Simulation evaluation methods for automotive infotainment automation testing equipment

Abstract

This invention discloses a simulation evaluation method for automotive infotainment system (IAS) automated testing equipment, belonging to the field of equipment simulation testing technology. The method includes collecting operational test data, processing it using a deep convolutional neural network model to obtain test task data, constructing a test action sequence, mapping the test action sequence to control commands, and optimizing the control commands according to a preset action rule library. The method drives the IAS automated testing equipment to execute the control commands to complete the test operation, records the operation response time and interface state change data during the test, calculates the motion trajectory deviation value, and calculates the test accuracy score. When the test accuracy score is less than a preset threshold, optimized parameters are added to the preset action rule library, and a new test is triggered. This invention uses the TCP / IP communication protocol to achieve precise control of the testing equipment and continuously optimizes the test strategy by dynamically updating the preset action rule library, thereby improving the efficiency of IAS automated testing.

Description

Technical Field

[0001] This invention relates to the field of equipment simulation testing technology, and in particular to a simulation evaluation method for vehicle automation testing equipment. Background Technology

[0002] Traditional vehicle infotainment system (VMS) testing relies heavily on manual operation, resulting in low efficiency, poor consistency, and high costs. In recent years, automated testing technologies have been increasingly applied to VMS testing, using machine vision and robotic arms to automate the testing process. However, existing automated VMS testing solutions often employ rule-based, fixed testing models, lacking in-depth analysis of test data and adaptive optimization capabilities, making it difficult to cope with complex and ever-changing testing scenarios and the differentiated requirements of different vehicle models. Furthermore, due to the complexity of the VMS interface and the precision requirements of operational trajectories, existing testing equipment frequently exhibits inaccurate operations and response delays when executing testing tasks, affecting the reliability of test results.

[0003] Currently, the main technical challenges in the field of automated vehicle infotainment testing are as follows: First, the intelligent recognition and action planning capabilities of test tasks are insufficient, making it difficult to accurately understand interface states and generate reasonable test sequences; second, the execution accuracy and real-time control of test actions are inadequate, failing to guarantee the stability and consistency of test operations; and third, the lack of effective test quality evaluation and optimization feedback mechanisms makes it difficult to continuously improve test results. These problems severely restrict the application and promotion of automated vehicle infotainment testing technology.

[0004] To address the aforementioned issues, this invention proposes a simulation evaluation method for vehicle-mounted automated testing equipment. This method achieves intelligent identification of test tasks through a deep convolutional neural network and, combined with a multi-dimensional scoring model and dynamic optimization strategy, effectively improves the accuracy and reliability of test execution. Summary of the Invention

[0005] In view of the problems of inaccurate test task identification, insufficient action execution accuracy and lack of adaptive optimization capability in existing vehicle-mounted automated testing equipment, this invention is proposed.

[0006] Therefore, the problem to be solved by this invention is how to achieve intelligent identification and adaptive execution of vehicle system testing tasks, and how to establish a scientific test evaluation and optimization mechanism, so as to improve the accuracy and reliability of vehicle system automated testing.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] In a first aspect, embodiments of the present invention provide a simulation evaluation method for vehicle infotainment system automated testing equipment, comprising: collecting operational test data of the vehicle infotainment system interface to be tested, and processing the operational test data through a deep convolutional neural network model to obtain test task data; constructing a test action sequence based on the test task data, mapping the test action sequence to control instructions, and optimizing and adjusting the control instructions according to a preset action rule library; driving the vehicle infotainment system automated testing equipment to execute the control instructions to complete the test operation, and recording the operation response time and interface state change data during the test process, and calculating the motion trajectory deviation value; calculating a test accuracy score based on the operation response time, interface state change data, and motion trajectory deviation value, and when the test accuracy score is less than a preset threshold, supplementing the preset action rule library with optimization parameters and triggering a new test.

[0009] As a preferred embodiment of the simulation evaluation method for vehicle automation testing equipment described in this invention, the method for calculating the test accuracy score is as follows: The operation response time is normalized using a weighted summation method, and a time deviation score is calculated based on the standard response time; a deep learning image matching algorithm is used to compare the interface state images before and after the test, calculate the structural similarity index and local feature point matching degree, and generate an interface consistency score; based on the motion trajectory deviation value, the root mean square error method is used to evaluate the fitting degree between the actual trajectory and the standard trajectory, and generate a trajectory accuracy score; the time deviation score, the interface consistency score, and the trajectory accuracy score are fused using a multi-dimensional weighted model to obtain the test accuracy score; the test accuracy score is compared with a preset threshold, the specific dimensions causing the score reduction are analyzed, and optimization parameters for the corresponding dimensions are extracted; the optimization parameters are updated to a preset action rule library, automatically adjusting the execution speed, waiting time, and operation intensity of the test action, and triggering a new round of testing until the test accuracy score reaches the preset threshold requirement.

[0010] As a preferred embodiment of the simulation evaluation method for vehicle automation testing equipment described in this invention, the specific formula for the test accuracy scoring is as follows:

[0011] ;

[0012] in, To score the accuracy of the test, The time-based scoring weighting coefficient. This is the weighting coefficient for the interface rating. For trajectory scoring weighting coefficients, Let i be the actual response time of the i-th operation. For standard response time, For the total number of operations, Let be the structural similarity before and after the i-th interface state. For the interface state image data before performing the test operation at the i-th test point, For the interface state image data after performing the test operation at the i-th test point, The number of interface status monitoring points. This represents the deviation value of the motion trajectory.

[0013] When the test accuracy score is less than the preset threshold, if there is only a single dimension score abnormality, the system will enter the rapid optimization mode; if there are two or more dimensions score abnormalities at the same time, the system will enter the collaborative optimization mode; when the test accuracy score is greater than or equal to the preset threshold, the adjustment range of the current optimal parameters and the initial parameters will be calculated respectively, and the current optimal parameter combination will be updated as the benchmark parameters for this type of test scenario.

[0014] As a preferred embodiment of the simulation evaluation method for vehicle automation testing equipment described in this invention, the method includes: driving the vehicle automation testing equipment to execute the control commands to complete the test operation, recording the operation response time and interface state change data during the test process, and calculating the motion trajectory deviation value. This includes: sending an optimized control command sequence to the main controller of the vehicle automation testing equipment according to the TCP / IP communication protocol, wherein the main controller parses the command parameters and drives a multi-degree-of-freedom robotic arm to execute the corresponding test operation; setting a microsecond-level timer to record the sending time and actual response time of each control command, calculating the command execution delay time, and writing the operation response time data into the test database in real time; using a camera module to collect interface images during the test process in real time, extracting interface state feature parameters, and recording interface state change data according to a time sequence; using a laser displacement sensor to collect the spatial coordinate data of the robotic arm's end effector in real time, comparing the actual motion trajectory with the theoretical trajectory, eliminating measurement noise through a Kalman filter algorithm, and calculating the motion trajectory deviation value.

[0015] As a preferred embodiment of the simulation evaluation method for vehicle automation testing equipment described in this invention, the specific formula for the motion trajectory deviation value is as follows:

[0016] ;

[0017] in, This represents the deviation value of the motion trajectory. Let be the position function of the actual trajectory in the i-th dimension. Let i be the position function of the theoretical trajectory in the i-th dimension. To test the time variable, Let be the weighting coefficient for the j-th measurement point. Let j be the instantaneous error of the j-th measurement point. The total number of measurement points. This is the Hessian matrix, used to evaluate trajectory curvature. This is a rotation matrix used for attitude deviation compensation. For matrix determinant operations, This is the trace operation of a matrix.

[0018] As a preferred embodiment of the simulation evaluation method for vehicle automation testing equipment described in this invention, the method includes: constructing a test action sequence based on the test task data, mapping the test action sequence to control instructions, and optimizing the control instructions according to a preset action rule library. This includes: decomposing the test task data into basic action units based on a depth-first search algorithm, constructing a directed acyclic graph according to operational logic dependencies, and generating a test action sequence, wherein the basic action units include touch operations, swipe operations, long-press operations, and multi-touch operations; converting the test action sequence into corresponding control instructions according to an action mapping table, wherein the action mapping table includes action type, coordinate position, execution duration, and triggering conditions; reading optimization rules from the preset action rule library and verifying the instruction parameters of the control instructions, wherein the optimization rules include action interval thresholds, triggering condition constraints, and conflict detection rules; and using a genetic algorithm to optimize the parameters of the control instruction sequence, iteratively optimizing instruction parameters that do not meet the optimization rules to generate an optimized control instruction sequence.

[0019] As a preferred embodiment of the simulation evaluation method for vehicle infotainment system automation testing equipment described in this invention, the method for acquiring test task data is as follows: A high-resolution camera array is used to collect multi-angle data of the vehicle infotainment system interface to be tested, acquiring runtime test data of the interface display content, control layout, and text information; the runtime test data is input into a deep convolutional neural network model, wherein the neural network model adopts a VGG-16 network structure and extracts hierarchical features of the interface image through several convolutional and pooling layers; an attention mechanism is set in the fully connected layer of the neural network model to perform object detection and semantic segmentation on the button area, text area, and input box area of ​​the vehicle infotainment system interface to be tested; the results of object detection and semantic segmentation are labeled to generate test task data, wherein the test task data includes control type, position information, and operation attributes.

[0020] In a second aspect, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program instructions, when executed by the processor, implement the steps of the simulation evaluation method for vehicle automation testing equipment as described in the first aspect of the present invention.

[0021] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of the simulation evaluation method for vehicle automation testing equipment as described in the first aspect of the present invention.

[0022] The beneficial effects of this invention are as follows: By employing a deep convolutional neural network model for intelligent recognition and feature extraction of the vehicle interface, combined with an attention mechanism, precise positioning of test tasks is achieved, effectively overcoming the limitations of traditional manually defined features; a directed acyclic graph of test action sequences is constructed using a depth-first search algorithm, and parameters are optimized using a genetic algorithm, significantly improving the execution efficiency and reliability of test instructions; multi-sensor fusion technology and Kalman filtering algorithm are used for real-time monitoring and data processing of the test process, combined with microsecond-level response time recording and laser displacement measurement, achieving high-precision test trajectory tracking and evaluation; simultaneously, a multi-dimensional scoring model based on time deviation, interface consistency, and trajectory accuracy is established, and adaptive adjustment of test parameters is achieved through two modes: rapid optimization and collaborative optimization, ensuring the accuracy and repeatability of test results. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0024] Figure 1 This is a flowchart of the simulation evaluation method for vehicle automation testing equipment in Example 1. Detailed Implementation

[0025] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0026] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0027] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0028] Example 1

[0029] Reference Figure 1 This is the first embodiment of the present invention, which provides a simulation evaluation method for vehicle automation testing equipment, including,

[0030] S1: Collect the running test data of the vehicle's infotainment system interface to be tested, and use a deep convolutional neural network model to identify and process the running test data to obtain test task data.

[0031] Specifically, the test task data acquisition method is to use a high-resolution camera array to collect multi-angle data on the vehicle's interface to be tested, and obtain the running test data of the interface display content, control layout and text information.

[0032] It should be noted that the camera array includes three fixed shooting positions: front view, left view, and right view.

[0033] Furthermore, the test data is input into a deep convolutional neural network model, which adopts the VGG-16 network structure and extracts hierarchical features of the interface image through several convolutional and pooling layers. An attention mechanism is set in the fully connected layer of the neural network model to perform object detection and semantic segmentation on the button area, text area and input box area of ​​the vehicle system test interface.

[0034] Furthermore, the results of object detection and semantic segmentation are labeled to generate test task data, which includes control type, location information, and operation attributes.

[0035] S2: Construct a test action sequence based on the test task data, map the test action sequence to control instructions, and optimize and adjust the control instructions according to the preset action rule library.

[0036] Specifically, the test task data is decomposed into basic action units based on the depth-first search algorithm, and a directed acyclic graph is constructed according to the operational logic dependencies to generate a sequence of test actions. The basic action units include touch operation, swipe operation, long press operation and multi-touch operation.

[0037] Furthermore, the test action sequence is converted into corresponding control instructions based on the action mapping table, which includes action type, coordinate position, execution duration, and triggering conditions. Optimization rules from the preset action rule library are read, and the instruction parameters of the control instructions are verified. The optimization rules include action interval thresholds, triggering condition constraints, and conflict detection rules.

[0038] It should be noted that the following rules apply to action interval thresholds: the minimum interval between consecutive touch operations should be ≥200ms to avoid accidental touches; the waiting time after a swipe operation should be ≥500ms to ensure the completion of the interface animation; the minimum duration of a long press operation should be ≥1000ms to ensure the long press effect is triggered; the touch point press time difference for multi-touch operations should be ≤50ms to ensure synchronization; and the interval between switching between different control areas should be ≥300ms to ensure state updates. Triggering condition constraints apply: before a swipe operation, it is necessary to check whether the target area supports the swipe attribute; before a long press operation, it is necessary to verify whether the control has a long press trigger function; for multi-touch operations, it is necessary to confirm whether the interface supports multi-touch mode; operations on pop-up controls must wait until the pop-up is fully displayed before execution; the execution of subsequent operation instructions must be paused during animation transitions. Conflict detection rules apply: multiple touch operations are prohibited from being executed simultaneously in the same area; swipe operations and long press operations cannot be executed in parallel in overlapping areas; multi-touch operations cannot be performed simultaneously with other types of operations; continuous operations on mutually exclusive controls must be avoided; and the execution areas of adjacent instructions in the operation sequence must not overlap spatially.

[0039] Furthermore, a genetic algorithm is used to optimize the parameters of the control command sequence, iteratively optimizing the command parameters that do not meet the optimization rules to generate an optimized control command sequence.

[0040] S3: Drive the vehicle's automated testing equipment to execute control commands to complete the test operation, record the operation response time and interface status change data during the test, and calculate the motion trajectory deviation value.

[0041] Specifically, the optimized control command sequence is sent to the main controller of the vehicle automation testing equipment according to the TCP / IP communication protocol. The main controller then parses the command parameters and drives the multi-degree-of-freedom robotic arm to perform the corresponding test operations.

[0042] Furthermore, a microsecond-level timer is set to record the sending time and actual response time of each control command, calculate the command execution delay time, and write the operation response time data into the test database in real time; the camera module is used to collect interface images in real time during the test process, extract interface state feature parameters, and record them in time series to form interface state change data.

[0043] Furthermore, a laser displacement sensor is used to collect the spatial coordinate data of the robotic arm's end effector in real time. The actual motion trajectory is compared with the theoretical trajectory, and a Kalman filter algorithm is used to eliminate measurement noise and calculate the motion trajectory deviation value. The specific formula is as follows:

[0044] ;

[0045] in, This represents the deviation value of the motion trajectory. Let be the position function of the actual trajectory in the i-th dimension. Let i be the position function of the theoretical trajectory in the i-th dimension. To test the time variable, Let be the weighting coefficient for the j-th measurement point. Let j be the instantaneous error of the j-th measurement point. The total number of measurement points. This is the Hessian matrix, used to evaluate trajectory curvature. This is a rotation matrix used for attitude deviation compensation. For matrix determinant operations, This is the trace operation of a matrix.

[0046] S4: Calculate the test accuracy score based on the operation response time, interface state change data and motion trajectory deviation value. When the test accuracy score is less than the preset threshold, add optimization parameters to the preset action rule library and trigger a new test.

[0047] Specifically, a weighted summation method is used to normalize the operation response time, and the time deviation score is calculated based on the standard response time. A deep learning image matching algorithm is used to compare the interface state images before and after the test, calculate the structural similarity index and the matching degree of local feature points, and generate an interface consistency score. Based on the motion trajectory deviation value, the root mean square error method is used to evaluate the degree of fit between the actual trajectory and the standard trajectory, and generate a trajectory accuracy score.

[0048] Furthermore, the time deviation score, interface consistency score, and trajectory accuracy score are fused using a multi-dimensional weighted model to derive the test accuracy score, as shown in the following formula:

[0049] ;

[0050] in, To score the accuracy of the test, The time-based scoring weighting coefficient. This is the weighting coefficient for the interface rating. For trajectory scoring weighting coefficients, Let i be the actual response time of the i-th operation. For standard response time, For the total number of operations, Let be the structural similarity before and after the i-th interface state. For the interface state image data before performing the test operation at the i-th test point, For the interface state image data after performing the test operation at the i-th test point, The number of interface status monitoring points. This represents the deviation value of the motion trajectory.

[0051] It should be noted that the multidimensional weighted model is constructed based on three dimensions: operation response time deviation score, interface state consistency score, and motion trajectory accuracy score.

[0052] Furthermore, the test accuracy score and preset threshold are compared to analyze the specific dimensions that led to the score decrease and extract the corresponding optimization parameters. When the test accuracy score is less than the preset threshold, if only a single dimension has an abnormal score, a rapid optimization mode is entered; if two or more dimensions have abnormal scores simultaneously, a collaborative optimization mode is entered. When the test accuracy score is greater than or equal to the preset threshold, the adjustment range of the current optimal parameters and the initial parameters are calculated separately, and the current optimal parameter combination is updated as the baseline parameters for this type of test scenario.

[0053] It should be noted that the preset threshold is determined based on a comprehensive analysis of the probability density distribution characteristics of historical test data, the sensitivity thresholds of scores in each dimension, and the complexity of the test scenario.

[0054] Specifically, the rapid optimization mode involves calculating the deviation rate between the score of the abnormal dimension and the historical mean. When the deviation rate is between 20% and 50%, the abnormality is marked as a mild abnormality, and only the core parameters of this dimension are fine-tuned. For example, if a mild abnormality occurs in the time dimension, only the waiting time parameter is adjusted; if a mild abnormality occurs in the interface consistency, only the image acquisition interval is adjusted; if a mild abnormality occurs in the trajectory accuracy, only the execution speed is adjusted. If the deviation rate exceeds 50%, the abnormality is marked as a severe abnormality, and all relevant parameters of this dimension are adjusted simultaneously. For example, if a severe abnormality occurs in the time dimension, both the waiting time and the number of retries are adjusted; if a severe abnormality occurs in the interface consistency, both the image acquisition interval and the contrast sensitivity are adjusted; if a severe abnormality occurs in the trajectory accuracy, both the execution speed is adjusted. The effect of each parameter adjustment is evaluated, and the optimal parameter combination is gradually approximated using a binary search method until the score of this dimension recovers to above the preset threshold.

[0055] Furthermore, the collaborative optimization model includes constructing a dimensional correlation matrix to analyze the mutual influence relationships between abnormal dimensions; calculating the contribution weight of each dimension using principal component analysis to determine the optimization order; if the correlation coefficient between two dimensions is greater than 0.7, these two dimensions are classified as a strongly correlated group, and a linked optimization strategy is adopted. For example, when the time dimension and the interface consistency dimension are both abnormal and strongly correlated, the waiting time and image acquisition interval are adjusted synchronously according to the correlation characteristics; if the correlation coefficient between dimensions is less than 0.3, they are classified as a weakly correlated group, and an independent optimization strategy is adopted, optimizing sequentially according to the contribution weight from high to low. For moderately correlated groups with correlation coefficients between 0.3 and 0.7, a hybrid optimization strategy is adopted. Through iterative optimization, the parameter combination is continuously adjusted until the scores of all dimensions simultaneously reach the preset threshold.

[0056] Furthermore, the optimization parameters are updated to the preset action rule library, automatically adjusting the execution speed, waiting time, and operational intensity of the test actions, and triggering a new round of testing until the test accuracy score reaches the preset threshold requirement.

[0057] This embodiment also provides a computer device applicable to the simulation evaluation method for vehicle automation testing equipment, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the simulation evaluation method for vehicle automation testing equipment as proposed in the above embodiment.

[0058] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0059] In summary, this invention employs a deep convolutional neural network model for intelligent recognition and feature extraction of the vehicle interface, combined with an attention mechanism to achieve precise positioning of test tasks, effectively overcoming the limitations of traditional manually defined features. It constructs a directed acyclic graph of test action sequences using a depth-first search algorithm and optimizes parameters using a genetic algorithm, significantly improving the execution efficiency and reliability of test commands. Furthermore, it utilizes multi-sensor fusion technology and a Kalman filter algorithm for real-time monitoring and data processing of the test process, combined with microsecond-level response time recording and laser displacement measurement, to achieve high-precision test trajectory tracking and evaluation. Simultaneously, it establishes a multi-dimensional scoring model based on time deviation, interface consistency, and trajectory accuracy, and achieves adaptive adjustment of test parameters through both rapid optimization and collaborative optimization modes, ensuring the accuracy and repeatability of test results.

[0060] Example 2

[0061] Referring to Table 1, which is the second embodiment of the present invention, this embodiment provides a simulation evaluation method for vehicle automation testing equipment. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0062] Specifically, a 2024 model intelligent vehicle infotainment system from a certain brand was selected as the test subject. This system features a 12.3-inch high-definition touchscreen display and runs on the Android Automotive operating system. The test equipment included: an automated testing platform equipped with a 6-axis robotic arm (accuracy ±0.02mm), a 4K high-speed industrial camera array (200fps sampling rate), a 16-channel laser displacement sensor (sampling accuracy 0.01mm), and a high-performance industrial computer (equipped with an RTX 4080 graphics card).

[0063] Furthermore, a high-resolution camera array is used to acquire data from the vehicle's infotainment interface. The camera array is distributed in a 120° fan shape to ensure complete coverage of the display screen surface. The acquired image data is preprocessed and then input into an optimized VGG-16 network. This network, with 5 convolutional blocks and 3 fully connected layers, combined with an improved attention mechanism, achieves a control recognition accuracy of 98.7%. Based on the recognition results, the system automatically generates a test task dataset containing 374 test points.

[0064] Furthermore, an improved depth-first search algorithm was used to decompose the test task, generating 2183 basic action units, including single-point touch, multi-point touch, swipe, and long-press operation types. These action units were then converted into standard control commands by a motion mapping engine and optimized using a pre-defined action rule library. The rule library contains 186 optimization rules summarized from extensive experiments, covering the optimal execution parameters for different operation types.

[0065] Specifically, during the test execution phase, the system sends optimized control commands to the test platform via Gigabit Ethernet. The main controller employs a real-time operating system with an interrupt response time of <10μs, ensuring the real-time execution of commands. Simultaneously, a laser displacement sensor array records the robotic arm's motion trajectory at a sampling rate of 1kHz. Noise is processed using an improved Kalman filter algorithm, achieving a trajectory reconstruction accuracy of ±0.05mm.

[0066] Furthermore, as shown in Table 1, a comparison of the data clearly demonstrates that the method of this invention significantly outperforms the traditional manual testing method in all performance indicators. Specific analysis is as follows: In terms of control recognition accuracy, the method of this invention achieves a high accuracy of 98.7%, an improvement of 6.4 percentage points compared to the 92.3% of the traditional manual testing method. This improvement is mainly attributed to the innovative application of the deep convolutional neural network model and attention mechanism used in this invention, enabling precise recognition and positioning of vehicle interface controls. The average response time indicator shows that the method of this invention reduces the response time to 98ms, a reduction of 147ms compared to the traditional method's 245ms, representing a 60% improvement in response speed. This significant improvement is attributed to the optimized TCP / IP communication protocol and precise control of the microsecond-level timer used in this invention, greatly enhancing the real-time performance of the testing system.

[0067] Table 1 Comparison between the method of the present invention and traditional methods

[0068] Performance indicators Traditional manual testing methods Method of the present invention Control recognition accuracy 92.3% 98.7% Average response time 245ms 98ms Interface consistency score 0.882 0.978 trajectory accuracy deviation 0.43mm 0.12mm Test coverage 78.5% 95.8% Test efficiency 12 times / hour 96 times / hour Overall score 0.823 0.947

[0069] Furthermore, in terms of interface consistency score, the method of this invention achieved a high score of 0.978, a 10.9% improvement over the traditional method's 0.882. This reflects that the present invention, through deep learning image matching algorithm, can more accurately evaluate changes in interface state during the test, ensuring the reliability of the test results. Trajectory accuracy deviation data shows that the method of this invention controls the deviation to 0.12mm, a 72.1% reduction compared to the traditional method's 0.43mm. This improvement in accuracy stems from the real-time monitoring by the laser displacement sensor and the noise processing technology of the Kalman filter algorithm used in this invention, ensuring high-precision execution of the test operation.

[0070] Specifically, regarding test coverage, the method of this invention achieves 95.8% coverage, an improvement of 17.3 percentage points compared to the traditional method's 78.5%. This indicates that the invention can more comprehensively cover test scenarios, significantly improving the completeness and reliability of the tests. In terms of test efficiency, the method of this invention can complete 96 tests per hour, eight times that of the traditional method's 12 tests per hour. This significant improvement in efficiency is attributed to the high degree of automation and optimized testing strategy of this invention, greatly improving the time efficiency of the testing process. Comprehensive scoring data shows that the method of this invention achieved a high score of 0.947, an improvement of 15.1% compared to the traditional method's 0.823. This comprehensive indicator fully demonstrates the superior overall performance of this invention.

[0071] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A simulation evaluation method for vehicle automation testing equipment, characterized in that: include, The system collects operational test data from the vehicle's infotainment system interface to be tested, and then uses a deep convolutional neural network model to identify and process the operational test data to obtain test task data. A test action sequence is constructed based on the test task data, the test action sequence is mapped to control instructions, and the control instructions are optimized and adjusted according to a preset action rule library. The vehicle-mounted automated testing equipment executes the control commands to complete the test operation, records the operation response time and interface status change data during the test, and calculates the motion trajectory deviation value. Based on the operation response time, interface state change data, and motion trajectory deviation value, the test accuracy score is calculated. When the test accuracy score is less than the preset threshold, the optimized parameters are added to the preset action rule library and a new test is triggered. The method for calculating the test accuracy score is as follows: The operation response time is normalized using a weighted summation method, and the time deviation score is calculated based on the standard response time. By using deep learning image matching algorithms to compare interface state images before and after testing, structural similarity index and local feature point matching degree are calculated to generate interface consistency score. Based on the motion trajectory deviation value, the root mean square error method is used to evaluate the degree of fit between the actual trajectory and the standard trajectory, and generate a trajectory accuracy score. The time deviation score, the interface consistency score, and the trajectory accuracy score are fused using a multi-dimensional weighted model to obtain the test accuracy score. The test accuracy score and preset threshold are compared to determine the specific dimensions that lead to the score reduction, and the optimization parameters for the corresponding dimensions are extracted. The optimized parameters are updated to the preset action rule library, the execution speed, waiting time and operation intensity of the test action are automatically adjusted, and a new round of testing is triggered until the test accuracy score reaches the preset threshold requirement. The specific formula for scoring the test accuracy is as follows: ； in, To score the accuracy of the test, The time-based scoring weighting coefficient. This is the weighting coefficient for the interface rating. For trajectory scoring weighting coefficients, Let i be the actual response time of the i-th operation. For standard response time, For the total number of operations, Let be the structural similarity before and after the i-th interface state. For the interface state image data before performing the test operation at the i-th test point, For the interface state image data after performing the test operation at the i-th test point, This represents the number of interface status detection points. This represents the deviation value of the motion trajectory. When the test accuracy score is less than the preset threshold, if there is only a single dimension score abnormality, it will enter the rapid optimization mode; if there are two or more dimensions score abnormalities at the same time, it will enter the collaborative optimization mode. When the test accuracy score is greater than or equal to the preset threshold, the adjustment range of the current optimal parameter and the initial parameter are calculated respectively, and the current optimal parameter combination is updated as the benchmark parameter for this type of test scenario.

2. The simulation evaluation method for vehicle automation testing equipment as described in claim 1, characterized in that: The automated testing equipment for the vehicle's infotainment system executes the control commands to complete the test operation, records the operation response time and interface state change data during the test, and calculates the motion trajectory deviation value, including: According to the TCP / IP communication protocol, the optimized control command sequence is sent to the main controller of the vehicle automation test equipment, where the main controller parses the command parameters and drives the multi-degree-of-freedom robotic arm to perform the corresponding test operation. Set a microsecond-level timer to record the sending time and actual response time of each control command, calculate the command execution delay time, and write the operation response time data into the test database in real time; The interface images during the test process are collected in real time using a camera module, the interface state feature parameters are extracted, and the interface state change data are recorded in a time series. A laser displacement sensor is used to collect the spatial coordinate data of the end effector of the robotic arm in real time. The actual motion trajectory is compared with the theoretical trajectory. The Kalman filter algorithm is used to eliminate measurement noise and calculate the motion trajectory deviation value.

3. The simulation evaluation method for vehicle automation testing equipment as described in claim 2, characterized in that: The specific formula for the motion trajectory deviation value is as follows: ； in, This represents the deviation value of the motion trajectory. Let be the position function of the actual trajectory in the i-th dimension. Let i be the position function of the theoretical trajectory in the i-th dimension. To test the time variable, Let be the weighting coefficient for the j-th measurement point. Let j be the instantaneous error of the j-th measurement point. The total number of measurement points. This is the Hessian matrix, used to evaluate trajectory curvature. This is a rotation matrix used for attitude deviation compensation. For matrix determinant operations, This is the trace operation of a matrix.

4. The simulation evaluation method for vehicle automation testing equipment as described in claim 3, characterized in that: A test action sequence is constructed based on the test task data, the test action sequence is mapped to control commands, and the control commands are optimized and adjusted according to a preset action rule base, including: The test task data is decomposed into basic action units based on the depth-first search algorithm, and a directed acyclic graph is constructed according to the operational logic dependency to generate a test action sequence. The basic action units include touch operation, swipe operation, long press operation and multi-touch operation. The test action sequence is converted into corresponding control instructions according to the action mapping table, wherein the action mapping table includes action type, coordinate position, execution duration and triggering conditions; The system reads optimization rules from a preset action rule library and verifies the command parameters of the control command, wherein the optimization rules include action interval thresholds, trigger condition constraints, and conflict detection rules. A genetic algorithm is used to optimize the parameters of the control command sequence. The command parameters that do not meet the optimization rules are iteratively optimized to generate an optimized control command sequence.

5. The simulation evaluation method for vehicle automation testing equipment as described in claim 4, characterized in that: The method for obtaining the test task data is as follows: A high-resolution camera array is used to collect multi-angle data on the vehicle's infotainment interface to obtain operational test data on the interface display content, control layout, and text information. The test data is input into a deep convolutional neural network model, wherein the neural network model adopts a VGG-16 network structure and extracts hierarchical features of the interface image through several convolutional and pooling layers. An attention mechanism is set in the fully connected layer of the neural network model to perform object detection and semantic segmentation on the button area, text area and input box area of ​​the vehicle system test interface; The results of object detection and semantic segmentation are labeled to generate test task data, which includes control type, location information and operation attributes.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the simulation evaluation method for vehicle automation testing equipment as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the simulation evaluation method for vehicle automation testing equipment as described in any one of claims 1 to 5.

Images