Test method for automobile accelerator pedal misuse suppression system

Abstract

The invention relates to the technical field of vehicle testing, in particular to a testing method for an automobile accelerator pedal misuse suppression system, which comprises the following steps of: establishing an accelerator pedal misapplication accident data set according to a traffic accident depth survey database; according to the established accelerator pedal misuse accident data set, automatically extracting corresponding test parameters when the accelerator pedal is misused; according to the extracted test parameters, significant correlation analysis is carried out on the test parameters, key parameters corresponding to the severity of the accident affecting the mistaken stepping of the accelerator pedal are matched, and a corresponding test matrix is designed; according to the parameters corresponding to the different test matrixes, on the basis of a preset test strategy of the automobile accelerator pedal misuse suppression system, executing the test strategy on the test vehicle, and outputting a corresponding test result; and according to the output test result, comprehensively analyzing the test result, and outputting a final analysis result.

Description

Technical Field

[0001] The invention relates to the technical field of vehicle testing, and in particular to a testing method for a vehicle accelerator pedal misuse suppression system. Background Art

[0002] In recent years, with the rapid development of new energy vehicles and advanced driver assistance systems (ADAS), pedal misoperation accidents, especially accidental accelerator pedal pressing, have become a new challenge in the field of traffic safety that cannot be ignored. Such accidents often arise from drivers' unfamiliarity with the operating characteristics of modern vehicles, especially in single-pedal driving mode, and improper handling of vehicle control in emergency situations. Such accidents occur frequently, often causing serious casualties, becoming a hot topic and triggering widespread discussion. The society is extremely concerned about such traffic safety issues.

[0003] To address this challenge, automakers have taken the lead in integrating pedal misapplication mitigation technologies, which aim to reduce the frequency of pedal misapplication accidents in the first place.

[0004] However, there is currently a lack of a unified and standardized test method and dedicated test equipment to comprehensively evaluate the functional integrity and scenario adaptability of the accelerator pedal misapplication suppression system. This technical gap not only hinders the standardization process of related safety technologies, but also limits the industry's objective verification and continuous optimization of the vehicle's accelerator pedal misapplication suppression capabilities. Summary of the invention

[0005] One of the purposes of the present invention is to provide a testing method for an automobile accelerator pedal misuse suppression system, which can realize reliable and scientific testing of the automobile accelerator pedal misuse suppression system, filling the gap in the automobile industry's accelerator pedal misuse suppression system functional testing.

[0006] In order to achieve the above object, a test method for a vehicle accelerator pedal misuse suppression system is provided, comprising the following steps:

[0007] S1. Establish an accelerator pedal mis-pressing accident data set based on the traffic accident in-depth investigation database;

[0008] S2. Automatically extracting test parameters corresponding to the accelerator pedal misuse according to the established accelerator pedal misuse accident data set;

[0009] S3. Based on the extracted test parameters, perform significant correlation analysis on the test parameters, match the key parameters corresponding to the severity of the accident affecting the accident of mis-pressing the accelerator pedal, and design a corresponding test matrix;

[0010] S4. According to the parameters corresponding to different test matrices, based on the preset test strategy of the automobile accelerator pedal misuse suppression system, the test strategy is executed on the test vehicle, and the corresponding test results are output;

[0011] S5. Based on the output test results, conduct a comprehensive analysis on the test results and output the final analysis results.

[0012] Technical principles and effects of this solution: In this solution, we first establish an accelerator pedal accident data set through the traffic accident in-depth investigation database. Through this step, the huge data in the traffic accident in-depth investigation database is divided, so that the extracted data are all related to the accelerator pedal accident, which greatly improves the reliability of the data and provides a solid foundation for subsequent analysis and testing.

[0013] Then, based on the established accelerator pedal accident data set, the test parameters corresponding to the accelerator pedal misuse are automatically extracted. By automatically extracting the test parameters related to the misuse event, the speed and accuracy of data processing are improved, human errors are reduced, and the effectiveness of subsequent analysis is ensured.

[0014] Then, according to the extracted test parameters, the test parameters are analyzed for significant correlation, the key parameters corresponding to the severity of accidents affecting the mis-stepping of the accelerator pedal are matched, and the corresponding test matrix is ​​designed; through the significant correlation analysis, it is determined which parameters have the greatest impact on the mis-stepping of the accelerator pedal, which helps to focus on the truly critical factors, optimize the design of the test matrix, improve the pertinence and efficiency of the test, and achieve the reliable establishment of the test matrix, making the test scenario more realistic and complex. After the establishment of the test matrix, according to the parameters corresponding to different test matrices, based on the preset test strategy of the automobile accelerator pedal misuse suppression system, the test strategy is executed on the test vehicle, and the corresponding test results are output, which can conduct targeted tests on the automobile accelerator pedal misuse suppression system, ensuring that the test covers various possible mis-stepping situations, thereby comprehensively evaluating the effectiveness of the system.

[0015] Finally, the analysis of the test results provides an objective evaluation of the effectiveness of the accelerator pedal misuse suppression system, which helps to identify the strengths and weaknesses of the system and provide direction for subsequent improvements.

[0016] In this solution, by selecting accident data of accelerator pedal misuse and establishing the corresponding test matrix, the corresponding suppression system is tested when the accelerator pedal is mistakenly pressed, thereby achieving in-depth testing of the performance of the accelerator pedal misuse suppression system and improving the effectiveness of the accelerator pedal misuse suppression system, thereby being able to provide more effective protection for the safety of passengers in the vehicle and the safety of VRUs (vulnerable road users) outside the vehicle, filling the gap in the automotive industry's functional testing of the accelerator pedal misuse suppression system, and can effectively guide automobile companies' development strategies for the accelerator pedal misuse suppression function, ultimately reducing the occurrence of accelerator pedal misuse accidents and reducing the number of casualties in accidents.

[0017] Further, the S1 includes:

[0018] S10, performing preliminary screening of accident data in the traffic accident in-depth investigation database based on preset preliminary screening conditions according to the traffic accident in-depth investigation database;

[0019] The initial screening condition is accident data where the cause of the accident is improper use of the accelerator / brake pedal;

[0020] S11. Based on a preset credibility judgment strategy, credibility judgment is performed on the selected accident data corresponding to the accident cause of improper use of the accelerator / brake pedal, accident data meeting the judgment conditions are selected, and the data are summarized to form a corresponding accelerator pedal misuse accident data set;

[0021] The credibility determination strategy is:

[0022] Identify the avoidance and braking conditions of the driver's return information in the accident data and generate corresponding identification information;

[0023] If the identification information is avoidance and braking, it is determined that the accident data does not meet the judgment conditions;

[0024] If the identification information is failure to avoid and acceleration, it is determined that the accident data meets the judgment conditions and the corresponding credibility is 0.8;

[0025] If the identification information is that there is no avoidance and no braking, the braking situation at the time of collision is judged on the accident video in the corresponding accident data. If the vehicle brake lights are on at the time of collision, it is judged that the accident data does not meet the judgment conditions. If the vehicle brake lights are not on at the time of collision, the vehicle EDR data in the accident data is used to identify the status of the vehicle 5 seconds before the collision. If the service brake of the vehicle is on 5 seconds before the collision, it is judged that the accident data does not meet the judgment conditions. If the service brake of the vehicle is off 5 seconds before the collision and the accelerator pedal increases, the accident data meets the judgment conditions and the corresponding credibility is 1.0.

[0026] Beneficial effects: In this solution, by setting specific initial screening conditions (such as accident data corresponding to accidents caused by improper use of the accelerator / brake pedal), irrelevant accident records can be effectively excluded, ensuring that the data set for subsequent analysis only contains accidents that are potentially related to the misuse of the accelerator pedal, thereby improving the pertinence and effectiveness of the data.

[0027] A comprehensive credibility judgment strategy is adopted, which not only considers the driver's immediate reaction (avoidance, braking), but also combines the accident video and vehicle electronic data (EVD Recording System EDR) information to ensure the accuracy and reliability of the accident data. That is, the credibility judgment strategy verifies the accident data in detail through multiple evidences such as driver return visit information, image information, and vehicle control information, avoiding false alarms and false positives, ensuring the purity and quality of the data set, and providing a reliable basis for subsequent analysis and testing, which is crucial for understanding the causes of accidental stepping accidents and developing effective preventive measures.

[0028] Further, the S3 includes:

[0029] S30, based on the extracted test parameters, determining the corresponding matrix dimensions, the matrix dimensions including the test vehicle element, the target object element and the environmental element, and determining the data corresponding to each matrix dimension based on each matrix dimension; the data corresponding to the strategy vehicle element includes the initial speed and motion state in the test parameters, the data corresponding to the target object element includes the collision target object category in the test parameters, and the data corresponding to the environmental element includes the weather, time period, road condition, road slope, and accident location in the test parameters;

[0030] S31. According to the determined data corresponding to each matrix dimension, a corresponding test matrix is ​​established, and the significant correlation between the data corresponding to each matrix dimension in the test matrix and the accident severity is judged to form a corresponding judgment result;

[0031] S32. Based on the judgment result, identify the key parameters corresponding to the severity of the accident affecting the accident of mis-stepping on the accelerator pedal corresponding to each matrix dimension;

[0032] S33, based on the identified key parameters, determine the data corresponding to the key parameters corresponding to the three matrix dimensions corresponding to each accident data in the accelerator pedal mis-stepping accident data set, and classify each accident data into a separate category;

[0033] S34, calculating the distance between each category, and merging the two categories with the best distance into a new category;

[0034] S35. Based on the formed new categories, a corresponding test matrix is ​​formed, and based on the preset weight determination strategy, the dimension weights corresponding to each matrix dimension and the parameter weights corresponding to the parameters under each matrix dimension are determined, and based on the dimension weights and parameter weights, the corresponding final test matrix is ​​formed.

[0035] Beneficial effects: By defining the test vehicle elements, target object elements and environmental elements, a comprehensive parameter framework was established to ensure the comprehensiveness and accuracy of the test coverage. By constructing a test matrix, the relationship between different elements and accident severity was quantified, which helps to identify which factors have the greatest impact on accidents and provides a data basis for subsequent analysis. By identifying key parameters through judgment results, focusing on the variables that have the greatest impact on accidents caused by accidentally pressing the accelerator pedal, this will help subsequent tests be more targeted and reduce resource waste.

[0036] At the same time, cluster mining of the pedal misapplication test scenarios corresponding to the accident data can better classify and process the data, facilitating subsequent targeted testing.

[0037] Furthermore, the preset weight determination strategy is:

[0038] Based on a pre-built neural network model, the neural network model includes an input layer, a hidden layer and an output layer; using the data corresponding to the key parameters of the three matrix dimensions corresponding to each accident data in the accelerator pedal mis-stepping accident data set as the input layer data, using the scene occurrence probability corresponding to each accident data and the number of deaths caused by the scene as the output layer data, identifying the weight coefficient between the input layer and the hidden layer, and the weight coefficient between the hidden layer and the output layer;

[0039] According to the weight coefficients between the input layer and the hidden layer, and between the hidden layer and the output layer, the weight coefficients between each neuron are determined, and the weight influence processing is performed on the weight coefficients between each neuron, and the parameter weights corresponding to each matrix dimension and the dimensional weights corresponding to each matrix dimension are output.

[0040] Beneficial effects: Using a neural network model to determine the weight determination strategy can not only automatically learn complex patterns in the data, but also provide more accurate prediction results and weight distribution, which is of great significance for understanding the key factors of pedal misapplication accidents and formulating effective prevention and response measures.

[0041] Furthermore, the S3 further includes:

[0042] S36, according to the final test matrix formed, based on the preset scenario severity calculation formula, the scenario severity value corresponding to each test matrix is ​​calculated, and each test matrix is ​​arranged in order from large to small according to the scenario severity value, and a corresponding evaluation weight is set, wherein the larger the order of the test matrix, the higher the corresponding evaluation weight;

[0043] The scenario severity calculation formula is:

[0044]

[0045] Where F is the severity value of the scene, C i is the severity of the i-th matrix dimension, σ i is the dimension weight of the i-th matrix dimension, c ij is the severity of the jth key parameter under the i-th matrix dimension, u ij is the parameter weight of the jth key parameter under the i-th matrix dimension, and n is the number of key parameters in a certain dimension.

[0046] Beneficial effects: In this solution, the severity of the scenario corresponding to the matrix dimension is first quantified, making the evaluation of the severity of the scenario more objective and quantitative.

[0047] Secondly, the test matrix is ​​arranged in descending order according to the scenario severity value, ensuring that high-severity scenarios are tested and focused on first. This is critical for the allocation of limited test resources and can prioritize solving potential high-risk problems. The higher the order of the test matrix, the higher the corresponding evaluation weight, which means that as the test progresses, the lower-severity scenarios ranked later are gradually given higher test priorities, ensuring that all scenarios are properly covered.

[0048] Furthermore, the test strategy of the preset automobile accelerator pedal misuse suppression system includes:

[0049] According to different test matrices, update the initial speed, road slope and target type of the corresponding test vehicle;

[0050] The target object to be tested is placed at a virtual collision position, and the center line of the target object coincides with the reference runway. After the test vehicle is driven to the initial test position, it is driven at a constant speed on the preset reference runway according to the updated initial data; the test vehicle is equipped with a high-precision positioning inertial navigation and a driving machine control device; the high-precision positioning inertial navigation collects vehicle GPS signals, and the driving machine control device controls the accelerator pedal, brake pedal, and steering wheel;

[0051] When the test vehicle reaches the test start position, the driving machine control device is automatically triggered to press the accelerator pedal with a constant force until the accelerator pedal opening reaches the full stroke, and the test vehicle drives until it collides with the target object or stops;

[0052] Determine whether the test vehicle collides with the target object. If a collision occurs, record the initial opening degree of the accelerator pedal of the test vehicle, the accelerator pedal depression time, the collision speed, and whether a warning sound is issued;

[0053] If no collision occurs, the initial opening degree of the accelerator pedal of the test vehicle, the time when the accelerator pedal is pressed, the time when braking begins, the distance from the test vehicle to the target object when braking, and whether a warning sound is sounded; at this time, the collision speed is recorded as 0 km / h;

[0054] Calculate the score for each test matrix, and after all test matrices are completed, integrate the corresponding test results.

[0055] Beneficial effects: In this solution, corresponding tests are carried out through different test matrices, covering a variety of driving conditions (such as different initial speeds, road slopes and target types), ensuring the comprehensiveness and representativeness of the test. By using virtual collision positions and reference runways, as well as high-precision positioning inertial navigation and driving machine control devices, the driving path of the test vehicle and the operation of the accelerator pedal can be accurately controlled, improving the accuracy and repeatability of the test results. The test process is automated, including triggering the accelerator pedal operation, recording test data, etc., which reduces the interference of human factors and ensures the objectivity and efficiency of the test. The test strategy takes into account both collision and non-collision results, and can comprehensively evaluate the performance of the system under emergency avoidance and normal driving conditions to ensure the reliability and stability of the system. By integrating the test results of all test matrices, the performance of the vehicle accelerator pedal misuse suppression system can be quantitatively evaluated, the advantages and disadvantages of the system can be identified, and data support can be provided for subsequent optimization and improvement.

[0056] Further, the S5 includes:

[0057] According to the test results, the corresponding total test evaluation score is calculated based on the preset evaluation score calculation formula;

[0058] The evaluation score calculation formula is:

[0059]

[0060] In the formula, ρ k is the evaluation weight corresponding to the kth test matrix, result k is the score corresponding to the test result of the kth test matrix.

[0061] Beneficial effects: In this scheme, the quantitative evaluation of the test results is achieved through the evaluation score calculation formula, which greatly improves the effective and reliable evaluation of the test results. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 The flowchart is a test method for the automobile accelerator pedal misuse suppression system in the first embodiment of the present invention.

[0063] Figure 2 This is a flow chart of establishing an accelerator pedal mis-pressing accident data set in the first embodiment of the present invention.

[0064] Figure 3 The flowchart is a test strategy of the automobile accelerator pedal misuse suppression system in the first embodiment of the present invention. DETAILED DESCRIPTION

[0065] The following is further described in detail through specific implementation methods:

[0066] Embodiment 1

[0067] A test method for a vehicle accelerator pedal misuse suppression system is basically as follows Figure 1 As shown, the following steps are included:

[0068] S1. Establish an accelerator pedal mis-pressing accident data set based on the traffic accident in-depth investigation database;

[0069] like Figure 2 As shown, the S1 includes:

[0070] S10, performing preliminary screening of accident data in the traffic accident in-depth investigation database based on a preset preliminary screening condition according to the traffic accident in-depth investigation database; the preliminary screening condition is accident data in which the cause of the accident is improper use of the accelerator / brake pedal;

[0071] The entry conditions corresponding to the data in the traffic accident in-depth investigation database are road traffic collisions involving at least one four-wheeled vehicle and at least one person injured;

[0072] S11. Based on a preset credibility judgment strategy, credibility judgment is performed on the selected accident data corresponding to the accident cause of improper use of the accelerator / brake pedal, accident data meeting the judgment conditions are selected, and the data are summarized to form a corresponding accelerator pedal misuse accident data set;

[0073] The credibility determination strategy is:

[0074] Identify the avoidance and braking conditions of the driver's return information in the accident data and generate corresponding identification information;

[0075] If the identification information is avoidance and braking, it is determined that the accident data does not meet the judgment conditions;

[0076] If the identification information is failure to avoid and acceleration, it is determined that the accident data meets the judgment conditions and the corresponding credibility is 0.8;

[0077] If the identification information is that there is no avoidance and no braking, the braking situation at the time of collision is judged on the accident video in the corresponding accident data. If the vehicle brake lights are on at the time of collision, it is judged that the accident data does not meet the judgment conditions. If the vehicle brake lights are not on at the time of collision, the vehicle EDR data in the accident data is used to identify the status of the vehicle 5 seconds before the collision. If the service brake of the vehicle is on 5 seconds before the collision, it is judged that the accident data does not meet the judgment conditions. If the service brake of the vehicle is off 5 seconds before the collision and the accelerator pedal increases, the accident data meets the judgment conditions and the corresponding credibility is 1.0.

[0078] S2. Automatically extracting test parameters corresponding to the accelerator pedal misuse according to the established accelerator pedal misuse accident data set; the test parameters include kinematic parameters, target object characteristics, initial speed, and vehicle motion state when the accelerator pedal misuse occurs;

[0079] In this embodiment, when extracting the features of the target object of the collision, the specific extraction steps are as follows: first, a corresponding training set is established, that is, image and labels folders are established in a local folder. The image folder stores images of collision targets such as pedestrians, two-wheeled vehicles, three-wheeled vehicles, and columns. The labels folder uses a .txt file format to mark the target category and the position of the object to be identified in the image. Use the YOLO V5 model to train and identify the images in the training set until the required number of training times is reached, and save the trained model, which is defined as trained YOLO V5. Capture the pre-collision image from the accident video, import it into the test_image folder in the local folder, use trained YOLO V5 to identify the target object in the image under the test_image folder, and output the identification result to the detect folder. Count the proportion of collision target objects: Count the number of recognitions under each label item, and output the proportion of each type.

[0080] When extracting the initial speed, the accident video is first preprocessed, including denoising, image enhancement, and edge detection. The machine learning-based target detection algorithm detects and identifies the vehicle in the video. The identified vehicle is tracked. The vehicle speed is estimated using the pixel position change of the vehicle in the image.

[0081] Vehicle motion state extraction: Vehicle identification and tracking: Detect and identify vehicles in the video based on machine learning target detection algorithm. Vehicle driving direction angle calculation: Based on opencv to identify the vehicle, calculate its position change in ten consecutive frames in the video to obtain its moving direction vector, and calculate the angle between the direction vector and the horizontal direction as α. Vehicle motion state judgment: Judge the vehicle motion state based on the calculated α. If α≈0° / 180°, it is considered that the vehicle is in a straight state; if the difference between α and 0° / 180° is large, it is considered that the vehicle is turning.

[0082] S3. Based on the extracted test parameters, perform significant correlation analysis on the test parameters, match the key parameters corresponding to the severity of the accident affecting the accident of mis-pressing the accelerator pedal, and design a corresponding test matrix;

[0083] The S3 includes:

[0084] S30, based on the extracted test parameters, determining the corresponding matrix dimensions, the matrix dimensions including the test vehicle element, the target object element and the environmental element, and determining the data corresponding to each matrix dimension based on each matrix dimension; the data corresponding to the strategy vehicle element includes the initial speed and motion state in the test parameters, the data corresponding to the target object element includes the collision target object category in the test parameters, and the data corresponding to the environmental element includes the weather, time period, road condition, road slope, and accident location in the test parameters;

[0085] S31. According to the data corresponding to each determined matrix dimension, a corresponding test matrix is ​​established, and the significant correlation between the data corresponding to each matrix dimension in the test matrix and the accident severity is judged to form a corresponding judgment result; in this embodiment, since the test vehicle element and the target object element are the test subject elements, it is not necessary to use this matrix dimension as a correlation analysis to measure its significant influence on the accident severity. Only the factors under the matrix dimension corresponding to the environmental element are subjected to correlation analysis.

[0086] Among them, the probability model of accident severity is:

[0087]

[0088] In the formula, p m is the probability that the accident severity is the mth category, x is the independent variable that has a significant impact on the accident severity, and β0 is the regression intercept.

[0089] Corresponding regression analysis variables are established, and the values ​​of each variable and their corresponding codes are shown in Table 1. When selecting, the significance level is selected as 0.05, and the variables with a significance level less than 0.05 after model fitting are extracted as key parameters affecting the severity of accidents.

[0090]

[0091]

[0092] Table 1

[0093] For example, through selection, the key parameters corresponding to the test matrix of pedal misapplication are summarized as shown in Table 2.

[0094]

[0095] Table 2

[0096] S32. Based on the judgment result, identify the key parameters corresponding to the severity of the accident affecting the accident of mis-stepping on the accelerator pedal corresponding to each matrix dimension;

[0097] S33, based on the identified key parameters, determine the data corresponding to the key parameters corresponding to the three matrix dimensions corresponding to each accident data in the accelerator pedal mis-stepping accident data set, and classify each accident data into a separate category;

[0098] S34, calculating the distance between each category, and merging the two categories with the best distance into a new category;

[0099] S35. Based on the formed new categories, a corresponding test matrix is ​​formed, and based on the preset weight determination strategy, the dimension weights corresponding to each matrix dimension and the parameter weights corresponding to the parameters under each matrix dimension are determined, and based on the dimension weights and parameter weights, the corresponding final test matrix is ​​formed.

[0100] The preset weight determination strategy is:

[0101] Based on a pre-built neural network model, the neural network model includes an input layer, a hidden layer and an output layer; using the data corresponding to the key parameters of the three matrix dimensions corresponding to each accident data in the accelerator pedal mis-stepping accident data set as the input layer data, using the scene occurrence probability corresponding to each accident data and the number of deaths caused by the scene as the output layer data, identifying the weight coefficient between the input layer and the hidden layer, and the weight coefficient between the hidden layer and the output layer;

[0102] According to the weight coefficients between the input layer and the hidden layer, and between the hidden layer and the output layer, the weight coefficients between each neuron are determined, and the weight influence processing is performed on the weight coefficients between each neuron, and the parameter weights corresponding to each matrix dimension and the dimensional weights corresponding to each matrix dimension are output.

[0103] The S3 further includes:

[0104] S36, according to the final test matrix formed, based on the preset scenario severity calculation formula, the scenario severity value corresponding to each test matrix is ​​calculated, and each test matrix is ​​arranged in order from large to small according to the scenario severity value, and a corresponding evaluation weight is set, wherein the larger the order of the test matrix, the higher the corresponding evaluation weight;

[0105] The scenario severity calculation formula is:

[0106]

[0107] Where F is the severity value of the scene, C i is the severity of the i-th matrix dimension, σ i is the dimension weight of the i-th matrix dimension, c ij is the severity of the jth key parameter under the i-th matrix dimension, u ij is the parameter weight of the jth key parameter under the i-th matrix dimension, and n is the number of key parameters in a certain dimension. In this embodiment, different test matrices correspond to different test scenarios, mainly in which the initial speed, road slope, and target type are different.

[0108] S4. According to the parameters corresponding to different test matrices, based on the preset test strategy of the automobile accelerator pedal misuse suppression system, the test strategy is executed on the test vehicle, and the corresponding test results are output;

[0109] like Figure 3 As shown, the test strategy of the preset automobile accelerator pedal misuse suppression system includes:

[0110] According to different test matrices, update the initial speed, road slope and target type of the corresponding test vehicle;

[0111] The target object to be tested is placed at a virtual collision position, and the center line of the target object coincides with the reference runway. After the test vehicle is driven to the initial test position, it is driven at a constant speed on the preset reference runway according to the updated initial data; the test vehicle is equipped with a high-precision positioning inertial navigation and a driving machine control device; the high-precision positioning inertial navigation collects vehicle GPS signals, and the driving machine control device controls the accelerator pedal, brake pedal, and steering wheel;

[0112] When the test vehicle reaches the test start position, the driving machine control device is automatically triggered to press the accelerator pedal with a constant force until the accelerator pedal opening reaches the full stroke, and the test vehicle drives until it collides with the target object or stops;

[0113] Determine whether the test vehicle collides with the target object. If a collision occurs, record the initial opening degree of the accelerator pedal of the test vehicle, the accelerator pedal depression time, the collision speed, and whether a warning sound is issued;

[0114] If no collision occurs, the initial opening degree of the accelerator pedal of the test vehicle, the time when the accelerator pedal is pressed, the time when braking begins, the distance from the test vehicle to the target object when braking, and whether a warning sound is sounded; at this time, the collision speed is recorded as 0 km / h;

[0115] Calculate the score of each test matrix, and integrate the corresponding test results after all test matrices are completed. In this embodiment, the test initial position refers to the position where the test vehicle is initially placed when the test starts. The reference path refers to the target route that the test vehicle is traveling. The initial opening of the accelerator pedal refers to the percentage of the opening of the accelerator pedal in the full stroke when the test vehicle is traveling at the initial speed. The virtual collision position refers to the position on the reference path where the test vehicle is assumed to collide with the target object. The test start position refers to the position on the reference path where the test vehicle begins to perform the accelerator pedal misuse operation. The accelerator pedal depression time refers to the time required from the initial opening of the accelerator pedal to the accelerator pedal opening reaching 100% of the full stroke. The braking time refers to the time when the vehicle pedal misuse suppression system begins to work during the accelerator pedal depression process. The collision speed refers to the test vehicle speed when it is less than 0.0m from the virtual collision position. When the virtual collision position is not reached, it is set to 0km / h.

[0116] S5. Analyze the test results according to the output test results, and output corresponding analysis results.

[0117] The S5 includes:

[0118] According to the test results, the corresponding total test evaluation score is calculated based on the preset evaluation score calculation formula;

[0119] The evaluation score calculation formula is:

[0120]

[0121] In the formula, ρ k is the evaluation weight corresponding to the kth test matrix, result k is the score corresponding to the test result of the kth test matrix.

[0122] In this embodiment, the braking efficiency value of a single scenario is calculated according to the test results corresponding to different test matrices. The specific calculation formula is as follows:

[0123]

[0124] Where, brake efficiency is the braking efficiency value of a single scenario, V initial is the initial velocity of the test vehicle, and V is the corresponding collision velocity.

[0125] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is described too much here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can know all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A test method for a vehicle accelerator pedal misuse suppression system, characterized in that: The following steps are involved: S1. Establish an accelerator pedal mis-pressing accident data set based on the traffic accident in-depth investigation database; S2. Automatically extracting test parameters corresponding to the accelerator pedal misuse according to the established accelerator pedal misuse accident data set; S3. Based on the extracted test parameters, perform significant correlation analysis on the test parameters, match the key parameters corresponding to the severity of the accident affecting the accident of mis-pressing the accelerator pedal, and design a corresponding test matrix; S4. According to the parameters corresponding to different test matrices, based on the preset test strategy of the automobile accelerator pedal misuse suppression system, the test strategy is executed on the test vehicle, and the corresponding test results are output; S5. Based on the output test results, conduct a comprehensive analysis on the test results and output the final analysis results.

2. A method for testing a vehicle accelerator pedal misuse suppression system according to claim 1, characterized in that: The S1 includes: S10, performing preliminary screening of accident data in the traffic accident in-depth investigation database based on preset preliminary screening conditions according to the traffic accident in-depth investigation database; The initial screening condition is accident data where the cause of the accident is improper use of the accelerator / brake pedal; S11. Based on a preset credibility judgment strategy, credibility judgment is performed on the selected accident data corresponding to the accident cause of improper use of the accelerator / brake pedal, accident data meeting the judgment conditions are selected, and the data are summarized to form a corresponding accelerator pedal misuse accident data set; The credibility determination strategy is: Identify the avoidance and braking conditions of the driver's return information in the accident data and generate corresponding identification information; If the identification information is avoidance and braking, it is determined that the accident data does not meet the judgment conditions; If the identification information is failure to avoid and acceleration, it is determined that the accident data meets the judgment conditions and the corresponding credibility is 0.8; If the identification information is that there is no avoidance and no braking, the braking situation at the time of collision is judged on the accident video in the corresponding accident data. If the vehicle brake lights are on at the time of collision, it is judged that the accident data does not meet the judgment conditions. If the vehicle brake lights are not on at the time of collision, the vehicle EDR data in the accident data is used to identify the status of the vehicle 5 seconds before the collision. If the service brake of the vehicle is on 5 seconds before the collision, it is judged that the accident data does not meet the judgment conditions. If the service brake of the vehicle is off 5 seconds before the collision and the accelerator pedal increases, the accident data meets the judgment conditions and the corresponding credibility is 1.

0.

3. A method for testing a vehicle accelerator pedal misuse suppression system according to claim 2, characterized in that: The S3 includes: S30, based on the extracted test parameters, determining the corresponding matrix dimensions, the matrix dimensions including the test vehicle element, the target object element and the environmental element, and determining the data corresponding to each matrix dimension based on each matrix dimension; the data corresponding to the strategy vehicle element includes the initial speed and motion state in the test parameters, the data corresponding to the target object element includes the collision target object category in the test parameters, and the data corresponding to the environmental element includes the weather, time period, road condition, road slope, and accident location in the test parameters; S31. According to the determined data corresponding to each matrix dimension, a corresponding test matrix is ​​established, and the significant correlation between the data corresponding to each matrix dimension in the test matrix and the accident severity is judged to form a corresponding judgment result; S32. Based on the judgment result, identify the key parameters corresponding to the severity of the accident affecting the accident of mis-stepping on the accelerator pedal corresponding to each matrix dimension; S33, based on the identified key parameters, determine the data corresponding to the key parameters corresponding to the three matrix dimensions corresponding to each accident data in the accelerator pedal mis-stepping accident data set, and classify each accident data into a separate category; S34, calculating the distance between each category, and merging the two categories with the best distance into a new category; S35. Based on the formed new categories, a corresponding test matrix is ​​formed, and based on the preset weight determination strategy, the dimension weights corresponding to each matrix dimension and the parameter weights corresponding to the parameters under each matrix dimension are determined, and based on the dimension weights and parameter weights, the corresponding final test matrix is ​​formed.

4. A method for testing a vehicle accelerator pedal misuse suppression system according to claim 3, characterized in that: The preset weight determination strategy is: Based on a pre-built neural network model, the neural network model includes an input layer, a hidden layer and an output layer; using the data corresponding to the key parameters of the three matrix dimensions corresponding to each accident data in the accelerator pedal mis-stepping accident data set as the input layer data, using the scene occurrence probability corresponding to each accident data and the number of deaths caused by the scene as the output layer data, identifying the weight coefficient between the input layer and the hidden layer, and the weight coefficient between the hidden layer and the output layer; According to the weight coefficients between the input layer and the hidden layer, and between the hidden layer and the output layer, the weight coefficients between each neuron are determined, and the weight influence processing is performed on the weight coefficients between each neuron, and the parameter weights corresponding to each matrix dimension and the dimensional weights corresponding to each matrix dimension are output.

5. A method for testing a vehicle accelerator pedal misuse suppression system according to claim 4, characterized in that: The S3 further includes: S36, according to the final test matrix formed, based on the preset scenario severity calculation formula, the scenario severity value corresponding to each test matrix is ​​calculated, and each test matrix is ​​arranged in order from large to small according to the scenario severity value, and a corresponding evaluation weight is set, wherein the larger the order of the test matrix, the higher the corresponding evaluation weight; The scenario severity calculation formula is: Where F is the severity value of the scene, C i is the severity of the i-th matrix dimension, σ i is the dimension weight of the i-th matrix dimension, c ij is the severity of the jth key parameter under the i-th matrix dimension, u ij is the parameter weight of the jth key parameter under the i-th matrix dimension, and n is the number of key parameters in a certain dimension.

6. A method for testing a vehicle accelerator pedal misuse suppression system according to claim 5, characterized in that: The test strategy of the preset automobile accelerator pedal misuse suppression system includes: According to different test matrices, update the initial speed, road slope and target type of the corresponding test vehicle; The target object to be tested is placed at a virtual collision position, and the center line of the target object coincides with the reference runway. After the test vehicle is driven to the initial test position, it is driven at a constant speed on the preset reference runway according to the updated initial data; the test vehicle is equipped with a high-precision positioning inertial navigation and a driving machine control device; the high-precision positioning inertial navigation collects vehicle GPS signals, and the driving machine control device controls the accelerator pedal, brake pedal, and steering wheel; When the test vehicle reaches the test start position, the driving machine control device is automatically triggered to press the accelerator pedal with a constant force until the accelerator pedal opening reaches the full stroke, and the test vehicle drives until it collides with the target object or stops; Determine whether the test vehicle collides with the target object. If a collision occurs, record the initial opening degree of the accelerator pedal of the test vehicle, the accelerator pedal depression time, the collision speed, and whether a warning sound is issued; If no collision occurs, the initial opening degree of the accelerator pedal of the test vehicle, the time when the accelerator pedal is pressed, the time when braking begins, the distance from the test vehicle to the target object when braking, and whether a warning sound is sounded; at this time, the collision speed is recorded as 0 km / h; Calculate the score for each test matrix, and after all test matrices are completed, integrate the corresponding test results.

7. A method for testing a vehicle accelerator pedal misuse suppression system according to claim 6, characterized in that: The S5 includes: According to the test results, the corresponding total test evaluation score is calculated based on the preset evaluation score calculation formula; The evaluation score calculation formula is: In the formula, ρ k is the evaluation weight corresponding to the kth test matrix, result k is the score corresponding to the test result of the kth test matrix.

Images