An automated real vehicle cabin test method

Abstract

This invention relates to the field of vehicle testing technology, specifically disclosing an automated method for testing a real vehicle cockpit. This invention achieves closed-loop control throughout the entire process, from action execution and quality assessment to dynamic adjustment, improving the adaptability and accuracy of the testing system while reducing manual intervention costs. Furthermore, this invention offers high testing efficiency; testing can begin immediately after parameter changes and adjustments, and testing time is short, allowing for 24-hour testing. Compared to manual testing, it eliminates the need for a large investment in testing personnel and equipment. After the initial platform-based test cases are developed, only a small amount of manpower is required for adaptation, significantly saving costs. In addition, during the testing process, relevant images and audio files are saved on the computer and can be printed to the test report via CAPL scripts for easy viewing and problem analysis.

Description

Technical Field

[0002] This invention relates to the field of vehicle testing technology, specifically to an automated method for testing the cockpit of a real vehicle. Background Technology

[0004] Currently, traditional real vehicle cockpit testing always faces the problem of hardware and human resource costs. Each test requires the investment of multiple test vehicles, professional test equipment, and multiple test engineers. The human resource and equipment investment costs for long-term testing are high, and real vehicle testing needs to be carried out again after frequent updates and iterations of cockpit software. All these factors lead to the problem of long test cycles and high test costs.

[0005] Real-vehicle cockpit testing primarily relies on manual operation, requiring testers to execute test cases one by one, a tedious and time-consuming process. To address the surge in testing time caused by frequent cockpit software iterations, this invention constructs and implements an automated real-vehicle testing solution. The core of this solution lies in using ADB (Automatic Database), robotic arms, cameras, voice prompts, and microphones to replace manual operations, thereby automating cockpit testing and generating relevant test reports according to scripts. After deploying the automated scripts, functional testing can be performed quickly and accurately on different vehicle models and after version updates. Simultaneously, this invention can automatically verify the completion rate of the replaced manual operations, thus achieving self-checking to ensure test effectiveness.

[0006] In view of this, the present invention proposes an automated method for testing the cockpit of a real vehicle. Summary of the Invention

[0008] The purpose of this invention is to provide an automated method for testing the cockpit of a real vehicle, thereby solving the following technical problems:

[0009] How to solve the problem of increased testing time caused by frequent cockpit software iterations.

[0010] The objective of this invention can be achieved through the following technical solutions:

[0011] An automated method for testing a real vehicle cockpit includes:

[0012] S1. Test cases are called through the test control unit, and the multimodal interactive device is controlled to perform automated actions through the human-computer interaction simulation unit;

[0013] S2. Collect execution data for each automated action. The execution data includes at least one of the following: position deviation, response time, execution consistency, environmental interference, and equipment status.

[0014] S3. Input the execution data into the action scoring module to obtain the dimension score, and calculate the completion score of the current automated action based on the dimension score by weighted summation.

[0015] S4. Input the action completion score into the dynamic control module, and dynamically adjust the execution parameters and test strategy of the corresponding action device according to the score result;

[0016] S5. Re-execute the test process based on the adjusted parameters and generate a test report containing scoring data.

[0017] The above technical solution achieves closed-loop control throughout the entire process, from action execution and quality assessment to dynamic adjustment. This enhances the adaptability and accuracy of the testing system, reduces manual intervention costs, and provides high testing efficiency. The platform-based scripts can be developed before the project begins, and testing can commence immediately after parameter adjustments. Testing is short, allowing for 24 / 7 testing, eliminating time constraints caused by frequent version updates and significantly improving efficiency. Compared to manual testing, it eliminates the need for a large number of testers and equipment. After the initial platform-based test cases are developed, only a small amount of manpower is required for adaptation, greatly saving costs. Test case execution is unaffected by version updates, preventing overlooked issues even when the functionality is working correctly. Furthermore, during testing, relevant images and audio files are saved on the computer and can be printed to the test report via CAPL scripts for easy review and problem analysis.

[0018] Completion rating is based on the following formula:

[0019]

[0020] Where n is the total number of rating dimensions, and k is a non-zero natural number not greater than n. It is the preset weight value of the k-th rating dimension. It is the normalized score function for the k-th rating dimension. It is the independent variable of the normalized score function of the k-th rating dimension.

[0021] The above technical solution provides a configurable and scalable scoring model that supports flexible adjustment of evaluation dimensions and weights according to different testing scenarios.

[0022] The normalized score function includes a distance decay function based on position deviation, a time window function based on response time, an inverse variance function based on execution consistency, a linear compensation function based on environmental interference, and a periodic decay function based on device health status. The distance decay function is expressed by the formula:

[0023] ;

[0024] Obtain the distance decay function ,in It is the actual displacement vector. It is the target displacement vector. It is the preset standard comparison distance. To determine the norm of the difference between two actual displacement vectors and the target displacement vector;

[0025] The time window function is defined by the formula:

[0026]

[0027] Get Time Window Function ,in This is the actual response time. and These are the preset maximum and minimum response times, respectively.

[0028] The inverse variance function is expressed by the formula:

[0029] ;

[0030] Obtain the inverse variance function ,in The variance of the results from multiple executions;

[0031] The linear compensation function is expressed by the formula:

[0032] ;

[0033] Obtaining the linear compensation function ,in and These are the preset interference coefficients. This is a noise level index. Light level index;

[0034] The periodic decay function is expressed by the formula:

[0035] ;

[0036] Obtain the periodic decay function ,in The total available duration for the current usage period. This represents the duration of use in the current usage cycle.

[0037] Dynamic adjustments include:

[0038] If the score is below the first threshold, the execution speed will be reduced;

[0039] If the score is lower than the second threshold, a retry mechanism is triggered;

[0040] If the score remains below the third threshold, the test will be paused and a maintenance alarm will be triggered. The first, second, and third thresholds are preset values ​​set based on the completion score.

[0041] The above technical solution enables a tiered response mechanism, which ensures the continuity of the testing process and allows for timely identification and handling of abnormal situations.

[0042] The retry mechanism includes:

[0043] Choose a retry strategy based on the action type and the reason for failure;

[0044] Dynamically adjust device execution parameters based on historical retry data;

[0045] Perform retry actions and collect retry execution data;

[0046] Evaluate the effectiveness of the retry; if the requirements are still not met, proceed to the next round of retrying.

[0047] The retry strategy includes: adjusting the device execution parameters and re-executing the current action and recording the number of retries; if the number of retries exceeds the preset number, the action is determined to have failed and the reason for failure is recorded.

[0048] It should be noted that a maximum number of retries can be set, and the retry limit can be dynamically adjusted according to the importance of the action; the parameters, results and scores of each retry are recorded to form a retry knowledge base to help improve the retry scheme.

[0049] The above technical solution provides a controllable recovery mechanism to avoid test interruptions caused by occasional interference and improve test robustness.

[0050] The retry strategy also includes:

[0051] Parameter fine-tuning and retry: After adjusting the equipment's execution parameters, the original action is executed again;

[0052] Position offset retry: Reselect the execution position within the preset offset range;

[0053] Timing delay retry: retry after extending the action interval and inserting a waiting time;

[0054] Equipment switchover retry: Switch to backup equipment and perform the same action;

[0055] Environmental compensation retry: Adjust the execution intensity and frequency according to environmental interference.

[0056] The adjustment methods for the parameter fine-tuning retry include:

[0057] Adjust the clicking force of the robotic arm;

[0058] Voice output volume adjustment;

[0059] Camera exposure time adjustment;

[0060] Execution speed is scaled proportionally.

[0061] The methods for evaluating the effectiveness of retry include:

[0062] Compare the improvement in performance score before and after retrying;

[0063] Determine whether the preset pass threshold has been reached after retrying;

[0064] The effectiveness of retrying is assessed by combining the equipment's health status.

[0065] The multimodal interaction device includes:

[0066] Robotic arms are used to perform touch and button operations;

[0067] The mouthpiece and microphone are used to perform voice interaction tests;

[0068] A camera is used to capture images of the screen and identify the state of the interface.

[0069] The beneficial effects of this invention are:

[0070] This invention provides closed-loop control of the entire process from action execution and quality assessment to dynamic adjustment, which improves the adaptability and accuracy of the testing system, reduces the cost of manual intervention, and also has high testing efficiency. The platform-based scripts can be developed before the project starts, and can be directly tested after changing and debugging parameters, resulting in short testing time.

[0071] The scoring model of this invention is configurable and scalable, and supports flexible adjustment of evaluation dimensions and weights according to different test scenarios. Attached Figure Description

[0073] The invention will now be further described with reference to the accompanying drawings.

[0074] Figure 1 This is a flowchart of the testing and solution steps of the present invention. Detailed Implementation

[0076] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0077] Please see Figure 1 As shown, in one embodiment, an automated real-vehicle cockpit testing method is provided, including:

[0078] S1. Test cases are called through the test control unit, and the multimodal interactive device is controlled to perform automated actions through the human-computer interaction simulation unit;

[0079] S2. Collect execution data for each automated action. The execution data includes at least one of the following: position deviation, response time, execution consistency, environmental interference, and equipment status.

[0080] S3. Input the execution data into the action scoring module to obtain the dimension score. Calculate the completion score of the current automated action based on the dimension score using a weighted summation method. The weight value of each dimension score in the weighted summation process adopts a preset value and is set based on the action type of the corresponding automated action.

[0081] S4. Input the action completion score into the dynamic control module, and dynamically adjust the execution parameters and test strategy of the corresponding action device according to the score result;

[0082] S5. Re-execute the test process based on the adjusted parameters and generate a test report containing scoring data.

[0083] In this embodiment, an HMI control module can be integrated into the CANoe test project to send action commands to devices such as robotic arms and voice mouths via the HTTP protocol; during execution, data is collected in real time through cameras, microphones, and ADB; the scoring module calculates the score in real time based on a preset algorithm; if the score is lower than 0.7, a retry is triggered; and the final report is automatically generated by a CAPL script.

[0084] In addition, in-vehicle function testing can be divided into touch control testing and voice interaction testing.

[0085] Touch tests include:

[0086] The HMI software uses a camera to capture images of the vehicle's central control screen, analyzes the current screen status and coordinates, and then uses relevant scripts to move, click, and return the robotic arm to its original position. By repeatedly executing the recognition and clicking process, it simulates the effect of a human clicking the screen and can also detect screen response time—functions that are impossible for a human to perform. The robotic arm can also operate physical buttons on the steering wheel, windshield wipers, lights, air conditioning, and other in-vehicle controls using preset coordinates and movements.

[0087] The HMI software uses ADB to monitor the images on the vehicle's infotainment system in real time. It can compare pre-captured screenshots to see if the current image status meets expectations. For tests involving flashing warning lights in the instrument cluster, images are transmitted via a camera. The HMI software uses these images to detect the indicator light pattern, flashing frequency, and image color to determine the test results. The HMI also supports performance testing, including click response time, smoothness, and device startup time.

[0088] Voice interaction testing includes:

[0089] The HMI software integrates a voice engine to achieve TTS (Text-to-Speech) conversion, allowing users to speak through their mouths. It supports male / female voices, dialects, and Chinese / English pronunciation. Speech recognition and semantic recognition are performed via a microphone, and test results are output based on visual feedback and CAN signals. The software also allows for testing of wake-up rate, recognition rate, voice interaction, and voice control.

[0090] Finally, CANoe generates relevant test reports based on the test data transmitted by the HMI and the signal status detected in the actual vehicle.

[0091] To facilitate the simulation of actions such as getting in and out of a vehicle and closing doors during real-world testing, a test vehicle and robotic arm can be used to mimic a series of test procedures, including locking the vehicle when leaving, unlocking it when approaching, opening and closing the doors, starting the vehicle, and closing the doors.

[0092] Based on the above testing plan, platform-based scripts can be developed for different projects and vehicle models, and then adapted to the corresponding vehicle models. A basic project can be built first: add an HTTP protocol script to CANoe for data transmission between CANoe and the HMI software; write relevant test cases in the HMI software; add robotic arm execution parameters; and finally add the test scripts to CANoe. For different projects, simply add the relevant channels, import the database, uniformly replace the signal parameters in the test scripts, replace the test case images in the HMI software, and then debug.

[0093] The above technical solution achieves closed-loop control throughout the entire process, from action execution and quality assessment to dynamic adjustment. This enhances the adaptability and accuracy of the testing system, reduces manual intervention costs, and provides high testing efficiency. The platform-based scripts can be developed before the project begins, and testing can commence immediately after parameter adjustments. Testing is short, allowing for 24 / 7 testing, eliminating time constraints caused by frequent version updates and significantly improving efficiency. Compared to manual testing, it eliminates the need for a large number of testers and equipment. After the initial platform-based test cases are developed, only a small amount of manpower is required for adaptation, greatly saving costs. Test case execution is unaffected by version updates, preventing overlooked issues even when the functionality is working correctly. Furthermore, during testing, relevant images and audio files are saved on the computer and can be printed to the test report via CAPL scripts for easy review and problem analysis.

[0094] Completion rating is based on the following formula:

[0095]

[0096] Where n is the total number of rating dimensions, and k is a non-zero natural number not greater than n. It is the preset weight value of the k-th rating dimension. It is the normalized score function for the k-th rating dimension. It is the independent variable of the normalized score function of the k-th rating dimension.

[0097] The above technical solution provides a configurable and scalable scoring model that supports flexible adjustment of evaluation dimensions and weights according to different testing scenarios.

[0098] The normalized score function includes a distance decay function based on position deviation, a time window function based on response time, an inverse variance function based on execution consistency, a linear compensation function based on environmental interference, and a periodic decay function based on device health status. The distance decay function is expressed by the formula:

[0099] ;

[0100] Obtain the distance decay function ,in It is the actual displacement vector. It is the target displacement vector. It is the preset standard comparison distance. To determine the norm of the difference between two actual displacement vectors and the target displacement vector;

[0101] The time window function is defined by the formula:

[0102]

[0103] Get Time Window Function ,in This is the actual response time. and These are the preset maximum and minimum response times, respectively.

[0104] The inverse variance function is expressed by the formula:

[0105] ;

[0106] Obtain the inverse variance function ,in The variance of the results of multiple executions is recorded as 1 if the execution is consistent, otherwise it is recorded as 0.

[0107] The linear compensation function is expressed by the formula:

[0108] ;

[0109] Obtaining the linear compensation function ,in and These are the preset interference coefficients. This is a noise level index. The illumination level index is set based on the type of action being performed. For example, for actions using a camera, the interference coefficient of the noise level index is increased, while for actions using speech, the interference coefficient of the noise level index is increased. When the action is not affected by either interference, the index remains unchanged. and All are 0;

[0110] The periodic decay function is expressed by the formula:

[0111] ;

[0112] Obtain the periodic decay function ,in The total available duration for the current usage period. This represents the duration of use in the current usage cycle.

[0113] Dynamic adjustments include:

[0114] If the score is below the first threshold, the execution speed will be reduced;

[0115] If the score is lower than the second threshold, a retry mechanism will be triggered or the device will be switched to a backup device.

[0116] If the score remains below the third threshold, the test will be paused and a maintenance alarm will be triggered. The first, second, and third thresholds are preset values ​​set based on the completion score.

[0117] The above technical solution enables a tiered response mechanism, which ensures the continuity of the testing process and allows for timely identification and handling of abnormal situations.

[0118] The retry mechanism includes:

[0119] Choose a retry strategy based on the action type and the reason for failure;

[0120] Dynamically adjust device execution parameters based on historical retry data;

[0121] Perform retry actions and collect retry execution data;

[0122] Evaluate the effectiveness of the retry; if the requirements are still not met, proceed to the next round of retrying.

[0123] The retry strategy includes: adjusting the device execution parameters and re-executing the current action and recording the number of retries; if the number of retries exceeds the preset number, the action is determined to have failed and the reason for failure is recorded.

[0124] It should be noted that a maximum number of retries can be set, and the retry limit can be dynamically adjusted according to the importance of the action; the parameters, results and scores of each retry are recorded to form a retry knowledge base to help improve the retry scheme.

[0125] In this embodiment, the retry process may be as follows: if the touch click fails, select "parameter fine-tuning retry"; retrieve the parameters of successful retry of similar actions from the history database as the initial value; re-evaluate after execution, and if it is still lower than 0.6, trigger device switching.

[0126] The above technical solution provides a controllable recovery mechanism to avoid test interruptions caused by occasional interference and improve test robustness.

[0127] The retry strategy also includes:

[0128] Parameter fine-tuning and retry: After adjusting the equipment's execution parameters, the original action is executed again;

[0129] Position offset retry: Reselect the execution position within the preset offset range;

[0130] Timing delay retry: retry after extending the action interval and inserting a waiting time;

[0131] Equipment switchover retry: Switch to backup equipment and perform the same action;

[0132] Environmental compensation retry: Adjust the execution intensity and frequency according to environmental interference.

[0133] The adjustment methods for the parameter fine-tuning retry include:

[0134] Adjust the clicking force of the robotic arm;

[0135] Voice output volume adjustment;

[0136] Camera exposure time adjustment;

[0137] Execution speed is scaled proportionally.

[0138] For example, when the robotic arm fails to click, the parameters are first fine-tuned (e.g., force +10%). If it still fails, the position is shifted by ±2 pixels and retried. When voice wake-up fails, it is first delayed for 500ms and then retried by increasing the volume by 5dB. This multi-strategy approach covers various failure scenarios, improving the system's fault tolerance and test continuity.

[0139] The methods for evaluating the effectiveness of retry include:

[0140] Compare the improvement in performance score before and after retrying;

[0141] Determine whether the preset pass threshold has been reached after retrying;

[0142] By comprehensively assessing the effectiveness of retrying based on the equipment's health status, it is clear that the judgment criteria need to be adjusted over time as the equipment is used.

[0143] The multimodal interaction device includes:

[0144] Robotic arms are used to perform touch and button operations;

[0145] The mouthpiece and microphone are used to perform voice interaction tests;

[0146] A camera is used to capture images of the screen and identify the state of the interface.

[0147] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. An automated method for testing a real vehicle cockpit, characterized in that, Includes the following steps: S1. Test cases are called through the test control unit, and the multimodal interactive device is controlled to perform automated actions through the human-computer interaction simulation unit; S2. Collect execution data for each automated action. The execution data includes at least one of the following: position deviation, response time, execution consistency, environmental interference, and equipment status. S3. Input the execution data into the action scoring module to obtain the dimension score, and calculate the completion score of the current automated action based on the dimension score by weighted summation. S4. Input the action completion score into the dynamic control module, and dynamically adjust the execution parameters and test strategy of the corresponding action device according to the score result; S5. Re-execute the test process based on the adjusted parameters and generate a test report containing scoring data.

2. The automated vehicle cockpit testing method according to claim 1, characterized in that, In step S3, the completion score is calculated using the following formula: Where n is the total number of rating dimensions, and k is a non-zero natural number not greater than n. It is the preset weight value of the k-th rating dimension. It is the normalized score function for the k-th rating dimension. It is the independent variable of the normalized score function of the k-th rating dimension.

3. The automated vehicle cockpit testing method according to claim 2, characterized in that, The normalized score function includes a distance decay function based on position deviation, a time window function based on response time, an inverse variance function based on execution consistency, a linear compensation function based on environmental interference, and a periodic decay function based on device health status. The distance decay function is expressed by the formula: ； Obtain the distance decay function ,in It is the actual displacement vector. It is the target displacement vector. It is the preset standard comparison distance. To determine the norm of the difference between two actual displacement vectors and the target displacement vector, the time window function is calculated using the formula: Get Time Window Function ,in This is the actual response time. and These are the preset maximum and minimum response times, respectively. The inverse variance function is expressed by the formula: ； Obtain the inverse variance function ,in The variance of the results from multiple executions; The linear compensation function is expressed by the formula: ； Obtaining the linear compensation function ,in and These are the preset interference coefficients. This is a noise level index. Light level index; The periodic decay function is expressed by the formula: ； Obtain the periodic decay function ,in The total available duration for the current usage period. This represents the duration of use in the current usage cycle.

4. The automated vehicle cockpit testing method according to claim 1, characterized in that, In step S4, dynamic adjustment includes: If the score is below the first threshold, the execution speed will be reduced; If the score is lower than the second threshold, a retry mechanism is triggered; If the score remains below the third threshold, the test will be paused and a maintenance alarm will be triggered. The first, second, and third thresholds are preset values ​​set based on the completion score.

5. The automated vehicle cockpit testing method according to claim 4, characterized in that, The retry mechanism includes: Choose a retry strategy based on the action type and the reason for failure; Dynamically adjust device execution parameters based on historical retry data; Perform retry actions and collect retry execution data; Evaluate the effectiveness of the retry; if the requirements are still not met, proceed to the next round of retrying. The retry strategy includes: adjusting the device execution parameters and re-executing the current action and recording the number of retries; if the number of retries exceeds the preset number, the action is determined to have failed and the reason for failure is recorded.

6. The automated vehicle cockpit testing method according to claim 5, characterized in that, The retry strategy also includes: Parameter fine-tuning and retry: After adjusting the equipment's execution parameters, the original action is executed again; Position offset retry: Reselect the execution position within the preset offset range; Timing delay retry: retry after extending the action interval and inserting a waiting time; Equipment switchover retry: Switch to backup equipment and perform the same action; Environmental compensation retry: Adjust the execution intensity and frequency according to environmental interference.

7. The automated vehicle cockpit testing method according to claim 6, characterized in that, The adjustment methods for the parameter fine-tuning retry include: Adjust the clicking force of the robotic arm; Voice output volume adjustment; Camera exposure time adjustment; Execution speed is scaled proportionally.

8. The automated vehicle cockpit testing method according to claim 5, characterized in that, The methods for evaluating the effectiveness of retry include: Compare the improvement in performance score before and after retrying; Determine whether the preset pass threshold has been reached after retrying; The effectiveness of retrying is assessed by combining the equipment's health status.

9. The automated vehicle cockpit testing method according to claim 1, characterized in that, The multimodal interaction device includes: Robotic arms are used to perform touch and button operations; The mouthpiece and microphone are used to perform voice interaction tests; A camera is used to capture images of the screen and identify the state of the interface.

Images