Vehicle test data processing method, device and storage medium

Abstract

The application provides a vehicle test data processing method and device and a storage medium. It relates to the technical field of data processing. The method comprises the following steps: obtaining multiple test scenes of a vehicle in a driving process, scene complexity and test duration corresponding to each test scene; determining a cognitive load parameter according to the scene complexity and the test duration, the cognitive load parameter being used to indicate the cognitive stress degree of a target user in the test process; determining a frequency threshold and a load threshold according to the test duration, the frequency threshold being used to restrict the appearance frequency of a complex scene, and the load threshold being used to restrict the cognitive load parameter; and determining a cognitive state of the target user according to the cognitive load parameter, the frequency threshold and the load threshold, the cognitive state being a normal state or a load state. The accuracy of the vehicle test data processing is improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, device and storage medium for processing vehicle test data. Background Technology

[0002] In intelligent driving scenarios, road testing is a core step in verifying the performance of autonomous vehicles and identifying potential defects. The completeness and accuracy of test problem records directly determine the quality of iterative optimization of intelligent driving systems and driving safety.

[0003] In related technologies, a reminder method based on fixed duration rules is usually used to assist in testing. That is, by pre-setting a fixed test duration, a rest reminder is pushed uniformly after the tester completes a continuous fixed duration of testing to assist in the testing work.

[0004] However, in the above process, the cognitive load on testers varies significantly depending on the test scenario. Relying solely on fixed-duration reminders can easily lead to warning judgments deviating from the actual working state, resulting in low accuracy in processing vehicle test data. Summary of the Invention

[0005] This application provides a method, device, and storage medium for processing vehicle test data, in order to solve the problem of low accuracy in processing vehicle test data.

[0006] Firstly, this application provides a method for processing vehicle test data, including:

[0007] Obtain multiple test scenarios during vehicle operation, along with the scenario complexity and test duration for each scenario.

[0008] Based on the complexity of the scenario and the duration of the test, the cognitive load parameter is determined. The cognitive load parameter is used to indicate the level of cognitive stress experienced by the target users during the test.

[0009] Based on the test duration, determine the frequency threshold and the load threshold. The frequency threshold is used to constrain the frequency of occurrence of complex scenarios, and the load threshold is used to constrain cognitive load parameters.

[0010] Based on cognitive load parameters, frequency thresholds, and load thresholds, the cognitive state of the target user is determined, which is either a normal state or a load state.

[0011] In one possible implementation, cognitive load parameters are determined based on scenario complexity and test duration, including:

[0012] Determine the average complexity based on the complexity of the scenario;

[0013] Cognitive load parameters are determined based on average complexity and test duration.

[0014] In one possible implementation, the average complexity is determined based on the scenario complexity, including:

[0015] Get the first quantity corresponding to the test scenario;

[0016] The average complexity is determined based on the scenario complexity and the number of elements in each test scenario.

[0017] In one possible implementation, the average complexity is determined based on the scenario complexity and the first quantity of each test scenario, including:

[0018] Determine the perturbation weights for each test scenario. The perturbation weights are used to indicate the degree of influence of the test scenario on the cognitive state.

[0019] The average complexity is determined based on the scenario complexity, multiple perturbation weights, and the first quantity.

[0020] In one possible implementation, cognitive load parameters are determined based on average complexity and test duration, including:

[0021] Based on the average complexity and test duration, a baseline cognitive load parameter is determined, which is the cognitive consumption rate of the target user during the test.

[0022] The baseline cognitive load parameters are corrected to obtain the cognitive load parameters.

[0023] In one possible implementation, the baseline cognitive load parameters are modified to obtain cognitive load parameters, including:

[0024] Determine the second quantity corresponding to test scenarios with the same complexity;

[0025] Based on the second quantity, the adjustment parameters are determined, and the baseline cognitive load parameters are corrected according to the adjustment parameters to obtain the cognitive load parameters.

[0026] In one possible implementation, determining the load threshold based on the test duration includes:

[0027] When the test duration is less than or equal to the preset duration, the load threshold is determined as the first parameter;

[0028] When the test duration exceeds the preset duration, the load threshold is determined based on the first parameter and the test duration.

[0029] In one possible implementation, determining the cognitive state of the target user includes:

[0030] Determine the third quantity corresponding to complex test scenarios;

[0031] When the third quantity is greater than or equal to the frequency threshold and the cognitive load parameter is greater than the load threshold, the cognitive state is determined to be a load state.

[0032] In one possible implementation, determining the cognitive state of the target user includes:

[0033] When the third quantity is less than the frequency threshold, or the cognitive load parameter is less than or equal to the load threshold, the cognitive state is determined to be normal.

[0034] Secondly, this application provides a vehicle test data processing apparatus, wherein:

[0035] The acquisition module is used to acquire multiple test scenarios during the vehicle's driving process, the scenario complexity and test duration of each test scenario;

[0036] The first determination module is used to determine the cognitive load parameter based on the scenario complexity and test duration. The cognitive load parameter is used to indicate the level of cognitive stress experienced by the target user during the test.

[0037] The second determining module is used to determine the frequency threshold and the load threshold based on the test duration. The frequency threshold is used to constrain the frequency of occurrence of complex scenarios, and the load threshold is used to constrain cognitive load parameters.

[0038] The processing module is used to determine the cognitive state of the target user, which is either a normal state or a load state, based on cognitive load parameters, frequency threshold, and load threshold.

[0039] Thirdly, embodiments of this application provide a vehicle test data processing apparatus, comprising: at least one processor and a memory; the memory storing computer execution instructions; the at least one processor executing the computer execution instructions stored in the memory, causing the at least one processor to perform the vehicle test data processing method as described in the first aspect and any one of the first aspects.

[0040] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the vehicle test data processing method described in the first aspect and any one of the first aspects.

[0041] The vehicle test data processing method, equipment, and storage medium provided in this application acquire multiple test scenarios during vehicle operation, along with the scenario complexity and test duration for each scenario. Based on the scenario complexity and test duration, cognitive load parameters are determined to indicate the cognitive stress level of the target user. Frequency thresholds and load thresholds are determined based on the test duration. These cognitive load parameters, frequency thresholds, and load thresholds are then combined to determine the target user's cognitive state. This method, combined with the scenario characteristics of intelligent driving testing, enables real-time and dynamic monitoring of the tester's cognitive state during testing. It effectively avoids warning deviations caused by differences in scenario complexity and dynamic changes in test conditions, and resolves omissions due to cognitive overload. This improves both the accuracy of vehicle test data processing and the reliability of intelligent driving testing. Attached Figure Description

[0042] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0043] Figure 1 This is a schematic diagram illustrating an application scenario provided in the embodiments of this application;

[0044] Figure 2 A flowchart illustrating a method for processing vehicle test data provided in an embodiment of this application;

[0045] Figure 3 A schematic diagram illustrating the process of determining cognitive load parameters provided in an embodiment of this application;

[0046] Figure 4 A schematic diagram illustrating the process of determining the frequency threshold and load threshold provided in the embodiments of this application;

[0047] Figure 5 A schematic diagram illustrating the process of determining cognitive state as provided in an embodiment of this application;

[0048] Figure 6 A flowchart illustrating another method for processing vehicle test data provided in this application embodiment;

[0049] Figure 7 A flowchart illustrating another method for processing vehicle test data provided in this application embodiment;

[0050] Figure 8 A schematic diagram of a vehicle test data processing device provided in an embodiment of this application;

[0051] Figure 9 This is a schematic diagram of the structure of the vehicle test data processing device provided in the embodiments of this application.

[0052] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0053] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0054] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0055] In the research and development and verification process of intelligent driving systems, road testing is a core step in evaluating system performance and identifying potential problems. Testers need to observe, judge, and accurately record abnormal scenarios or system behavior problems encountered by autonomous vehicles in complex dynamic environments in real time (e.g., covering complex conditions such as unexpected pedestrian appearances, vehicles cutting in front, and basic straight driving). The completeness of these records is directly related to the safety and reliability of the product.

[0056] In related technologies, the main approach to recording test issues is a fixed-duration reminder scheme. This scheme sets a uniform test duration threshold, and a break reminder pops up after the tester has continuously tested for the preset duration. While this scheme is logically simple and easy to implement, it relies solely on the cumulative time for warning judgment and cannot adaptively adjust to dynamic changes in scenario complexity and cognitive load. This results in a significant discrepancy between the warning timing and the actual cognitive state, leading to an uncontrollable problem omission rate.

[0057] Furthermore, during testing, passive recording tools provide testers with a manually triggered problem recording interface. When a problem is observed, the tester must manually click, select the problem type, and enter a description. This solution is essentially a passive data entry terminal, lacking the ability to actively monitor and intervene in the tester's cognitive state during the recording process, and thus unable to provide effective early warnings before cognitive overload occurs.

[0058] In summary, existing technologies face dual limitations in recording test issues: passive recording and fixed alerts. Fixed-duration alerts, due to their single-time-dimensional, fixed-threshold determination, lack comprehensive consideration of multi-dimensional factors such as scenario complexity and cognitive load, resulting in insufficient alert accuracy. Passive recording tools, lacking a cognitive state monitoring mechanism, cannot proactively intervene before cognitive resources are exhausted, further exacerbating the risk of missed issues.

[0059] In addition, existing solutions mostly focus on real-time monitoring of the driver's status, mainly based on monitoring methods such as physiological signal detection. They are only applicable to vehicle driving safety management scenarios, and their evaluation results lack quantifiable and measurable correlation with the core quality indicator of problem omission rate, making it difficult to implement in real test engineering environments.

[0060] Therefore, this application provides a method for processing vehicle test data. By acquiring multiple test scenarios, their corresponding scenario complexity, and test duration during the testing process, the method determines the average complexity based on the scenario complexity and combines the average complexity with the test duration to determine cognitive load parameters, thereby quantifying the real-time cognitive stress of the test personnel. Simultaneously, it dynamically determines frequency and load thresholds based on the test duration, and uses the cognitive load parameters, frequency thresholds, and load thresholds to comprehensively judge the cognitive state of the target user. This method solves the problem of the disconnect between fixed-duration reminders and actual cognitive load by constructing a cognitive load quantification model that combines scenario complexity and test duration. By introducing a dual judgment mechanism of frequency and load thresholds, it achieves accurate identification and proactive warning of cognitive state. Furthermore, this method is adaptable to the operational needs of various intelligent driving test scenarios, effectively reducing the problem omission rate and improving the accuracy of test data processing and the reliability of intelligent driving testing.

[0061] To facilitate understanding, the following will be combined with... Figure 1 The application scenarios applicable to the embodiments of this application will be described.

[0062] Figure 1 This is a schematic diagram illustrating an application scenario provided in an embodiment of this application. Please refer to [link / reference]. Figure 1 This includes vehicles, vehicle test data processing equipment, and terminal equipment, with vehicle test data processing equipment deployed in the vehicles.

[0063] During intelligent driving road tests, test personnel interact with the vehicle test data processing equipment via terminal devices to collect relevant data in real time. Based on this data, the electronic equipment (or vehicle test data processing device) determines the test personnel's cognitive state and sends corresponding warning prompts or status information to the terminal devices. This enables real-time monitoring and proactive warnings of the test personnel's cognitive load, improving the completeness of test records and the accuracy of data processing.

[0064] The technical solution of this application and how it solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0065] Figure 2 This is a flowchart illustrating a method for processing vehicle test data provided in an embodiment of this application. Please refer to [link / reference]. Figure 2 As shown, the method may include the following steps:

[0066] S201. Obtain multiple test scenarios during vehicle operation, the scenario complexity and test duration corresponding to each test scenario.

[0067] The execution subject of this application embodiment can be an electronic device or a vehicle test data processing device installed in the electronic device. The vehicle test data processing device can be implemented by software or by a combination of software and hardware.

[0068] Test scenarios can refer to the specific traffic situations or driving events experienced by the test vehicle during intelligent driving road testing.

[0069] For example, test scenarios include, but are not limited to: "ghost peek" scenarios (pedestrians or non-motorized vehicles suddenly appear in front of a vehicle from behind an obstacle), "emergency cut-in" scenarios (vehicles in adjacent lanes cut in front of the vehicle at close range), basic straight-ahead scenarios (vehicles are traveling normally in a straight line on a structured road without any obvious traffic incidents), pedestrian crossing scenarios (pedestrians crossing the road in front of a vehicle), emergency braking scenarios of the vehicle in front, and traffic light scenarios. Different types of test scenarios place different demands on the test engineer's attention allocation and judgment abilities, and their cognitive demands vary significantly.

[0070] Scene complexity can be a quantitative indicator that reflects the degree to which a test scene consumes the cognitive resources of testers.

[0071] The scene complexity corresponds one-to-one with the scene type. The higher the scene complexity parameter value, the higher the requirements for the tester's attention, judgment and reaction speed, and the greater the cognitive input required for the tester to observe and record problems in the scene.

[0072] For example, the ghost peek scenario involves sudden and unpredictable pedestrian intrusion. Testers need to complete observation, judgment and recording in a very short time. The scenario complexity is high. For example, assuming the scenario complexity ranges from [0, 5], the complexity of the ghost peek scenario can be preset to 4.2.

[0073] The scenario of an emergency vehicle cutting in front involves the rapid recognition and response to the dynamic intentions of other vehicles. It is less complex than the scenario of a vehicle suddenly appearing out of the way, and its scenario complexity can be preset to 4.0. The basic straight-ahead scenario is a normal driving state with no sudden abnormal events, and the cognitive consumption is low. Its scenario complexity can be preset to 2.0.

[0074] It is understood that the specific values ​​mentioned above are only illustrative examples. In actual applications, they can be calibrated and adjusted based on real test data. Furthermore, the complexity of different scenario types can be pre-stored in the database of electronic devices to form a mapping table between scenario types and basic complexity for subsequent steps to query and use.

[0075] Test duration can refer to the time period elapsed from the current test cycle to the current moment. For example, test duration can be measured in minutes to indicate the time that testers have been continuously performing test work.

[0076] In some embodiments, the electronic device identifies the scenario type of the currently executed test scenario. For example, it determines the current scenario as a "ghost pedestrian" (a vehicle suddenly appearing out of the way), an emergency cut-in by a vehicle ahead, or basic straight driving by using vehicle sensor data (e.g., camera images, radar point clouds) or preset labels in the test script. Then, based on the identified scenario type, the electronic device queries a preset mapping table between scenario types and basic complexity, which can be pre-stored in the electronic device. Finally, the electronic device determines the basic complexity value corresponding to the scenario type as the scenario complexity of the test scenario.

[0077] For example, if the electronic device identifies the current scene as a ghost peek-out scene, the corresponding scene complexity obtained by looking up the table is 4.2. If it identifies it as a basic straight-ahead scene, the corresponding scene complexity obtained by looking up the table is 2.0.

[0078] In some embodiments, when acquiring the test duration, the electronic device starts a timer at the beginning of the test cycle (e.g., when the system is powered on or the test task is started) to record the test start time; at the current time, the electronic device acquires the cumulative duration recorded by the timer from the start time to the current time as the test duration. For example, if the test starts at 9:30 AM and the current time is 10:30 AM, the test duration is 60 minutes.

[0079] S202. Determine the cognitive load parameters based on the scenario complexity and test duration.

[0080] The cognitive load parameter is used to indicate the level of cognitive stress experienced by the target user during the testing process.

[0081] Understandably, cognitive load parameters comprehensively reflect the total amount or rate of cognitive resources consumed by the target user in the current testing state for observing, judging, and recording problems. A higher cognitive load parameter indicates greater cognitive stress on the tester, and a higher risk of cognitive overload and problem omission.

[0082] The target user can refer to the test personnel who perform intelligent driving test tasks and record problems, that is, the monitoring object targeted by the cognitive load parameter in this application embodiment. The target user can also refer more broadly to the operators who need to perform cognitive state assessment during intelligent driving testing.

[0083] In some embodiments, cognitive load parameters can be determined by constructing a cognitive load assessment model. Specifically, the electronic device can input scenario complexity and test duration into a pre-trained cognitive load assessment model, which is trained based on a large amount of historical test data (including annotations of the engineer's cognitive state under different scenario types and test durations) and is used to predict the current engineer's cognitive load parameters based on the input scenario complexity and test duration.

[0084] For example, the model can be built based on a regression algorithm, outputting a continuous numerical value as a cognitive load parameter; or it can be built based on a classification algorithm, outputting a cognitive load level (e.g., mild, moderate, severe), and further determining the corresponding cognitive load parameters. The advantage of this model is its ability to learn the nonlinear relationships between multidimensional parameters, improving the accuracy of cognitive load prediction.

[0085] S203. Determine the frequency threshold and load threshold based on the test duration.

[0086] Among them, the frequency threshold is used to constrain the occurrence frequency of complex scenes.

[0087] Understandably, the frequency threshold is a critical value used to determine whether high-difficulty scenarios occur too frequently within the current testing period. When the frequency of complex scenarios exceeds this threshold, it indicates that testers are dealing with too many high-cognitive-load scenarios per unit of time, reaching a level that requires attention.

[0088] Load thresholds are used to constrain cognitive load parameters.

[0089] Understandably, the cognitive load threshold is used to determine whether the current cognitive load parameters have reached a critical value requiring intervention. When the cognitive load parameters exceed this threshold, it indicates that the test subject's current cognitive stress has exceeded the acceptable range, and there is a risk of missing problems.

[0090] The load threshold can be a fixed value or it can change dynamically over time. For example, as the test duration increases, the load threshold can be reduced accordingly to reflect the objective law that engineers' cognitive tolerance decreases after long hours of work.

[0091] In some embodiments, the frequency threshold can be a preset fixed value. Specifically, the electronic device acquires a fixed frequency threshold pre-stored in a database. This threshold can be directly set based on prior industry knowledge; for example, the average number of complex scenarios that can be tolerated per unit time can be calculated based on a large amount of actual test data, and this average value is used as the frequency threshold. During the test, this frequency threshold remains unchanged regardless of changes in the test duration. This approach is suitable for scenarios with relatively stable test environments and relatively simple scenario types.

[0092] Optionally, the frequency threshold can be set based on the segmented intervals of the current test duration. Specifically, the test duration is pre-divided into multiple duration intervals, each corresponding to a frequency threshold. For example, if the test duration is in the 0-60 minute interval, the frequency threshold is set to value A; if it is in the 60-120 minute interval, the frequency threshold is set to value B; and if it exceeds 120 minutes, the frequency threshold is set to value C. As the test duration increases, the frequency threshold can be gradually decreased to reflect the objective law that engineers' tolerance for complex scenarios decreases after long-term testing.

[0093] In some embodiments, the load threshold can be a preset fixed value. Specifically, the electronic device acquires a fixed load threshold pre-stored in a database. This threshold can be derived based on statistical analysis of a large amount of test feedback data. For example, by statistically analyzing the cognitive load parameters and corresponding subjective fatigue scores of testers in historical tests, a critical value that can effectively distinguish between tolerable and overload states can be determined. During the test, this load threshold remains unchanged regardless of changes in test duration.

[0094] Optionally, the load threshold can be set in segments based on the test duration. Specifically, the test duration is pre-divided into multiple duration intervals, each corresponding to a load threshold. For example, if the test duration is in the 0-120 minute interval, the load threshold is set to value D; if it is in the 120-180 minute interval, the load threshold is set to value E, where the second parameter is lower than the first parameter; if it exceeds 180 minutes, the load threshold is set to value F, where value E is less than value D, and value F is less than value E. As the test duration increases, the load threshold gradually decreases to reflect the objective law that the cognitive tolerance of testers gradually declines after long-term work.

[0095] S204. Determine the cognitive state of the target user based on cognitive load parameters, frequency threshold, and load threshold.

[0096] Among them, the cognitive state is either a normal state or a state of overload.

[0097] Cognitive state refers to the level of cognitive resource utilization or fatigue experienced by the target user at the current testing moment. It reflects whether the tester has sufficient cognitive capacity to accurately complete the task of observing and recording problems.

[0098] A normal state refers to a state where the target user's current cognitive load is within a tolerable range, enabling them to maintain effective problem observation, judgment, and recording capabilities, and allowing testing to proceed normally. Under normal conditions, the testers' problem omission rate is at a low level.

[0099] The "load state" refers to a condition where the target user's current cognitive load exceeds their tolerance level, posing a risk of cognitive resource exhaustion and missed issues. Under load conditions, testers' observational acuity and judgment accuracy may decrease, requiring intervention in the testing process.

[0100] In some embodiments, the electronic device determines the number of test scenarios whose complexity exceeds a preset complexity threshold (i.e., the number of complex test scenarios) based on the scenario complexity of each test scenario. For example, the preset complexity threshold can be 3.0. If the scenario complexity of a test scenario is 4.2, it is considered a complex test scenario; if it is 2.0, it is not considered a complex test scenario. Then, the electronic device compares the number of complex test scenarios with a frequency threshold and compares the cognitive load parameter with a load threshold. When the number of complex test scenarios is greater than or equal to the frequency threshold, and the cognitive load parameter is greater than the load threshold (i.e., both conditions are met), the electronic device determines the target user's cognitive state as a load state. When the number of complex test scenarios is less than the frequency threshold, or the cognitive load parameter is less than or equal to the load threshold (i.e., neither condition is met), the electronic device determines the target user's cognitive state as a normal state.

[0101] Optionally, cognitive state can be determined using a pre-built classification model. Specifically, the electronic device uses cognitive load parameters, frequency thresholds, and load thresholds as input features, which are then fed into a pre-trained state classification model. This model is trained on a large amount of sample data labeled with cognitive states and is used to automatically identify the target user's current cognitive state category (normal state or load state) based on the input features.

[0102] This classification model can be implemented using algorithms such as support vector machines, decision trees, random forests, or neural networks. Compared to judgment methods based on a single comparison rule, the classification model can learn the complex relationships between multi-dimensional features, improving the robustness and accuracy of state recognition, and is especially suitable for testing environments with large individual differences or high scene diversity.

[0103] In some embodiments, when the cognitive state is determined to be normal, the electronic device maintains the normal operation of the testing process and does not trigger early warning intervention. Specifically, the electronic device can continue to perform data acquisition and cognitive state monitoring, updating cognitive load parameters, frequency thresholds, and load thresholds in real time to continuously monitor changes in the engineer's cognitive state. Optionally, the electronic device can also output cognitive state prompts at a lower frequency in the normal state (e.g., displaying an indicator that the current cognitive state is normal in the testing interface) so that the tester is aware of their own status.

[0104] In some embodiments, when the cognitive state is determined to be under load, the electronic device sends a warning command to the tester's operating terminal, triggering a pause recording warning. Specifically, after the warning is triggered, the electronic device may perform at least one of the following operations: pop up a prompt window on the operating interface and display a pause recording suggestion; illuminate a warning light or emit a buzzer sound for an audible and visual alarm; automatically pause the delivery of test tasks to prevent new test scenarios from further increasing the cognitive load; and automatically save the cached data of currently unfinished problem records to ensure that observed information is not lost.

[0105] Optionally, after triggering the alert, the electronic device can also start a preset countdown (e.g., set to 5 minutes by default) to wait for rest or adjustment; if the test status is detected to have been restored during the countdown (e.g., through keyboard interaction, confirmation prompts to close, etc.), the alert will be terminated and normal testing will resume; if no restoration instruction is received after the countdown ends, the alert status will continue or higher-level mandatory intervention measures will be implemented, such as recording the current test status and pausing all test tasks until the tester manually confirms restoration.

[0106] In this embodiment, multiple test scenarios during vehicle operation, along with the scenario complexity and test duration for each scenario, are acquired. Cognitive load parameters, used to indicate the cognitive stress level of the target user, are determined based on the scenario complexity and test duration. Frequency and load thresholds are determined based on the test duration. The cognitive state of the target user is then determined by combining the cognitive load parameters, frequency thresholds, and load thresholds. This method, tailored to the scenario characteristics of intelligent driving testing, enables real-time, dynamic monitoring of the tester's cognitive state during testing. This effectively avoids warning deviations caused by differences in scenario complexity and dynamic changes in test conditions, and resolves issues caused by cognitive overload. While improving the accuracy of vehicle test data processing, it also enhances the reliability of intelligent driving testing.

[0107] Based on any of the above embodiments, the following, in conjunction with Figure 3 The process of determining cognitive load parameters based on scenario complexity and test duration is explained in detail.

[0108] Figure 3 This is a schematic diagram illustrating the process of determining cognitive load parameters as provided in an embodiment of this application. Please refer to... Figure 3 The method may include:

[0109] S301. Determine the average complexity based on the scenario complexity.

[0110] Average complexity refers to the average value obtained by comprehensively quantifying the difficulty level of all test scenarios experienced in the current testing cycle. Average complexity is used to represent the overall technical difficulty level of the current testing task. The higher the average complexity, the more complex the overall testing task faced by the testers in the current testing cycle.

[0111] In some embodiments, the average complexity can be determined based on the scenario complexity by: obtaining a first quantity corresponding to the test scenario; and determining the average complexity based on the scenario complexity and the first quantity of each test scenario.

[0112] The first quantity can refer to the total number of test scenarios experienced or evaluated by the tester within the current testing cycle. For example, if the tester handles 10 test scenarios in a continuous testing process, then the first quantity is 10.

[0113] In some embodiments, when determining the average complexity, the electronic device obtains the scenario complexity values ​​corresponding to each of the test scenarios within the current testing period, then sums these scenario complexity values ​​to obtain the total complexity, and finally divides the total complexity by a first quantity to obtain the average complexity. For ease of understanding, this process can be expressed as: Average complexity = (Sum of the complexities of each scenario) / Total number of scenarios. For example, if there are 3 test scenarios within the current testing period, and the complexities of each scenario are 4.2, 2.0, and 4.0 respectively, then the average complexity is the arithmetic mean of these values, i.e., the average complexity is 3.4. This calculation method reflects the overall scenario difficulty through equal-weighted averaging, but does not consider the nonlinear impact of scenario type distribution on cognitive load. Subsequently, dynamic adjustments are made by introducing perturbation weights to more accurately match actual changes in cognitive pressure.

[0114] By using the above method, the complexity differences of different scenario types can be averaged to obtain a quantitative indicator reflecting the overall difficulty level of the current test task. This avoids the shortcomings of evaluating different difficulty scenarios in a fragmented manner and improves the consistency and operability of cognitive load quantification.

[0115] In some embodiments, the average complexity can be determined based on the scenario complexity and a first quantity of each test scenario as follows: determining the perturbation weights corresponding to each test scenario, whereby the perturbation weights are used to indicate the degree of influence of the test scenario on the cognitive state; and determining the average complexity based on the scenario complexity, multiple perturbation weights, and the first quantity.

[0116] Perturbation weights refer to coefficients that adjust the proportion of influence of different test scenarios in the calculation of average complexity. Perturbation weights reflect the contribution of a particular test scenario's complexity to the overall average complexity. These perturbation weights can be dynamically determined based on the scenario type; different types and difficulties of scenarios can correspond to different perturbation weight values. A larger perturbation weight indicates a greater impact of that scenario on the tester's cognitive state.

[0117] In some embodiments, one way to determine the perturbation weight is as follows: the electronic device analyzes the combination type of each test scenario in the current test cycle in real time, and dynamically adjusts the perturbation weight of each scenario according to the distribution of high-complexity scenarios and low-complexity scenarios.

[0118] Specifically, the electronic device first counts the percentage of high-complexity scenarios within the current testing period. When the percentage of high-complexity scenarios exceeds a preset threshold (e.g., 60%), it indicates that the current test is under high-pressure and intensive conditions. In this case, the electronic device increases the perturbation weight of high-complexity scenarios from their initial values ​​(e.g., from 1.0 to 1.5) and decreases the perturbation weight of low-complexity scenarios from their initial values ​​(e.g., from 1.0 to 0.5). Conversely, if the percentage of high-complexity scenarios does not exceed the preset threshold, the perturbation weight of each scenario remains at its default value (e.g., all are 1.0).

[0119] For example, a high-complexity scenario can refer to a test scenario whose complexity exceeds a preset complexity threshold (e.g., 3.0), while a low-complexity scenario can refer to a test scenario whose complexity is less than or equal to that threshold. This adjustment method is based on the following cognitive principle: in a test environment where high-complexity scenarios occur frequently, the tester's attention is under continuous high tension. At this time, the consumption of cognitive resources by high-complexity scenarios is amplified, while the consumption of cognitive resources by low-complexity scenarios interspersed among them is relatively reduced.

[0120] In some embodiments, when determining the average complexity, the electronic device obtains the scenario complexity value and perturbation weight value corresponding to each of the test scenarios in the current test cycle; then, it multiplies the scenario complexity of each test scenario with its corresponding perturbation weight to obtain the weighted complexity of each test scenario; next, it sums all the weighted complexities to obtain the total weighted complexity; finally, it divides the total weighted complexity by a first quantity to obtain the average complexity after weighted averaging.

[0121] For ease of understanding, this process can be expressed as: Average complexity = (∑(complexity of each scenario × perturbation weight)) / total number of scenarios. By introducing perturbation weights, the average complexity can more realistically reflect the actual cognitive load experienced by testers under mixed scenario combinations, avoiding overestimation or underestimation of the actual cognitive load under the fixed weight model.

[0122] By using the above methods, the influence ratio of different scenarios in the average complexity calculation can be dynamically adjusted according to the actual distribution of high-complexity and low-complexity scenarios in the current testing cycle. This makes the average complexity calculation results more consistent with the actual cognitive load felt by testers under mixed scenario combinations, thereby improving the accuracy of average complexity calculation and scenario adaptability.

[0123] S302. Determine the cognitive load parameters based on the average complexity and test duration.

[0124] In some embodiments, the cognitive load parameter can be determined based on the average complexity and test duration as follows: a baseline cognitive load parameter is determined based on the average complexity and test duration; the baseline cognitive load parameter is then corrected to obtain the cognitive load parameter.

[0125] The baseline cognitive load parameter represents the cognitive consumption rate of the target user during the testing process; that is, the amount of cognitive resources consumed by the tester per unit of time in processing the test scenario. The baseline cognitive load parameter reflects the basic cognitive consumption rate of the tester under given average complexity and test duration.

[0126] In some embodiments, the electronic device calculates a baseline cognitive load parameter based on a first quantity (total number of scenarios), average complexity, test duration, and a preset keyboard input cost baseline coefficient. Specifically, the baseline cognitive load parameter = (first quantity + keyboard input cost baseline coefficient × average complexity) / (normalization coefficient × average complexity × test duration), and the relevant formula can be expressed as:

[0127]

[0128] Wherein, CL1 is the baseline cognitive load parameter, N is the first quantity, that is, the total number of scenarios processed by the testers in the current test cycle, C_avg is the average complexity, T is the test duration, 30 is the keyboard input cost baseline coefficient, which is used to convert the semantic encoding and text input burden when the testers describe the scenario into an equivalent number of scenarios to be processed, and 10 is the normalization coefficient, which is used to adjust the calculation results to a reasonable numerical range.

[0129] Among them, the keyboard input cost benchmark coefficient is used to convert the mental and operational burden of testers performing semantic encoding and text input when describing test scenarios into a calculable numerical quantity. That is, when testers describe a problem for a basic straight scenario (complexity calibrated value of 2.0), they need to input an average of 15 characters. After model normalization derivation, that is, (15 / 2.0)×4=30, the keyboard input cost benchmark coefficient is obtained.

[0130] In this formula, the numerator (N + 30 × C_avg) comprehensively reflects the cognitive cost in terms of both scene processing volume and semantic description depth, while the denominator (10 × C_avg × T) introduces a time dimension to calculate the cognitive consumption rate. The specific values ​​in the above formula (e.g., 30, 10) are for illustrative purposes only and can be calibrated and adjusted based on actual measurement data in practical applications.

[0131] In some embodiments, the baseline cognitive load parameter only reflects the intensity of basic task processing under the condition of excluding repetitive interference. However, in actual testing, when test scenarios with the same complexity are repeated continuously, the vigilance of testers will gradually decrease due to the habituation effect, and cognitive fatigue will accumulate more rapidly. This repetitive fatigue effect requires additional correction to the baseline cognitive load parameter.

[0132] Therefore, electronic devices can acquire information on the repetition of scenarios with the same complexity that occur consecutively within the current testing cycle. Based on this repetition, adjustment parameters are determined to represent the cumulative effect of repetitive fatigue. These adjustment parameters are then used to amplify and correct the baseline cognitive load parameters, resulting in the final cognitive load parameters. The corrected cognitive load parameters can more comprehensively reflect the actual cognitive stress experienced by test subjects when facing repetitive tasks.

[0133] By employing the above method, a more comprehensive set of cognitive load parameters can be obtained by calculating the basic cognitive depletion rate and then compensating for repetitive fatigue effects with a correction factor. This approach integrates factors from both task intensity and time accumulation, improving the completeness and accuracy of cognitive load parameter calculations compared to single-dimensional assessments.

[0134] In some embodiments, the baseline cognitive load parameter can be corrected to obtain the cognitive load parameter by: determining a second quantity corresponding to test scenarios with the same complexity; determining an adjustment parameter based on the second quantity; and correcting the baseline cognitive load parameter based on the adjustment parameter to obtain the cognitive load parameter.

[0135] The second quantity refers to the number of test scenarios with the same basic complexity value that occur consecutively within the current testing period. For example, if a tester handles 5 consecutive ghost-probe scenarios (with the same complexity), the second quantity is 5. The second quantity is used to quantify the degree of repetitive cognitive fatigue accumulated by the tester due to continuously handling scenarios of the same difficulty level.

[0136] The adjustment parameter refers to the multiplier used to amplify and correct the baseline cognitive load parameter. The value of the adjustment parameter is positively correlated with the second quantity; that is, the more times a scenario of the same complexity is repeated consecutively, the larger the value of the adjustment parameter, and the more significant the amplification effect on the baseline cognitive load parameter. The adjustment parameter is used to simulate the cumulative amplification effect of cognitive fatigue caused by repetitive tasks.

[0137] In some embodiments, the process of determining the adjustment parameter can be as follows: the electronic device obtains a preset repetition influence weight coefficient, then multiplies this coefficient by a second quantity and adds 1 to obtain the adjustment parameter. For ease of understanding, this process can be expressed as: Adjustment parameter = 1 + Repetition influence weight coefficient × Second quantity.

[0138] For example, if the repetition effect weighting coefficient is 0.21 and the second quantity is R, then the adjustment parameter = 1 + 0.21 × R. The repetition effect weighting coefficient (i.e., 0.21) can be determined through regression analysis on a large amount of real-world scenario data, reflecting the incremental impact of each repetition of a scenario of the same complexity on the cognitive load rate.

[0139] Specifically, the process of determining the weight coefficient of repeated effects can be as follows: obtain a large amount of actual test scenario data (e.g., more than 300 actual test scenarios) and the corresponding cognitive state labels of the testers; use the number of consecutive repetitions of scenarios with the same complexity as the independent variable and the incremental change of cognitive load rate as the dependent variable to perform regression analysis; and determine the regression coefficient of the independent variable as the weight coefficient of repeated effects based on the regression analysis results.

[0140] For example, regression analysis determined the repetition effect weighting coefficient to be 0.21, which statistically means that for each consecutive repetition of a test scenario of the same complexity, the cognitive load rate increases by 21%. Regression analysis verified that the coefficient of determination (R²) reached 0.95, indicating a strong correlation between the independent and dependent variables, thus confirming the statistical rationality and reliability of the coefficient value. This coefficient represents the cumulative amplification effect of consecutive repetitions of scenarios of the same complexity on cognitive load, making the quantification of cognitive load parameters more closely reflect the actual cognitive stress changes of test subjects in repetitive task scenarios.

[0141] In some embodiments, after obtaining the adjustment parameters, the baseline cognitive load parameters are corrected to obtain the cognitive load parameters. One implementation method is that the electronic device multiplies the baseline cognitive load parameters by the adjustment parameters to obtain the corrected cognitive load parameters.

[0142] This process can be expressed as: Cognitive load parameter = Baseline cognitive load parameter × Adjustment parameter. For example, if the baseline coefficient for keyboard input cost is 30, the repetition frequency influence coefficient is 0.21, and the normalization coefficient is 10, then the relevant formula for the cognitive load parameter can be expressed as:

[0143]

[0144] By using the multiplicative correction method described above, the cumulative effect of repetitive fatigue is quantitatively amplified, so that the final cognitive load parameter can more realistically reflect the actual cognitive stress level of the tester when continuously dealing with scenarios of the same difficulty.

[0145] By using the above method, the cumulative amplification effect of repetitive tasks on cognitive fatigue can be accurately quantified based on the number of consecutive repetitions of scenarios with the same complexity. This effect can then be used to correct the baseline cognitive load parameters in the form of adjustment parameters. This ensures that the final cognitive load parameters not only consider the difficulty level and test duration of the scenario itself, but also take into account the additional impact of task repetition on cognitive consumption, thereby improving the comprehensiveness and accuracy of the cognitive load parameter calculation.

[0146] In this embodiment, the average complexity is determined by acquiring the scenario complexity of multiple test scenarios. A baseline cognitive load parameter is then determined based on the average complexity and test duration. Finally, the baseline cognitive load parameter is corrected based on the number of consecutive repetitions of scenarios with the same complexity to obtain the final cognitive load parameter. This method comprehensively quantifies the cognitive consumption rate of test personnel based on multiple dimensions such as the difficulty level of the test scenario, the cumulative effect of time, and repetitive fatigue. This avoids the problem of single-dimensional assessment being disconnected from actual cognitive load and improves the accuracy of cognitive load parameter calculation.

[0147] Based on any of the above embodiments, the following, in conjunction with Figure 4 The process of determining the frequency threshold and load threshold based on the test duration is explained in detail.

[0148] Figure 4 This is a schematic diagram illustrating the process of determining the frequency threshold and load threshold provided in the embodiments of this application. Please refer to... Figure 4 The method may include:

[0149] S401. Determine the frequency threshold based on the test duration.

[0150] In some embodiments, a preset density threshold is obtained, which is a critical value for the number of complex scenarios per unit time (unit: scenarios / hour). Then, the density threshold is multiplied by the test duration (in hours) to obtain the frequency threshold. The relevant formula can be expressed as: Frequency threshold = 10.25 × (T / 60). Where T is the test duration (minutes), T / 60 is used to convert the test duration to hours; 10.25 is the density threshold, representing the upper limit critical value for the number of complex scenarios that can be tolerated per hour, which is determined based on statistical analysis of a large amount of actual test scenario data.

[0151] When the density of complex scenarios exceeds this threshold, it indicates that the testers are facing sustained high levels of cognitive stress. The physical meaning of the frequency threshold is: if the number of complex scenarios reaches this value within the current test duration, it means that the frequency of complex scenarios is already at a high level, requiring further judgment in conjunction with cognitive load parameters.

[0152] S402. Determine the load threshold based on the test duration.

[0153] In some embodiments, the load threshold can be determined based on the test duration in the following manner: when the test duration is less than or equal to a preset duration, the load threshold is determined as a first parameter; when the test duration is greater than the preset duration, the load threshold is determined based on the first parameter and the test duration.

[0154] The preset duration can refer to a pre-defined critical value for the length of time. In this embodiment, the preset duration is a reference duration for which testers can maintain a stable level of cognitive load without additional intervention. For example, the preset duration can be 120 minutes (i.e., 2 hours). It should be understood that this value is illustrative and can be adaptively adjusted according to factors such as the characteristics of the testers and the testing environment in actual applications.

[0155] The first parameter can refer to a pre-set baseline load threshold, which is used as the load threshold when the test duration does not exceed the preset duration. For example, the first parameter can be 0.12. When the cognitive load parameter reaches this value, most testers indicated that they felt high cognitive stress and needed to take a break.

[0156] In some embodiments, when the test duration is less than or equal to a preset duration, the load threshold is determined as the first parameter. Specifically, if the preset duration is 120 minutes, then when the test duration T ≤ 120 minutes, the load threshold DT is set to the first parameter 0.12.

[0157] Since the cognitive tolerance of test subjects is relatively stable and sufficient when the test duration is within 2 hours, a fixed baseline load threshold can be used as the judgment standard to effectively identify the risk of cognitive overload without the need for additional attenuation adjustments. The relevant formula can be expressed as: if T≤120 minutes, then DT=0.12.

[0158] In some embodiments, when the test duration exceeds a preset duration, a load threshold is determined based on a first parameter and the test duration. Specifically, if the preset duration is 120 minutes, then when the test duration T > 120 minutes, the load threshold DT is attenuated based on the first parameter: DT = 0.12 - 0.02 × ((T - 120) / 60). Here, (T - 120) represents the excess duration (in minutes) beyond the preset duration, (T - 120) / 60 is used to convert the excess duration to hours, and 0.02 is the attenuation coefficient, meaning that after the test time exceeds 2 hours, the trigger threshold decreases by 0.02 for every hour exceeding 2 hours.

[0159] As the testing duration exceeds two hours, the cognitive tolerance of testers gradually decreases with the extension of the testing time. If the same load threshold is used as at the beginning of the test, the optimal warning time may be missed due to the threshold being too high. Therefore, it is necessary to appropriately reduce the load threshold to make the warning more sensitive. This attenuation process reflects the objective physiological law of the decline in testers' tolerance after prolonged continuous work, making the warning triggering conditions more stringent as the testing duration increases, and enabling more timely identification of cognitive overload risks in the later stages of the test.

[0160] By using the above methods, the load threshold can be dynamically adjusted according to changes in test duration, so that the warning triggering conditions match the tolerance decay pattern of testers, thereby improving the accuracy and adaptability of cognitive load warnings in long-term test scenarios.

[0161] In this embodiment, the frequency threshold and load threshold are dynamically determined based on the test duration. In this method, the frequency threshold increases linearly with the test duration to accommodate a reasonable tolerance range for the density of complex scenarios under different test durations. The load threshold remains fixed within a preset duration, and then linearly decreases with increasing test duration after exceeding the preset duration to match the objective law of decreased tester tolerance after prolonged testing. Through this approach, dynamic adaptation of the two thresholds to the test duration is achieved, providing a benchmark reference value that matches the real-time test status for accurate subsequent judgment of cognitive state, thus improving the accuracy and reliability of cognitive load warning.

[0162] Figure 5 This is a schematic diagram illustrating the process of determining cognitive states as provided in an embodiment of this application. Please refer to... Figure 5 The method may include:

[0163] S501. Determine the third quantity corresponding to the complex test scenario.

[0164] The third quantity can refer to the total number of test scenarios whose scenario complexity exceeds a preset complexity threshold within the current test cycle.

[0165] The third quantity is used to quantify the frequency of occurrence of high-difficulty scenarios within the current testing cycle, and is an important basis for judging whether testers are facing a continuous high-intensity cognitive challenge. For example, the preset complexity threshold can be 3.0. If the scenario complexity of a certain test scenario is 4.2, then the scenario is included in the third quantity; if the scenario complexity is 2.0, then it is not included in the third quantity.

[0166] S502. When the third quantity is greater than or equal to the frequency threshold and the cognitive load parameter is greater than the load threshold, the cognitive state is determined to be a load state.

[0167] In some embodiments, the electronic device determines whether the third quantity is greater than or equal to a frequency threshold, and simultaneously determines whether the cognitive load parameter is greater than a load threshold. When both conditions are met, the electronic device determines that the target user's cognitive state is a load state.

[0168] The frequency condition is used to confirm that the density of high-difficulty scenarios in the current testing period has reached a warning level requiring attention, while the load condition is used to confirm that the tester's current cognitive load parameter has exceeded the tolerable critical range. Only when high-difficulty scenarios occur frequently and the tester's cognitive load parameter exceeds the threshold is it determined to be a load state and an early warning is triggered.

[0169] The aforementioned dual confirmation mechanism avoids the misjudgment that may be caused by triggering an early warning based on a single condition. It also avoids triggering an early warning when the cognitive load parameter only increases temporarily due to a few extreme scenarios but the overall testing environment is not dense. Furthermore, it avoids premature intervention when complex scenarios occur densely but the cognitive load parameter has not yet exceeded the threshold due to individual differences.

[0170] S503. When the third quantity is less than the frequency threshold, or the cognitive load parameter is less than or equal to the load threshold, the cognitive state is determined to be normal.

[0171] In some embodiments, the electronic device determines whether the third quantity is less than a frequency threshold, or whether the cognitive load parameter is less than or equal to a load threshold. When either of these two conditions is met, the electronic device determines that the target user's cognitive state is normal.

[0172] Understandably, only when the third quantity is less than the frequency threshold (i.e., the density of high-difficulty scenarios has not yet reached the level that needs attention), or the cognitive load parameter does not exceed the load threshold (i.e., the tester's current cognitive pressure has not exceeded the tolerable range), is the current cognitive state considered normal, no warning is triggered, and the test process can continue.

[0173] The above approach ensures the conservatism and security of early warning triggering, avoiding unnecessary early warning interventions in relatively relaxed testing environments or when testers are in a good cognitive state, thus maintaining the continuity of testing. Simultaneously, after determining a normal state, the electronic device continues to collect data and monitor cognitive state, updating parameters in real time to continuously monitor changes in the testers' cognitive state.

[0174] In this embodiment, a third quantity corresponding to a complex test scenario is determined, and this third quantity is compared with a frequency threshold and a cognitive load parameter is compared with a load threshold. If the third quantity is greater than or equal to the frequency threshold and the cognitive load parameter is greater than the load threshold, the cognitive state is determined to be a load state; otherwise, the cognitive state is determined to be a normal state. This method introduces a dual-condition judgment mechanism, requiring both the frequency condition and the load condition to be met simultaneously to trigger an alert. This effectively distinguishes between scenarios where high-difficulty scenarios occur frequently but the tester's cognition is still manageable, and scenarios where high-difficulty scenarios occur frequently and the tester's cognition is overloaded. This avoids misjudgment or missed judgment caused by a single condition trigger, improving the accuracy and reliability of cognitive state judgment.

[0175] Figure 6 This is a flowchart illustrating another method for processing vehicle test data provided in an embodiment of this application. Please refer to... Figure 6 The method may include:

[0176] S601. Obtain each test scenario, as well as the corresponding scenario complexity and test duration.

[0177] Specifically, the electronic device determines the scenario complexity corresponding to each test scenario based on the scenario type of the current test period. Specifically, the electronic device obtains the scenario type of the current test period, queries a preset mapping table between scenario types and basic complexity, and determines the obtained basic complexity value as the scenario complexity of that test scenario.

[0178] S602, Determine keyboard input cost.

[0179] The electronic device determines the keyboard input cost based on the average problem complexity and a preset baseline value. Specifically, the keyboard input cost = measured baseline value × average problem complexity, where the baseline value is calculated based on the number of descriptive words and input efficiency in a basic straight-line scenario. The keyboard input cost represents the cognitive burden incurred by testers due to semantic encoding and text input when describing test scenarios. It transforms the abstract cost of cognitive depth into a calculable numerical value, which is then incorporated into the subsequent quantitative calculation of cognitive load rate.

[0180] S603. Calculate cognitive load parameters.

[0181] Among them, the cognitive load rate of electronic devices is dynamically calculated based on the total number of scenarios, average problem complexity, number of consecutive repetitions of scenarios with the same complexity, keyboard input cost, and test duration.

[0182] Specifically, the electronic device first determines the average problem complexity based on the scenario complexity of each test scenario, then determines the baseline cognitive load parameter based on the total number of scenarios, the average problem complexity, and the test duration, and then corrects the baseline cognitive load parameter based on the number of consecutive repetitions of scenarios with the same complexity to obtain the cognitive load parameter.

[0183] S604. Based on the cognitive load parameters, frequency threshold, and load threshold, determine whether to trigger an early warning.

[0184] Specifically, the electronic device determines whether to trigger an alert based on cognitive load parameters, frequency thresholds, and load thresholds. Specifically, the electronic device first determines the third number of complex test scenarios within the current test cycle and compares this third number with the frequency threshold, while simultaneously comparing the cognitive load parameters with the load threshold.

[0185] When the third quantity is greater than or equal to the frequency threshold and the cognitive load parameter is greater than the load threshold, it indicates that the tester is currently under the dual pressure of frequent occurrences of high-difficulty scenarios and cognitive load exceeding the tolerable range, and the electronic device triggers a pause recording warning.

[0186] When the third quantity is less than the frequency threshold, or the cognitive load parameter is less than or equal to the load threshold, it indicates that the density of high-difficulty scenarios has not yet reached a level that requires attention, or that the current cognitive pressure of the tester has not exceeded the acceptable range. The electronic device does not trigger an alarm, and the test process continues.

[0187] S605, Execute early warning operation.

[0188] Specifically, when a pause recording alert is triggered, the electronic device executes an alert operation. Specifically, the electronic device sends an alert command to the test management system or the tester's terminal, prompting the tester to pause issue recording.

[0189] Meanwhile, the electronic device automatically saves the current test status, ensuring that acquired test data is not lost. Upon receiving the warning, testers pause recording and take a break to adjust, resuming testing once their cognitive state has recovered. This warning system effectively reduces the rate of missed problems due to cognitive overload, improving the completeness and reliability of test records.

[0190] In this application embodiment, regarding test quality assurance, a measured correlation model between cognitive load and problem omission rate is established by using cognitive load rate as an early warning indicator for problem omission rate. When the cognitive load rate exceeds a critical threshold, a pause recording warning is triggered, significantly reducing the problem omission rate. Furthermore, regarding the accuracy of the warning, multiple key parameters verified based on extensive actual test operations ensure that the warning can be triggered promptly when the engineer's cognitive load reaches a critical point, achieving proactive warning and real-time intervention for cognitive fatigue. Moreover, regarding test efficiency and user experience, this application achieves proactive control of the problem omission rate through the warning mechanism, avoiding repeated testing and time waste caused by problem omissions, significantly increasing the average number of daily test scenarios, and significantly improving satisfaction. In addition, by establishing a measured correlation between cognitive load rate and problem omission rate, a scientific and effective solution is provided for the field of intelligent driving testing, covering typical scenarios such as ghost peeks, basic straight driving, and vehicles cutting in front, fully verifying the effectiveness and practicality of this solution.

[0191] Figure 7 This is a flowchart illustrating another method for processing vehicle test data provided in this application. Please refer to... Figure 7 ,include:

[0192] In the actual testing phase S1, testers perform intelligent driving test operations in the actual test environment and monitor the test process in real time through the data acquisition client. In the data acquisition phase S2, four types of key data are collected and uploaded in real time, including the total number of scenarios (first quantity), the scenario type of each test scenario, the test duration from the start of the test cycle to the current moment, and the number of times scenarios with the same basic complexity value appear consecutively (third quantity). These data serve as the basic input for subsequent calculations.

[0193] In the core parameter calculation stage S3, the corresponding basic complexity value is first queried from the predefined mapping table according to each scenario type and the average problem complexity is calculated. Then, the cognitive load parameter is calculated by integrating the number of scenarios, average complexity, test duration and repetitiveness. At the same time, dynamic thresholds (including frequency thresholds and load thresholds) are dynamically calculated based on the test duration.

[0194] In the condition judgment stage S4, the frequency condition judgment is first performed, that is, to determine whether the frequency of occurrence of complex scenarios has reached the warning level. Only when the frequency condition is met will the load condition judgment be further performed, that is, to compare whether the cognitive load parameters exceed the load threshold under the current test duration.

[0195] In the decision output phase S5, when both the frequency and load conditions are met, the system decision output needs to pause and send an early warning to the tester's interface. This instruction can be associated with specific operations such as automatically pausing test case deployment, popping up a forced rest prompt, or lighting up an alarm light. If any of the above conditions are not met, the system decision output remains in the current state, and the test process continues normally.

[0196] Furthermore, the decision-making instructions are fed back to the execution end, i.e., the execution feedback stage S6, completing a full closed loop from perception, calculation, judgment to execution. This method achieves real-time monitoring and proactive early warning of the cognitive load of test personnel through the above process.

[0197] For example, taking the precise triggering under high-intensity testing as an example, the tester performed a continuous test task lasting 60 minutes, experiencing a total of 28 test scenarios, including 22 ghost peeking scenarios (scenario complexity of 4.2) and 6 basic straight-line scenarios (scenario complexity of 2.0). Seven ghost peeking scenarios occurred consecutively in the middle of the test.

[0198] First, calculate the average problem complexity (if the perturbation weight is ignored), which is to sum the complexity of each scenario and divide by the total number of scenarios, to get the average complexity C_avg = (22 × 4.2 + 6 × 2.0) / 28 ≈ 3.73; then calculate the cognitive load rate, which is to use the formula for calculating the cognitive load parameter CL according to the total number of scenarios, average complexity, test duration and number of repetitions, and substitute N=28, C_avg≈3.73, T=60, R=7, to get the cognitive load rate CL≈0.154.

[0199] Since the current test duration is 60 minutes, which does not exceed the preset duration threshold of 120 minutes, the system sets the load threshold to DT=0.12. In the frequency condition judgment, the system identifies ghost peeks as a complex scene with a quantity of 22. The frequency threshold is calculated as 10.25×(60 / 60)=10.25, and 22≥10.25, so the frequency condition is met. In the load condition judgment, the cognitive load rate of 0.154 is greater than the load threshold of 0.12, so the load condition is also met.

[0200] Since both conditions are met simultaneously, the system outputs a pause command. At this point, the cognitive load rate has exceeded the baseline point at which testers indicated the need to pause and rest, and complex scenarios occur frequently. The system's intervention is synchronized with the subjective fatigue feelings of most testers, achieving accurate early warning under high load and avoiding the risk of decreased efficiency or misjudgment that might result from testers continuing testing in an overly fatigued state.

[0201] For example, taking the stable operation in a regular test as an example, the tester performs a test task with a duration of 95 minutes, experiencing a total of 18 test scenarios, including 5 scenarios of vehicles cutting in front (complexity 4.0), 8 scenarios of pedestrians crossing (complexity 3.5) and 5 basic straight-through scenarios (complexity 2.0). The different types of scenarios appear alternately, with a maximum of 3 consecutive repetitions.

[0202] First, the average problem complexity (ignoring perturbation weights) is calculated as C_avg = (5 × 4.0 + 8 × 3.5 + 5 × 2.0) / 18 ≈ 2.44. Then, the cognitive load rate is calculated. Using the formula for the cognitive load parameter CL, substituting N=18, C_avg ≈ 2.44, T=95, and R=3, the cognitive load rate CL ≈ 0.064 is obtained. Since the current test duration is 95 minutes, which does not exceed the preset duration threshold, the system sets the load threshold to DT = 0.12.

[0203] In the frequency condition judgment, scenarios with a complexity higher than 3.0 are considered complex scenarios, and their number is 5+8=13. The frequency threshold is calculated as 10.25×(95 / 60)≈16.23. Since 13<16.23, the frequency condition is not met. The cognitive load parameter 0.064<0.12 also does not meet the condition. That is, the system can directly decide to output instructions, and the test process continues normally.

[0204] Although the test lasted nearly 100 minutes, the overall complexity of the scenarios was moderate and dispersed, and neither the scenario frequency nor the cognitive load rate reached the trigger threshold. Therefore, the system did not intervene unnecessarily, allowing testers to maintain a consistent workflow. This demonstrates the importance of frequency conditions as a pre-filter, effectively suppressing false triggers and ensuring that the system only intervenes during truly high-load, high-density challenging phases, thereby increasing testers' trust and satisfaction with the system.

[0205] In this embodiment, the proposed solution constructs a multi-factor coupled computational model based on empirical data. This model integrates multiple factors such as the number of scenarios, complexity, test duration, and repetitiveness through cognitive load rate, and combines a dynamic threshold that adaptively adjusts with working duration, as well as a frequency filter to prevent misjudgments in low-density, high-load scenarios, forming a stable, robust, and accurate decision-making system. This method can effectively distinguish between high-intensity saturated working states and regular variable working states, ensuring a high degree of synchronization between system decisions and the subjective cognitive perception of testers, achieving the technical effects of significantly reducing false trigger rates and significantly improving tester satisfaction.

[0206] Figure 8 This is a schematic diagram of a vehicle test data processing device provided in an embodiment of this application. Please refer to... Figure 8 The vehicle test data processing device 800 includes: an acquisition module 801, a first determination module 802, a second determination module 803, and a processing module 804, wherein:

[0207] The acquisition module 801 is used to acquire multiple test scenarios of the vehicle during driving, the scenario complexity and test duration of each test scenario;

[0208] The first determining module 802 is used to determine the cognitive load parameter based on the scenario complexity and test duration. The cognitive load parameter is used to indicate the level of cognitive stress of the target user during the test.

[0209] The second determining module 803 is used to determine the frequency threshold and the load threshold based on the test duration. The frequency threshold is used to constrain the frequency of occurrence of complex scenarios, and the load threshold is used to constrain cognitive load parameters.

[0210] The processing module 804 is used to determine the cognitive state of the target user, which is either a normal state or a load state, based on the cognitive load parameters, frequency threshold, and load threshold.

[0211] In one possible implementation, the first determining module 802 is specifically used for:

[0212] Determine the average complexity based on the complexity of the scenario;

[0213] Cognitive load parameters are determined based on average complexity and test duration.

[0214] In one possible implementation, the first determining module 802 is specifically used for:

[0215] Get the first quantity corresponding to the test scenario;

[0216] The average complexity is determined based on the scenario complexity and the number of elements in each test scenario.

[0217] In one possible implementation, the first determining module 802 is specifically used for:

[0218] Determine the perturbation weights for each test scenario. The perturbation weights are used to indicate the degree of influence of the test scenario on the cognitive state.

[0219] The average complexity is determined based on the scenario complexity, multiple perturbation weights, and the first quantity.

[0220] In one possible implementation, the first determining module 802 is specifically used for:

[0221] Based on the average complexity and test duration, a baseline cognitive load parameter is determined, which is the cognitive consumption rate of the target user during the test.

[0222] The baseline cognitive load parameters are corrected to obtain the cognitive load parameters.

[0223] In one possible implementation, the first determining module 802 is specifically used for:

[0224] Determine the second quantity corresponding to test scenarios with the same complexity;

[0225] Based on the second quantity, the adjustment parameters are determined, and the baseline cognitive load parameters are corrected according to the adjustment parameters to obtain the cognitive load parameters.

[0226] In one possible implementation, the second determining module 803 is specifically used for:

[0227] When the test duration is less than or equal to the preset duration, the load threshold is determined as the first parameter;

[0228] When the test duration exceeds the preset duration, the load threshold is determined based on the first parameter and the test duration.

[0229] In one possible implementation, the processing module 804 is specifically used for:

[0230] Determine the third quantity corresponding to complex test scenarios;

[0231] When the third quantity is greater than or equal to the frequency threshold and the cognitive load parameter is greater than the load threshold, the cognitive state is determined to be a load state.

[0232] In one possible implementation, the processing module 804 is specifically used for:

[0233] When the third quantity is less than the frequency threshold, or the cognitive load parameter is less than or equal to the load threshold, the cognitive state is determined to be normal.

[0234] The vehicle test data processing device provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0235] Figure 9 This is a schematic diagram of the structure of a vehicle test data processing device provided in an embodiment of this application. Figure 9 As shown, the vehicle test data processing device 900 may include: a transceiver 901, a processor 902, and a memory 903.

[0236] Processor 902 executes computer execution instructions stored in memory, causing processor 902 to perform the scheme in the above embodiments. Processor 902 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0237] The memory 903 is connected to the processor 902 via the system bus and completes communication between them. The memory 903 is used to store computer program instructions.

[0238] Transceiver 901 can be used to obtain the task to be run and its configuration information.

[0239] The system bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus. Transceivers are used to enable communication between database access devices and other computers (e.g., clients, read-write libraries, and read-only libraries). Memory may include random access memory (RAM) and may also include non-volatile memory.

[0240] This application also provides a chip for executing instructions, which is used to execute the technical solution of the vehicle test data processing method in the above embodiments.

[0241] This application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer performs the technical solution of the vehicle test data processing method described in the above embodiments.

[0242] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, it can implement the technical solution of the vehicle test data processing method in the above embodiments.

[0243] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0244] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to implement the solution of this embodiment according to actual needs.

[0245] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The unit composed of the above modules can be implemented in hardware or in the form of hardware plus software functional units.

[0246] The integrated modules described above, implemented as software functional modules, can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods of the various embodiments of this application.

[0247] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0248] The memory may include high-speed RAM, and may also include non-volatile storage (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0249] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0250] The aforementioned storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0251] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Alternatively, the storage medium can be an integral part of the processor. The processor and storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and storage medium can exist as discrete components in an electronic control unit or main control device.

[0252] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0253] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for processing vehicle test data, characterized in that, The method includes: Obtain multiple test scenarios during vehicle operation, along with the scenario complexity and test duration for each scenario. Based on the scenario complexity and the test duration, a cognitive load parameter is determined, which is used to indicate the level of cognitive stress experienced by the target user during the test. Based on the test duration, a frequency threshold and a load threshold are determined. The frequency threshold is used to constrain the frequency of occurrence of complex scenarios, and the load threshold is used to constrain the cognitive load parameter. The cognitive state of the target user is determined based on the cognitive load parameter, the frequency threshold, and the load threshold. The cognitive state is either a normal state or a load state.

2. The method according to claim 1, characterized in that, Based on the scenario complexity and the test duration, cognitive load parameters are determined, including: Determine the average complexity based on the scenario complexity. The cognitive load parameters are determined based on the average complexity and the test duration.

3. The method according to claim 2, characterized in that, Based on the scenario complexity, the average complexity is determined, including: Obtain the first quantity corresponding to the test scenario; The average complexity is determined based on the scenario complexity of each test scenario and the first quantity.

4. The method according to claim 3, characterized in that, The average complexity is determined based on the scenario complexity of each test scenario and the first quantity, including: Determine the perturbation weights corresponding to each test scenario, wherein the perturbation weights are used to indicate the degree of influence of the test scenario on the cognitive state; The average complexity is determined based on the scenario complexity, multiple perturbation weights, and the first quantity.

5. The method according to claim 3, characterized in that, The cognitive load parameters are determined based on the average complexity and the test duration, including: Based on the average complexity and the test duration, a baseline cognitive load parameter is determined, which is the cognitive consumption rate of the target user during the test. The cognitive load parameters are obtained by correcting the baseline cognitive load parameters.

6. The method according to claim 5, characterized in that, The cognitive load parameters are obtained by correcting the baseline cognitive load parameters, including: Determine the second quantity corresponding to test scenarios with the same complexity; Based on the second quantity, an adjustment parameter is determined, and based on the adjustment parameter, the baseline cognitive load parameter is corrected to obtain the cognitive load parameter.

7. The method according to claim 1, characterized in that, Based on the test duration, determine the load threshold, including: When the test duration is less than or equal to the preset duration, the load threshold is determined as the first parameter; When the test duration exceeds the preset duration, the load threshold is determined based on the first parameter and the test duration.

8. The method according to claim 1, characterized in that, Determining the cognitive state of the target user includes: Determine the third quantity corresponding to complex test scenarios; When the third quantity is greater than or equal to the frequency threshold, and the cognitive load parameter is greater than the load threshold, the cognitive state is determined to be the load state.

9. The method according to claim 8, characterized in that, Determining the cognitive state of the target user includes: When the third quantity is less than the frequency threshold, or the cognitive load parameter is less than or equal to the load threshold, the cognitive state is determined to be the normal state.

10. A device for processing vehicle test data, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-9.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-9.

Images