions, the present invention has the following advantages: This invention provides an intelligent analysis method for intelligent connected vehicle test data, identifying test events corresponding to the test data and adding a unique test item identifier to each initial data entry. When retrieving data, the target test item identifier is used as the search keyword to locate and retrieve related data. This achieves the binding of test data with projects, improving retrieval efficiency. Fields such as test object model and functional direction are extracted to generate a candidate test item option set sorted by matching degree. Environmental parameters and data flow characteristics are collected at high-frequency cycles. A pre-set anomaly threshold and anomaly type repair strategy matching table is used. When an anomaly occurs, targeted repair operations are triggered, and the recovery status is immediately checked after repair. The anomaly event is then associated with and recorded with the target test item identifier. This improves anomaly resolution efficiency, reduces the frequency of test process interruptions, and ensures continuous and stable testing.

[0018] This invention extracts the target test item identifier associated with test cases, queries the database for unresolved exception events within the corresponding test time period, verifies the correlation between exceptions and failures, and determines whether failures are caused by exceptions, thereby improving the accuracy of test result evaluation. It retrieves basic scenarios, steps, and metrics from the automotive testing knowledge base by searching the target test item identifier, and adapts the basic content by combining it with the test object model and other user requirements. The testing strategy aligns with specific vehicle models and personalized user needs, covering key testing points and improving testing effectiveness. Attached Figure Description

[0019] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 Flowchart of intelligent analysis method for test data of intelligent connected vehicles; Figure 2 A schematic diagram of an intelligent analysis system for test data of intelligent connected vehicles; Figure 3 This is a schematic diagram of an electronic device. Detailed Implementation

[0021] The intelligent analysis method for test data of intelligent connected vehicles involved in this application will be described in detail below. Specific details such as particular system structures and technologies are proposed for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details.

[0022] It should be understood that, when used in this specification, terms include indicating the presence of a described feature, integral, step, operation, element, and / or component, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof. The terms include, encompass, have, and variations thereof mean including but not limited to, unless otherwise specifically emphasized.

[0023] The statements such as "one embodiment" or "some embodiments" described in this application mean that one or more embodiments of this application include the specific features, structures, or characteristics described in that embodiment. Therefore, the statements such as "in one embodiment," "in some embodiments," "in other embodiments," and "in still other embodiments" in this application do not necessarily refer to the same embodiment, but rather mean one or more, but not all, embodiments, unless otherwise specifically emphasized.

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Please see Figure 1 The diagram shows a flowchart of an intelligent analysis method for test data of intelligent connected vehicles in a specific embodiment. The method includes: S101. Receive initial vehicle test data, identify the test events corresponding to the data, associate the initial vehicle test data with test events and add test item identifiers, and store the labeled initial vehicle test data in the database.

[0026] In some embodiments, the initial vehicle test data is received from the cameras, radars, millimeter-wave radars, etc., the on-board controller ECU, VCU, test bench equipment, and roadside perception equipment of the intelligent connected vehicle under test.

[0027] Test event recognition is based on a preset event feature library, which contains data features corresponding to different test items, such as the brake signal triggering identifier and obstacle distance data for automatic emergency braking test, and the steering control signal and lane line deviation data for lane keeping test. The corresponding test event is determined by comparing the matching degree between the received data and the feature fields in the feature library.

[0028] As one implementation of this embodiment, S101 specifically includes the following steps: S1011 receives the raw data stream from the vehicle controller, vehicle sensors and test bench, and parses the raw data stream and converts it into standardized data frames with a unified timestamp sequence.

[0029] In some embodiments, for each set of raw input data, the driver restores its payload to physical values ​​according to the protocol specifications, appends a global timestamp taken from a high-precision synchronous clock source to each physical value data point, and finally packages it to generate a standardized data frame required for internal processing. The standardized data frame has a fixed structure and includes a timestamp, signal ID, physical value, and data quality status bits.

[0030] S1012. Configure the event recognition rule base. The event recognition rule base pre-stores the trigger conditions corresponding to various test items. Input the standardized data frame into the event recognition rule base for frame-by-frame matching and state machine judgment. When the data frame content continuously meets the specific trigger conditions, a test event is determined to have occurred, and the event type label of the test event is output.

[0031] In some embodiments, data-driven event detection is performed based on predefined rules. The event recognition rule base is a configurable set of logical judgments, with each rule associated with a specific test event.

[0032] The rule definition methods in this embodiment include, but are not limited to: determining the threshold of a single signal value, determining the logical combination state of multiple signals, or determining the sequence of a specific signal pattern appearing within a time window.

[0033] The system internally maintains a state machine corresponding to each rule. When a standardized data frame stream is continuously input, the system substitutes the signal values ​​within the frame into each activation rule for judgment, and the state machine transitions according to the judgment result. When the state machine corresponding to a certain rule transitions to the trigger completion state, it signifies that a complete test event defined by that rule has been successfully captured, and the system then generates and outputs an event type label with the event name and trigger time.

[0034] Step S1013: Establish the matching relationship between event type labels and test item identifiers; Based on the event type label output in step S1012, query the matching relationship to obtain one or more corresponding test item identifiers; write the obtained test item identifiers, corresponding standardized data frames and their timestamps into the designated data table of the database in the form of associated records.

[0035] In some embodiments, the matching entries explicitly record correspondences such as matching an emergency braking event to an AEB (Automatic Emergency Braking) test item and a braking performance test item.

[0036] After receiving the event type label, the system uses it as a key to query this table, thereby obtaining one or more test item identifiers with higher semantics and greater stability.

[0037] The system performs a database write operation: The test project identifier, used as a metadata field, along with the original standardized data frame that triggered the event and the event's start and end timestamps, forms a test data record, which is then written to the designed data table. This ensures a connection is established between the original data, the test event to which it belongs, and the final test project to which it belongs.

[0038] S102. Receive the user's test requirement information, parse the test requirement information, and generate a test item option set containing several candidate test item identifiers.

[0039] In some embodiments, a multi-channel user requirement reception system is established, and the requirements are extracted by extracting key fields to clarify the user's actual testing needs. Option sets are generated by sorting by matching degree to complete the transformation of user requirements into candidate test items.

[0040] As one implementation of this embodiment, S102 specifically includes the following steps: S1021. Receive the test requirement description text input by the user through the human-computer interaction interface.

[0041] In some embodiments, users express their testing intent by inputting information such as needing to test the automatic following function on rainy urban roads or predefined test conditions such as scenario: urban road; weather: rainy; function: adaptive cruise control. This input is captured by the interface backend service and encapsulated as a test requirement description text.

[0042] S1022, Call the word segmentation and keyword extraction operation on the test requirement description text, and input the extracted keyword set into the pre-built test project identifier database for fuzzy matching and semantic association query.

[0043] In some embodiments, the user-input description of the requirement is segmented into individual word or phrase units. Key keywords are then selected from these units based on part-of-speech and domain importance.

[0044] For example, keywords like "rainy day," "city road," and "autopilot" are fed into a database of test item identifiers. This database stores all valid test item identifiers and associates multiple feature keywords and synonyms with each identifier. The query process is a retrieval process based on text similarity calculations and semantic network expansion, aiming to identify test item identifiers that are highly relevant to the input keywords in literal or semantic terms.

[0045] S1023. Based on the matching query results, retrieve all test item identifiers with a correlation higher than a preset threshold from the test item identifier database, sort them in descending order of correlation, and generate a candidate test item identifier list as the test item option set.

[0046] In some embodiments, the system sets a minimum acceptable relevance threshold to filter out identifiers with low scores that may be irrelevant. For all identifiers that pass the threshold, the system sorts them in descending order of their relevance scores to form an ordered list. This list is the final set of test item options presented to the user. The ordered arrangement means that the options at the top are the candidate items that the system considers to best match the user's needs.

[0047] S103. Receive a selection instruction triggered by the user's selection from the test item option set, and determine the target test item identifier selected by the user based on the selection instruction.

[0048] In some embodiments, user-triggered selection commands are monitored in real time, and invalid operations are filtered out through verification. Encoded information from the commands is extracted and converted into a target test item identifier, thus accurately pinpointing the user's requirements to the specific test item.

[0049] S104. Based on the target test project identifier, retrieve and collect all vehicle test initial data associated with the identifier from the database, aggregate them, and form a target test project dataset.

[0050] In some embodiments, the target test item identifier is used as the primary retrieval method to locate related data in the database. Batch transmission and breakpoint resumption ensure the integrity and efficiency of data retrieval. Data aggregation is achieved by classifying by type, sorting by time, and deduplicating data to form the target dataset.

[0051] S105. Based on the target test item identifier and the test requirement information, and combined with a pre-set automotive test knowledge base, an analysis is performed to generate a test strategy for the target test item; the test strategy includes test scenarios, test steps, and evaluation indicators.

[0052] In some embodiments, basic test content for the target test project is obtained based on a distributed automotive testing knowledge base. This basic content is then adapted and optimized by incorporating personalized parameters reflecting user testing needs, and integrated to form a testing strategy. The testing strategy clearly defines the requirements for each stage of the testing process, ensuring its standardization.

[0053] S106: Based on the target test project dataset, perform data sufficiency and feature conformity analysis on the test scenarios and evaluation indicators defined in the test strategy; based on the analysis results, generate a test execution plan that includes test parameter suggestions and test case priority ranking.

[0054] In some embodiments, the completeness and sufficiency of the target dataset for the test scenario are assessed, and the conformity of data features with evaluation metrics is verified. Based on the analysis conclusions, test parameters are optimized, test cases are prioritized, a test plan is formed, and the test strategy is transformed into test execution.

[0055] S107: During the test execution according to the test execution plan, the test environment and test data flow are monitored in real time. When an anomaly is detected, a preset repair operation is performed, and the abnormal event and its associated target test item identifier are recorded in the database.

[0056] In some embodiments, real-time monitoring of the test environment and data stream is achieved through sensor and data frame capture, and an anomaly threshold is set as the judgment criterion. When an anomaly occurs, targeted repair operations are triggered, and the effect is verified after repair. Anomaly events are associated with and stored with target identifiers.

[0057] S108: When any test case in the test execution plan fails, query the database to see if there is an associated unresolved exception event record based on the target test project identifier associated with the test case; if so, determine that the failure was caused by an exception, and re-execute the test case after the exception is fixed.

[0058] In some embodiments, during the execution of test cases, the success, failure, and execution status of each test case are recorded in real time. When a test case fails to execute, the target test project identifier associated with the test case is extracted, and this identifier is used as a search keyword to query the abnormal event records in the database.

[0059] In this embodiment, the query scope is limited to unresolved exception events within the execution time period of the test case, that is, exceptions that were not successfully repaired or were not repaired completely.

[0060] The query results are categorized into two cases: those with unresolved exceptions and those without. If an exception exists, the correlation between the exception type and the reason for the test case execution failure is further verified. If the correlation is confirmed, the failure is determined to be caused by an exception. If no exception exists, the failure information is recorded, and the cause is manually investigated.

[0061] If the failure is determined to be due to an anomaly, the test case will be re-executed after the anomaly is fixed and the environment is verified to be back to normal. During the re-execution, the corresponding parameters will be monitored more closely to ensure that the test results are valid.

[0062] By identifying the correlation between anomalies and failures through related queries, the time spent troubleshooting failures is reduced, and the accuracy of test result evaluation is improved. Re-execution after repair ensures that test cases are effectively validated, without interrupting the overall test process due to anomalies, thus guaranteeing test integrity.

[0063] In one embodiment of the present invention, based on step S104, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S104 includes the following steps: S1041. Construct a query statement based on the target test item identifier; the query statement uses the target test item identifier as the query condition and points to the data table in the database that stores the initial test data of the labeled vehicles.

[0064] In some embodiments, after receiving the target test item identifier confirmed via S103, a predefined query template is used to fill in the target test item identifier as a parameter value in the query condition position of the corresponding test item identifier field in the template, generating a query statement. This statement explicitly specifies which database and which data table, i.e., the table storing the results of step S101, will select all data rows whose test item identifier field values ​​completely match the input parameters.

[0065] S1042. Extract all records that meet the conditions from the data table, and encapsulate the vehicle test initial data entity and its metadata in each record into an intermediate data object.

[0066] In some embodiments, the standardized query statement generated in step S1041 is sent to the database server. The database server performs query optimization and data retrieval internally. After receiving the result set consisting of multiple rows of records, the data access layer does not directly adopt the database row format. Instead, it iterates through all result rows, extracting key fields such as timestamps, original data frame content, and data quality markers, and encapsulates them into an intermediate data object with an internal structure that is easier to process in the application's memory. All retrieved records are ultimately transformed into a list of intermediate data objects.

[0067] S1043. Perform timestamp alignment and data field merging operations on the acquired multiple intermediate data objects, remove duplicate time segment data, and integrate them according to the time series to generate a data set as the target test project dataset.

[0068] In some embodiments, data fragments are cleaned, aligned, and merged. The system traverses a list of intermediate data objects, each containing a high-precision timestamp.

[0069] These objects are sorted based on their timestamps. The timestamps of adjacent objects are checked for overlap or gaps. For overlapping data, redundant parts are removed based on data source priority or quality markers. Data with time gaps is retained.

[0070] The signal data carried by each object from different buses or sensors are reorganized according to their signal identifiers and merged into a data structure framework. The horizontal dimension of this framework is time series, and the vertical dimension is each signal channel, ultimately generating a data set.

[0071] In one embodiment of the present invention, based on step S105, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S105 includes the following steps: S1051. Select key scenario elements based on environment, road, and performance from the extracted demand keywords, and combine them with the target test project identifier to generate a query vector.

[0072] In some embodiments, specific requirement keywords such as low temperature, high-speed loop, and response time are extracted from the user's original test requirement information text.

[0073] S1052. Using the elements in the aforementioned query vector as entry nodes, start graph traversal in the pre-set automotive test knowledge base; the knowledge base is stored in the form of a graph, where nodes represent test scenario elements, vehicle parts or evaluation indicators, and edges represent logical, subordinate or causal relationships between them. The traversal process is based on pre-defined heuristic rules, which retrieve and collect all relevant nodes and their attributes that are directly connected to the entry node and associated with it within a specified number of hops, forming a strategy knowledge fragment.

[0074] In some embodiments, knowledge mining is performed using the relational query capabilities of graph databases. The automotive testing knowledge base is constructed as a directed graph. One node in the graph represents an icy road surface, with attributes including the range of friction coefficients. Another node represents braking distance, with attributes including national standard thresholds.

[0075] Nodes are connected by edges of types such as "belongs to," "influences," and "prerequisites." After receiving a query vector, the system finds the group of entry nodes in the graph that best matches the values ​​of each field of the vector.

[0076] Initiate a traversal algorithm based on breadth-first search and rule-guided search.

[0077] For example, starting from the high-speed loop node, find all test action nodes that can be applied to this scenario, and further find the evaluation index nodes corresponding to these actions.

[0078] The traversal process collects all nodes along the way that satisfy the rules, along with their attributes and relationships, forming a fragment of strategy knowledge.

[0079] S1053. Based on a pre-set strategy template, strategy knowledge fragments are sorted, filtered, and assembled in a fill-in-the-blank manner; the template defines the test scenario, test steps, and evaluation indicators; according to the type and attribute weight of the knowledge fragments, they are filled into the corresponding positions in the template to finally generate a test strategy document.

[0080] In some embodiments, a pre-defined test strategy document template is provided, clearly indicating the locations of test scenario descriptions, preconditions, execution step lists, data recording requirements, evaluation metrics, and thresholds. The set of knowledge fragments collected in step S1052 is input into a strategy assembly engine. This engine embeds business rules, such as: filling environment variable node attributes into precondition slots, filling test action node attributes into the execution step list after sorting them according to execution logic, and filling evaluation metric node attributes and their threshold attributes into the evaluation metric table.

[0081] The engine formats the fragments according to their type and rules, fills them into the corresponding slots, and finally instantiates a strategy document that can be directly reviewed and executed by test engineers.

[0082] In one embodiment of the present invention, based on step S106, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S106 includes the following steps: S1061. Parse the test strategy document and extract the defined test scenario parameters and evaluation index thresholds; according to the preset signal matching dictionary, match each scenario parameter and evaluation index to one or more specific data signal channels in the target test project dataset to form a matching relationship table.

[0083] In some embodiments, the signal matching dictionary is a pre-defined standardized mapping table containing the correspondence between commonly used test strategy terms and dataset signal channels, and supports dynamic updates based on the test item type. The generated test strategy document is read. The marked fields in the document are identified, such as the test scenario: vehicle speed parameters maintained between 80-120 km / h, and the evaluation index: longitudinal acceleration index with an absolute longitudinal acceleration value less than 2.5 m / s² and its threshold of 2.5.

[0084] Query the pre-configured signal matching dictionary. This dictionary records, for example, vehicle speed corresponding to the VehicleSpeed ​​signal in the dataset, and longitudinal acceleration corresponding to the LongitudinalAccel signal.

[0085] Based on this matching relationship, a relationship table is generated, recording that strategy element A (such as speed range requirements) needs to check the X signal channel in the dataset, while strategy element B (such as acceleration threshold) needs to analyze the Y signal channel in the dataset.

[0086] S1062. Load the predefined data sufficiency check rule base, and scan and analyze the corresponding data signal channels in the target test project dataset based on the matching relationship table; the check rules include the time coverage of effective data in each channel, the number of occurrences and duration integrity of key scene segments, and the statistical proportion of missing or invalid data flags.

[0087] In some embodiments, a rule base is built-in, containing quantitative criteria for determining whether the data is sufficient.

[0088] For example, one rule is that for key signals, the effective data coverage within the target scene time period should be higher than 95%. Another rule is that sharp turn scene segments should appear at least 5 times in the entire dataset, with each segment lasting more than 2 seconds.

[0089] Based on the matching table generated by S1061, checks are performed one by one according to the rules. All check results, including coverage percentage, occurrence frequency, average duration, missing rate, etc., are calculated and compared with the minimum requirements preset by the rules.

[0090] S1063. For each evaluation indicator pointed to in the matching relationship table, calculate its actual feature value based on the statistical method of the maximum value and average value defined in the strategy definition; compare the actual feature value with the pass / fail threshold specified in the strategy document, and record the compliance status of each indicator.

[0091] In some embodiments, the system calculates the actual feature values ​​of relevant data segments according to the statistical methods specified for the indicators in the strategy document. It then reads the threshold values ​​set for the indicators in the strategy document. A logical comparison is performed, which may involve comparing the calculated actual feature values ​​with the threshold values. This determines whether the indicator meets or does not meet the threshold. The compliance status of all indicators is summarized, and those at the threshold edge are marked as boundary compliance for manual review.

[0092] In one embodiment of the present invention, based on step S107, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S107 specifically includes the following steps: S1071. Load the environment monitoring probe set and data flow inspector set according to the target test item identifier defined in the test execution plan; The environmental monitoring probe set includes probes for test bench power, network latency, and sensor power status; the data stream checker set includes verification procedures for data format, frame rate, and numerical physical validity.

[0093] In some embodiments, if the test is GPS-based, a GPS signal quality probe and an inertial navigation data synchronization checker are loaded and activated. If the test is bench power supply testing, a DC power supply ripple probe is loaded.

[0094] These probes and inspectors are pre-compiled small program modules configured to listen to specific hardware interfaces or data buses. Once started, they run in the background as services or daemons, thus reducing the resource consumption of the test host and improving the efficiency and effectiveness of monitoring.

[0095] S1072. Parallel execution of probe detection and inspector analysis, triggering repair instructions: The environmental monitoring probe set sends detection requests to each subsystem of the test environment at fixed intervals and analyzes the responses. The data stream inspector set verifies the real-time acquired test data stream frame by frame or batch by batch. When the output of any probe or inspector exceeds its predefined normal range threshold, a standardized abnormal alarm event is generated; based on the abnormal type code carried by the alarm event, the preset repair operation library is queried, and a repair instruction is generated.

[0096] In some embodiments, the environmental probe periodically executes its probing logic, such as sending a ping packet to the network switch and calculating the round-trip time, or reading the voltage readings of the power management module.

[0097] The data inspector continuously monitors the data stream, applying its verification rules to each incoming frame. For example, it checks if the CAN frame ID is in the expected list and if the signal value is within the possible physical range. All these probes and inspectors compare their outputs in real time with a pre-configured normal threshold table for the project.

[0098] If a certain indicator exceeds the limit, such as network latency greater than 50 milliseconds, or an impossible vehicle speed of 500 km / h is received, the corresponding detection program immediately generates an anomaly alert. The anomaly type code of this alert is used as a keyword to query a pre-prepared repair operation library.

[0099] The repair operation library is a matching table that stores exception codes. When an exception occurs, an entry such as "restart network switch port" is executed, thereby generating a repair command.

[0100] S1073. Send the generated repair instructions to the corresponding execution mechanism; the execution mechanism performs the action and records the abnormal alarm event, the triggered repair instructions, the execution timestamp, and the associated target test item identifier as abnormal events.

[0101] In some embodiments, a command distribution service sends the generated repair command to the corresponding target. This target is a control point of a programmable logic controller (PLC) used to operate a relay or reset switch. After the command is sent, a feedback timeout period is waited to confirm whether the command has been received and executed. Regardless of whether the repair action ultimately successfully resolves the problem, a crucial operation is performed: the entire event is logged.

[0102] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0103] The following are embodiments of the intelligent analysis system for intelligent connected vehicle test data provided in this disclosure. This system and the intelligent analysis method for intelligent connected vehicle test data in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the intelligent analysis system for intelligent connected vehicle test data, please refer to the embodiments of the intelligent analysis method for intelligent connected vehicle test data described above.

[0104] like Figure 2 As shown, the system includes: The receiving and association module 201 is used to receive vehicle test initial data, identify the test events corresponding to the data, associate the vehicle test initial data with test events and add test item identifiers, and store the labeled vehicle test initial data in the database. The parsing and generation module 202 is used to receive the user's test requirement information, parse the test requirement information, and generate a test item option set containing several candidate test item identifiers; The project selection and determination module 203 is used to receive a selection instruction triggered by the user's selection from the test project option set, and determine the target test project identifier selected by the user based on the selection instruction; The retrieval and aggregation module 204 is used to retrieve and call up the initial vehicle test data of all associated identifiers from the database according to the target test project identifier, and aggregate them to form the target test project dataset; The strategy generation module 205 analyzes the target test item identifier and the test requirement information in conjunction with a pre-set automotive test knowledge base to generate a test strategy for the target test item; the test strategy includes test scenarios, test steps, and evaluation indicators. The solution generation module 206 performs data sufficiency and feature conformity analysis on the test scenarios and evaluation indicators defined in the test strategy based on the target test project dataset; and generates a test execution plan that includes test parameter suggestions and test case priority ranking based on the analysis results. The exception handling module 207 is used to monitor the test environment and test data flow in real time during the test execution process according to the test execution plan, execute preset repair operations when an exception is detected, and record the exception event and its associated target test item identifier to the database. The test case review module 208 is used to query the database for any associated unresolved exception event records when any test case in the test execution plan fails, based on the target test project identifier associated with the test case; if such records exist, the failure is determined to be caused by an exception, and the test case is re-executed after the exception is fixed.

[0105] like Figure 3 As shown, this application also provides an electronic device, including a display module 103, a memory 102, a processor 101, a communication module 104, and a computer program stored in the memory and executable on the processor 101. When the processor 101 executes the program, it implements the steps of an intelligent analysis method for intelligent connected vehicle test data.

[0106] In embodiments of the present invention, electronic devices include, but are not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments described and / or claimed herein.

[0107] In this embodiment, processor 101 may be implemented using at least one of an application-specific integrated circuit, a programmable logic device, a field-programmable gate array, a processor, a controller, a microcontroller, a microprocessor, or an electronic unit designed to perform the functions described herein. In some cases, such an implementation may be implemented within a controller. For software implementation, implementations such as processes or functions may be implemented with separate software modules that allow the performance of at least one function or operation. Software code may be implemented by a software application (or program) written in any suitable programming language, and the software code may be stored in memory and executed by the controller.

[0108] The display module 103 is used to display information input by the user or information provided to the user. The display module 103 may include a display panel, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like.

[0109] The memory 102 can be used to store software programs and various data. The memory 102 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0110] The communication module 104 transmits radio signals to and / or receives radio signals from at least one of a base station, an external terminal, and a server. Such radio signals may include voice call signals, video call signals, or various types of data sent and / or received according to text and / or multimedia messages.

[0111] The present invention also provides a storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of the intelligent analysis method for test data of intelligent connected vehicles.

[0112] The storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0113] The storage medium stores a program product capable of implementing the methods described above in this specification. In some possible implementations, various aspects of this disclosure can also be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the exemplary methods section of this specification according to various exemplary embodiments of this disclosure.

[0114] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent analysis method for test data of intelligent connected vehicles, characterized in that, The methods include: S101. Receive initial vehicle test data, identify the test events corresponding to the data, associate the initial vehicle test data with test events and add test item identifiers, and store the labeled initial vehicle test data in the database. S102. Receive the user's test requirement information, parse the test requirement information, and generate a test item option set containing several candidate test item identifiers; S103. Receive a selection instruction triggered by the user's selection from the test item option set, and determine the target test item identifier selected by the user based on the selection instruction; S104. Based on the target test item identifier, retrieve and call up the initial vehicle test data of all associated identifiers from the database, aggregate them, and form the target test item dataset; S105. Based on the target test item identifier and the test requirement information, and combined with the pre-set automotive test knowledge base, an analysis is performed to generate a test strategy for the target test item. The testing strategy includes testing scenarios, testing steps, and evaluation metrics; S106. Based on the target test project dataset, perform data sufficiency and feature conformity analysis on the test scenarios and evaluation indicators defined in the test strategy; based on the analysis results, generate a test execution plan that includes test parameter suggestions and test case priority ranking; S107. During the test execution process according to the test execution plan, the test environment and test data flow are monitored in real time. When an anomaly is detected, a preset repair operation is performed, and the abnormal event and its associated target test item identifier are recorded in the database. S108. When any test case in the test execution plan fails, query the database to see if there is an associated unresolved exception event record based on the target test project identifier associated with the test case; if so, determine that the failure was caused by an exception, and re-execute the test case after the exception is fixed.

2. The intelligent analysis method for test data of intelligent connected vehicles according to claim 1, characterized in that, S101 includes the following steps: It receives raw data streams from the vehicle controller, vehicle sensors, and test bench, and parses and converts the raw data streams into standardized data frames with a unified timestamp sequence. Configure an event recognition rule base, which pre-stores trigger conditions corresponding to various test items; The standardized data frames are input into the event recognition rule base for frame-by-frame matching and state machine judgment. When the content of the data frame continuously meets the specific triggering conditions, the event type label of the test event is determined and output. Establish a matching relationship between event type labels and test item identifiers; Based on the output event type label, query the matching relationship to obtain one or more corresponding test item identifiers; write the obtained test item identifiers, corresponding standardized data frames and their timestamps into the specified data table of the database in the form of associated records.

3. The intelligent analysis method for test data of intelligent connected vehicles according to claim 1, characterized in that, S102 includes the following steps: The test requirement description text is received from the user through a human-computer interaction interface. The system performs word segmentation and keyword extraction on the test requirement description text, and inputs the extracted keyword set into a pre-built test project identifier database for fuzzy matching and semantic association query. Based on the matching query results, all test item identifiers with a relevance higher than a preset threshold are retrieved from the test item identifier database and sorted in descending order of relevance to generate a candidate test item identifier list as the test item option set.

4. The intelligent analysis method for test data of intelligent connected vehicles according to claim 1, characterized in that, S104 includes the following steps: Based on the target test item identifier, construct a query statement; the query statement uses the target test item identifier as the query condition and points to the data table in the database that stores the initial test data of the labeled vehicles. Extract all records that meet the conditions from the data table, and encapsulate the vehicle test initial data entity and its metadata in each record into an intermediate data object; The acquired intermediate data objects are timestamped and their data fields are merged to remove duplicate time segments. The data is then integrated according to the time series to generate a dataset for the target test project.

5. The intelligent analysis method for test data of intelligent connected vehicles according to claim 1, characterized in that, S105 includes the following steps: The key scenario elements based on environment, road, and performance are selected from the extracted demand keywords and combined with the target test project identifier to generate a query vector; Using the elements in the aforementioned query vector as entry nodes, a graph traversal is initiated in the pre-set automotive test knowledge base. The knowledge base is stored in the form of a graph, where nodes represent test scenario elements, vehicle parts, or evaluation indicators, and edges represent logical, subordinate, or causal relationships between them. The traversal process is based on preset heuristic rules, which retrieve and collect all relevant nodes and their attributes that are directly connected to the entry node and associated within a specified number of hops, forming a strategy knowledge fragment; Based on a pre-defined strategy template, strategy knowledge fragments are sorted, filtered, and assembled using a fill-in-the-blank method. According to the type and attribute weight of the knowledge fragments, they are filled into the corresponding positions in the template, and finally, a test strategy document is generated.

6. The intelligent analysis method for test data of intelligent connected vehicles according to claim 1, characterized in that, S106 includes the following steps: Parse the test strategy document to extract the defined test scenario parameters and evaluation index thresholds; based on the preset signal matching dictionary, match each scenario parameter and evaluation index to one or more specific data signal channels in the target test project dataset to form a matching relationship table; Load a predefined data sufficiency check rule base, and scan and analyze the corresponding data signal channels in the target test project dataset based on the matching relationship table; For each evaluation index pointed to in the matching relationship table, its actual feature value is calculated based on the statistical method of the maximum and average values ​​defined by the strategy. The actual feature values ​​are compared with the pass / fail thresholds specified in the strategy document, and the compliance status of each indicator is recorded.

7. The intelligent analysis method for test data of intelligent connected vehicles according to claim 1, characterized in that, S107 includes the following steps: Based on the target test item identifier defined in the test execution plan, load the environment monitoring probe set and data flow inspector set; The environmental monitoring probe set includes probe programs for test bench power, network latency, and sensor power status; The data stream inspector set contains verification procedures for validating data format, frame rate, and the physical validity of numerical values; Parallel execution of probe detection and inspector analysis, triggering repair instructions: The environmental monitoring probe set sends detection requests to each subsystem of the test environment at fixed intervals and analyzes the responses; the data stream inspector set verifies the real-time acquired test data stream frame by frame or batch by batch. When the output of any probe or inspector exceeds its predefined normal range threshold, a standardized abnormal alarm event is generated. Based on the exception type code carried by the alarm event, query the preset repair operation library, match and generate repair instructions; The generated repair instructions are sent to the corresponding execution mechanism; the execution mechanism performs the action and records the abnormal alarm event, the triggered repair instructions, the execution timestamp, and the associated target test project identifier as abnormal events.

8. An intelligent analysis system for test data of intelligent connected vehicles, characterized in that, The system is used to implement the intelligent analysis method for test data of intelligent connected vehicles as described in any one of claims 1 to 7; the system includes: The receiving and association module is used to receive initial vehicle test data, identify the test events corresponding to the data, associate the initial vehicle test data with test events and add test item identifiers, and store the labeled initial vehicle test data in the database. The parsing and generation module is used to receive the user's test requirement information, parse the test requirement information, and generate a test item option set containing several candidate test item identifiers; The project selection and determination module is used to receive a selection instruction triggered by the user's selection from the test project option set, and determine the target test project identifier selected by the user based on the selection instruction; The retrieval and aggregation module is used to retrieve and call up the initial vehicle test data of all associated identifiers from the database according to the target test project identifier, and aggregate them to form the target test project dataset; The strategy generation module analyzes the target test item identifier and the test requirement information in conjunction with a pre-built automotive test knowledge base to generate a test strategy for the target test item; the test strategy includes test scenarios, test steps, and evaluation indicators. The solution generation module performs data sufficiency and feature conformity analysis on the test scenarios and evaluation indicators defined in the test strategy based on the target test project dataset; and generates a test execution plan that includes test parameter suggestions and test case priority ranking based on the analysis results. The exception handling module is used to monitor the test environment and test data flow in real time during the test execution according to the test execution plan. When an exception is detected, a preset repair operation is performed, and the exception event and its associated target test item identifier are recorded in the database. The test case review module is used to query the database for related unresolved exception event records when any test case in the test execution plan fails, based on the target test project identifier associated with the test case; if such records exist, the failure is determined to be caused by an exception, and the test case is re-executed after the exception is fixed.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the intelligent analysis method for test data of intelligent connected vehicles as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent analysis method for test data of intelligent connected vehicles as described in any one of claims 1 to 7.

Images