Child seat type selection evaluation method and device, electronic equipment and storage medium

Abstract

The invention relates to a child seat type selection evaluation method and device, electronic equipment and a storage medium. The method comprises the steps that a child seat type selection data set is obtained, and the child seat type selection data set at least comprises collision pulse data of multiple vehicles, seat measurement data of multiple seats and multiple damage result data; performing data correction on the child seat type selection data set; performing key feature parameterization on the standardized data set through a Pearson's correlation coefficient; training a child seat model selection model pre-constructed by adopting a random forest regression algorithm by utilizing the key feature data; inputting the collision pulse data of the target vehicle into the trained child seat model selection model to obtain a child seat adapted to the target vehicle; and in response to updating of the child seat model selection data set, updating the child seat model selection model by using the updated child seat model selection data set.

Description

Technical Field

[0001] The present disclosure relates to the technical field of data processing, and particularly to a method, device, electronic device, and storage medium for evaluating the selection of child seats. Background Art

[0002] Child safety in vehicle is a global focus. Every year, there are numerous child injury incidents caused by traffic accidents, and many of these casualties can be prevented by appropriate child restraint systems (such as child safety seats). However, traditional child seat design and selection mainly rely on simulation research and development, crash tests, and expert experience. This method is time-consuming, laborious, costly, and it is difficult to comprehensively evaluate the safety in various crash scenarios. Summary of the Invention

[0003] To solve the above technical problems, the present disclosure provides a method, device, electronic device, and storage medium for evaluating the selection of child seats.

[0004] The present disclosure provides a method for evaluating the selection of child seats, including: obtaining a child seat selection data set, where the child seat selection data set at least includes collision pulse data of multiple vehicles, seat measurement data of multiple seats, and multiple injury result data, and among them, the injury result data is calculated based on the collision pulse data, the standard driving condition simulation model of the vehicle, and the seat simulation model of the child seat; performing data correction on the child seat selection data set to obtain a standardized data set; performing key feature parameterization on the standardized data set, and screening out waveform features significantly related to the target variable in the standardized data set through the Pearson correlation coefficient to obtain key feature data; using the key feature data to train a pre-constructed child seat selection model using the random forest regression algorithm, including: performing ensemble learning on the input key feature data through multiple decision trees; inputting the collision pulse data of the target vehicle into the trained child seat selection model to obtain a child seat suitable for the target vehicle; in response to the update of the child seat selection data set, updating the child seat selection model using the updated child seat selection data set.

[0005] Optionally, the standard driving condition simulation model is calculated based on the rear seat data, front seat data, cockpit data, and safety configuration data of multiple vehicles.

[0006] Optionally, the seat measurement data includes the size data and installation data of the child seat, and the injury result data includes the head 3ms synthetic acceleration data, head injury data, neck force data, and chest compression displacement data of the child seat.

[0007] Optionally, data correction is performed on the child seat selection data set to obtain a standardized data set, including: using the interpolation method to repair the incomplete data in the collision pulse data.

[0008] Optionally, outliers, extreme points, and missing values in the seat measurement data and injury result data are processed, and noise reduction processing and data correction processing are performed on the seat measurement data and injury result data to unify the data units of the seat measurement data and the data units of the injury result data.

[0009] Optionally, the waveform features include waveform activation duration, signal intensity integration, root mean square power, dominant harmonic amplitude, spectral extreme points, band energy integration, power spectral centroid, and spectral randomness measure.

[0010] Optionally, the collision pulse data of the target vehicle is input into the trained child seat selection model to obtain a child seat suitable for the target vehicle, including: inputting the collision pulse data of the target vehicle into the child seat selection model to respectively obtain the injury result data of multiple candidate child seats on the target vehicle; calculating the selection scores of multiple candidate child seats on the target vehicle according to the injury result data and the target regulations; sorting the multiple selection scores, and taking the child seat with the highest selection score as the child seat suitable for the target vehicle.

[0011] Based on the same inventive concept, the present disclosure also provides a child seat selection evaluation device, including: a data and model acquisition module, configured to acquire a child seat selection data set, where the child seat selection data set at least includes collision pulse data of multiple vehicles, seat measurement data of multiple seats, and multiple injury result data, where the injury result data is calculated based on the collision pulse data, a standard working condition simulation model of the vehicle, and a seat simulation model of the child seat; a data correction module, configured to perform data correction on the child seat selection data set to obtain a standardized data set; a feature extraction module, configured to perform key feature parameterization on the standardized data set, and screen out waveform features significantly correlated with the target variable in the standardized data set through the Pearson correlation coefficient to obtain key feature data; a model training module, configured to use the key feature data to train a child seat selection model pre-constructed by using a random forest regression algorithm, including: performing ensemble learning on the input key feature data through multiple decision trees; an evaluation module, inputting the collision pulse data of the target vehicle into the trained child seat selection model to obtain a child seat suitable for the target vehicle; an update module, configured to, in response to the update of the child seat selection data set, update the child seat selection model by using the updated child seat selection data set.

[0012] Based on the same inventive concept, the present disclosure further provides an electronic device, including: a processor; a memory for storing executable instructions; wherein, the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the method of any one of the above.

[0013] Based on the same inventive concept, the present disclosure further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the processor is caused to implement the method of any one of the above.

[0014] The technical solution provided by the present disclosure has the following advantages compared with the prior art: By training the model, the verification and evaluation of the child seat in the present disclosure do not require simulating and testing each product one by one. Only by inputting the pulse data of vehicle collision, the selection of child seats in the early stage of new vehicle development can be quickly carried out, greatly saving manpower, time and cost, and accelerating the vehicle model R & D process. Moreover, the method provided by the present disclosure can dynamically adapt to the update of regulations. After the data related to child seats provided in the regulations are updated, only the child seat selection data set needs to be updated, and then an evaluation scheme for child seat selection can be quickly provided for all the R & D vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present disclosure and used together with the specification to explain the principles of the present disclosure.

[0016] To more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 It is a schematic flowchart of a method for evaluating the selection of child seats provided by an embodiment of the present disclosure; Figure 2 It is a schematic flowchart of another method for evaluating the selection of child seats provided by an embodiment of the present disclosure; Figure 3 It is a schematic structural diagram of a device for evaluating the selection of child seats provided by an embodiment of the present disclosure; Figure 4 It is a schematic hardware structure diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to more clearly understand the above-mentioned objects, features, and advantages of the embodiments of the present disclosure, the solutions of the embodiments of the present disclosure will be further described below. It should be noted that, without conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other.

[0019] In the following description, many specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. However, the embodiments of the present disclosure may also be implemented in other ways different from those described herein. Obviously, the embodiments in the specification are only a part of the embodiments of the present disclosure, rather than all of the embodiments.

[0020] In the related art, the design and selection of child seats usually require a large amount of simulation research and development and crash tests. These tests not only require high-cost equipment and site investment, but also require professional technicians to operate and analyze data, and the development cycle is relatively long, making it difficult to meet the requirements of the rapid iteration of modern automotive products. Therefore, the traditional method has certain limitations in improving the safety of children in the car, and there is an urgent need for a more scientific, efficient, and economical method to optimize the selection of child seats.

[0021] In view of this, an embodiment of the present disclosure provides a method for evaluating the selection of child seats, as Figure 1 shown, including: S1. Obtain a child seat selection data set, which at least includes collision pulse data of multiple vehicles, seat measurement data of multiple seats, and multiple injury result data. Among them, the injury result data is calculated based on the collision pulse data, the standard working condition simulation model of the vehicle, and the seat simulation model of the child seat.

[0022] Specifically, before S1, the method provided by the embodiments of the present disclosure further includes constructing a vehicle-seat coupling simulation model based on the standard working condition simulation model of the vehicle and the seat simulation model of the child seat. Among them, the seat simulation model is mainly multiple child seat models recommended by C-NCAP, covering child seats of different brands, installation methods, and material characteristics. The standard working condition simulation model is the simulation model of a conventional vehicle, which will be further described in the following embodiments. Using the collision pulse data of the layout position obtained by the structural simulation standard collision as the input condition of the vehicle-seat coupling simulation model, and using finite element analysis software to perform dynamic collision simulation to calculate the dynamic response of the child seat during the collision, that is, the above-mentioned injury score, so as to obtain a complete child seat selection data set.

[0023] S2. Correct the data in the child seat selection data set to obtain a standardized data set.

[0024] The method provided by the embodiments of the present disclosure ensures the consistency and availability of data through the integration and correction processing of the data set, providing a solid data foundation for the training and prediction of the model.

[0025] S3. Parametrize the key features of the standardized data set, and screen out the waveform features significantly related to the target variable in the standardized data set through the Pearson correlation coefficient to obtain the key feature data.

[0026] The method provided by the embodiments of the present disclosure extracts key feature parameters from the standardized data set through feature extraction, thereby realizing the precise input processing of the data.

[0027] S4. Use the key feature data to train the child seat selection model pre-constructed by the random forest regression algorithm. Specifically, it includes: performing ensemble learning on the input key feature data through multiple decision trees.

[0028] The method provided by the embodiments of the present disclosure obtains a high-precision child seat selection model based on the extracted key feature parameters. This model is trained with a large amount of data, significantly improving its prediction and evaluation ability, making the child seat selection process more efficient and accurate, and providing guarantee for the selection and safety evaluation of child seats.

[0029] S5. Input the collision pulse data of the target vehicle into the trained child seat selection model to obtain the child seat adapted to the target vehicle.

[0030] Since the seat measurement data and the seat simulation model in the above S1 include the information of the child seat to be selected, for the trained child seat selection model, only the collision pulse data of the vehicle needs to be input to obtain the child seat adapted to the vehicle.

[0031] S6. In response to the update of the child seat selection data set, use the updated child seat selection data set to update the child seat selection model.

[0032] The method provided by the embodiments of the present disclosure enables the verification and evaluation of child seats without performing simulation and tests on each product one by one. Only by inputting the pulse data of vehicle collisions, the child seat selection in the early development process of new vehicles can be quickly carried out, greatly saving manpower, time and costs, and accelerating the vehicle model R & D process. Moreover, the method provided by the embodiments of the present disclosure can dynamically adapt to the update of regulations. After the data related to child seats provided in the regulations is updated, only the child seat selection data set needs to be updated to quickly provide a child seat selection evaluation plan for all R & D vehicles.

[0033] In some embodiments, the above-mentioned standard operating condition simulation model is calculated and constructed based on rear seat data, front seat data, cockpit data and safety configuration data of multiple vehicles.

[0034] Specifically, the above standard operating condition simulation model is calculated and constructed based on the statistical mean of rear seat data, the statistical mean of front seat data, the statistical mean of cabin data and the statistical mean of safety configuration data of multiple vehicles.

[0035] The front seat data and rear seat data include data related to backrest angle, seat material, anti-submarine bracket, etc.; the cockpit data include data related to the body of the vehicle, B-pillar, etc.; the safety configuration data include data related to seat belt pre-tensioning and buckle arrangement.

[0036] Seat material-related data (such as elastic modulus, yield strength) and safety configuration data (such as seat belt ignition time, force limit, and layout space) are determined based on the statistical mean of simulation data of historical R&D vehicles to ensure that the standard operating condition simulation model meets industry general standards.

[0037] In some embodiments, the seat measurement data includes the size data of the child seat and the installation data of the child seat, and the injury result data includes the 3ms synthetic acceleration data of the child seat's head, the child seat's head injury data, the child seat's neck force data, and the child seat's chest pressure displacement data.

[0038] Specifically, the size data of the child seat includes data such as the length, width, height, thickness, etc. of each component of the seat; the installation data of the child seat includes the corresponding installation method and installation direction of the seat.

[0039] In specific implementation, the method for obtaining the size data and installation data of the above-mentioned child seat includes: using a three-dimensional laser scanner to perform an all-round scan of the actual child seat to obtain accurate geometric size data, measuring the sizes of key components, including the length, width, height and thickness of the seat base, the height and inclination of the backrest, the adjustment range of the headrest, the size of the side protection device, establishing a size parameter database, and recording the size characteristics of each model of seat; the installation method and installation direction are obtained through the seat usage method.

[0040] Specifically, the 3ms synthetic acceleration of the head refers to the synthetic value of the acceleration experienced by the head within 3 milliseconds, which is one of the key factors in measuring the risk of head injury; the neck force data reflects the tensile, compressive or shear force experienced by the neck under impact, and the chest pressure injury data records the synthetic acceleration of the chest when it is stressed, which is an important basis for assessing the risk of chest injury. Among them, the calculation formulas for the head injury HIC value are shown in formulas (1) and (2):

[0041]

[0042] Among them, , , are the acceleration values of the head in three directions, with the unit of g, .

[0043] The calculation formula of the chest injury value is shown in Equation (1).

[0044] In some embodiments, the above S2 includes: Using the interpolation method to repair the non-complete data in the collision pulse data.

[0045] Specifically, the collision pulse data is specifically a pulse collision curve. When correcting the collision pulse data, it is necessary to check the continuity and integrity of the data, and remove the non-complete data; for some non-smooth curves with fewer nodes, the interpolation method needs to be used for repair. When correcting the collision pulse data, it is also necessary to correct the collision pulse curve data under different unit systems and convert them into collision pulse curve data under a unified unit system.

[0046] In some embodiments, the above S2 includes: Processing the outliers, extreme points, and missing values in the seat measurement data and the injury result data, performing denoising processing and data correction processing on the seat measurement data and the injury result data, unifying the data units of the seat measurement data, and unifying the data units of the injury result data.

[0047] Specifically, when correcting the numerical data such as the seat measurement data and the injury result data, it is necessary to detect and process the outliers and extreme points in the data, fill in the missing values, perform numerical smoothness processing, and at the same time perform data denoising and data correction. When correcting the numerical data, it is also necessary to correct the data under different unit systems and convert them into data under a unified unit system.

[0048] In some embodiments, in the above S3, the waveform features include waveform activation duration, signal intensity integral, root mean square power, dominant harmonic amplitude, spectral extreme points, band energy integral, power spectrum centroid, and spectral randomness measure. These waveform features are multiple key feature pairs with the highest Pearson correlation with the target value.

[0049] Specifically, the above S3 includes: using multi-dimensional feature Pearson correlation coefficient feature extraction. The Pearson correlation coefficient feature extraction significantly improves the model's understanding ability of complex collision dynamics, and improves the accuracy and robustness of model prediction, ensuring that the model can capture complex collision dynamics and child seat performance.

[0050] In a specific embodiment, the above S3 includes: establishing a correlation model between the pulse signal matrix and the injury result data (the above-mentioned head 3ms synthetic acceleration data, head injury data HIC, neck force data, and chest pressure injury data), and independently calculating the Pearson correlation coefficient for each injury index; adopting a multi-dimensional feature screening method, taking the average value of the Pearson correlation coefficients of each index to obtain the total Pearson correlation coefficient value of each feature for the multi-dimensional target quantity, and then selecting the key feature data, including waveform activation duration, signal intensity integral, root mean square power, dominant harmonic amplitude, spectral extreme point, band energy integral, power spectrum centroid, and spectral randomness measure.

[0051] Among them, the calculation method of the Pearson correlation coefficient is shown in Equation (3):

[0052] In some embodiments, before the above S4, the method provided by the embodiments of the present disclosure further includes: Constructing a child seat selection score model based on the random forest regression algorithm.

[0053] The method provided by the above embodiments of the present disclosure constructs a child seat selection score model using the random forest regression algorithm. This algorithm has high robustness when dealing with complex data sets, can effectively cope with the challenges of high-dimensional features, and thus ensures the stability and accuracy of the model in the face of a large amount of diverse input data. In addition, the random forest regression algorithm makes a comprehensive judgment by integrating multiple decision trees, reduces the variance of the model, significantly reduces the risk of overfitting, and improves the generalization ability of the model. By reasonably setting parameters such as the number of decision trees and the number of splitting features in the model, it can ensure that the model is neither overly complex nor fails to fully learn the data features.

[0054] Random forest is a powerful machine learning algorithm widely used in classification and regression problems. Its core lies in improving the prediction performance and model robustness by integrating multiple decision tree models. As the basic unit of the random forest, the decision tree recursively divides the data set into smaller subsets and finally forms a prediction result. Each internal node represents a feature, the branch represents the possible value of the feature, and the leaf node represents the final prediction result. On the basis of the traditional decision tree, the random forest introduces the bagging method and feature randomization technology to further optimize the model performance.

[0055] Specifically, the random forest regression algorithm uses the bagging method during training. By randomly sampling the original dataset with replacement, multiple different training subsets are generated, and each subset is used to train a decision tree. At the same time, during the splitting process of each node, the random forest does not consider all features but randomly selects a subset of features for splitting. This method of feature randomization can effectively reduce the correlation between models, enhance the diversity of the models, thereby further reducing the variance of the models and enhancing their generalization ability.

[0056] When constructing the model, it is necessary to reasonably set the key hyperparameters of the random forest regression algorithm. These hyperparameters include the number of decision trees, the number of features considered when splitting a single tree, the maximum depth of a single tree, the minimum number of samples required for node splitting, and the minimum number of samples required for leaf nodes, etc. By reasonably setting these hyperparameters, it can be ensured that the model can fully learn the specific information in the feature dataset, and at the same time avoid problems such as one-sided information mining and inaccurate weight setting caused by a single decision tree, thereby reducing calculation errors and improving the prediction accuracy of the model.

[0057] During the model training process, it is necessary to select a suitable model type according to the specific task objective. For the embodiments of the present disclosure, a method combining a classification model and a regression model can be adopted. Specifically, when the target variable is N different child seats, the model outputs the category to which each sample belongs; while when the target variable is N different injury results, the model will output the prediction score of each sample. During the training process, it is necessary to adjust the hyperparameters of the algorithm so that the prediction results of the classification model and the regression model can be effectively fused at the decision-making level.

[0058] In some embodiments, the above S5 includes: S501. Input the collision pulse data of the target vehicle into the child seat selection model to obtain the injury result data of multiple child seats to be selected on the target vehicle respectively.

[0059] S502. Calculate the selection scores of multiple child seats to be selected on the target vehicle according to the injury result data and the target regulations. Specifically, the selection score includes the total score calculated according to the target regulations for the injury results of each part (head, neck, chest).

[0060] S503. Sort the multiple selection scores, and take the child seat with the highest selection score as the child seat suitable for the target vehicle.

[0061] In the field of automotive safety, the management rules of NCAP (New Car Assessment Program) in various countries (hereinafter referred to as regulations) have requirements for the dynamic evaluation of child protection, including the frontal 100% overlap rigid barrier collision condition (FRB), the frontal 50% overlap moving progressive deformation barrier (MPDB) condition, and the side pole collision condition (pole collision).

[0062] In a specific embodiment, as Figure 2 shown, it is necessary to first select the target regulation, such as Chinese GB, Chinese C-NCAP, Chinese C-IASI, etc., and then select the automotive passive safety collision conditions. In this embodiment, it can be divided into the frontal collision condition and the pole collision condition, and determine the total target score for the frontal collision and the total target score for the pole collision. The input data is the collision pulse data, and by inputting it into the child seat selection score model, the selection scores of multiple child seats to be selected on the target vehicle can be obtained.

[0063] The child seat selection model provided in the above embodiments of the present disclosure provides important decision-making support for vehicle design and child seat selection. By inputting vehicle pulse data, it quickly generates selection suggestions for child seats, and provides corresponding selection scores and ranking results. This not only provides a scientific basis for vehicle developers, but also improves the design efficiency and accuracy of the child protection system. At the same time, the full-process automation from data collection to solution output makes the optimization process of child seat selection more efficient and transparent, providing strong technical support for the intelligentization of vehicle design.

[0064] In some embodiments, the above S6 includes: In response to the update of the above target regulation, calculate the selection score according to the updated target regulation.

[0065] In a specific embodiment, after the child seat data in a certain target regulation (such as the C-NCAP regulation) is updated, new child seats need to be referred to for selection in child seat safety development. Therefore, the current child seat selection model needs to re-incorporate the characteristic correlation information of the new seats specified in this regulation, and then repeat the above actions of S1 to S5, so as to obtain child seats that meet the requirements of the new version of the regulation and dynamically adapt to the update of the regulation.

[0066] In a specific embodiment, the method provided in the embodiments of the present disclosure further includes testing, dynamically optimizing and updating the above child seat selection model to ensure the adaptability and accuracy of the model.

[0067] To ensure the objectivity and reliability of test results, it is necessary to divide the standardized key feature data into a training set, a validation set, and a test set. For the current model, the validation set is divided for model selection and hyperparameter tuning, and the test set is divided for the performance evaluation of the final model. The model is tested according to the division ratio while ensuring the independence and representativeness of each part.

[0068] Specifically, the performance of the model can be comprehensively evaluated through indicators such as classification accuracy, F1-score, mean squared error (MSE), and coefficient of determination (R²) to identify deficiencies. For the regression task of the prediction model, the focus is on the prediction accuracy of the model for the injury selection score. The injury selection score is an important indicator to measure the safety performance of child seats, and its prediction accuracy directly affects the practical application value of the model. In the test, mean squared error (MSE) and coefficient of determination (R²) are used as the main evaluation indicators. Mean squared error can reflect the average error between the predicted value and the true value of the model, while the coefficient of determination can quantify the explanatory ability of the model for data changes. By analyzing these indicators, the performance of the model in the regression task can be comprehensively evaluated. For the classification task of the prediction model, the goal of the classification task is to accurately predict the type of child seat, which is of great significance for the selection decision in vehicle design. Classification accuracy and F1-score will be used as the main evaluation indicators. Classification accuracy can reflect the performance of the model in the overall classification task, while the F1-score can comprehensively consider the precision and recall of the model to comprehensively evaluate the prediction ability of the model on different categories. By analyzing these indicators, the advantages and disadvantages of the model in the classification task can be found, providing a direction for subsequent optimization.

[0069] Specifically, the prediction accuracy and efficiency can be improved by systematically adjusting the model hyperparameters, including the number of decision trees, the number of features considered when splitting a single tree, the maximum depth of a single tree, the minimum number of samples required for node splitting, and the minimum number of samples required for leaf nodes, etc., to accelerate the convergence speed and improve the model performance.

[0070] In model validation, it is necessary to evaluate the generalization ability of the model, which is an important indicator to measure the performance of the model on unseen data. To verify the generalization ability of the model, the method of cross-validation will be used. By repeatedly dividing the standardized key feature data into a training set and a validation set, cross-validation can effectively reduce the bias caused by data division and ensure that the performance of the model on different data subsets is highly stable. The test process and cross-validation results are recorded, and the mean value and standard deviation are calculated to evaluate the generalization ability of the model.

[0071] The continuous optimization mechanism of the model not only improves the long-term performance of the model, but also enhances the practicality and adaptability of the model, ensuring its effectiveness in different design scenarios.

[0072] For the child seat selection model with the random forest regression algorithm as the core, the interpretability of the model is mainly evaluated. Although the random forest regression algorithm has high prediction performance, its model structure is relatively complex and difficult to directly interpret. For model evaluation, the method of feature importance analysis will be used to evaluate the influence degree of each feature on the model prediction result. By analyzing the feature importance, the testing department can provide valuable reference information for designers to help them understand the prediction logic of the model.

[0073] It should be noted that the method of the embodiment of the present application can be executed by a single device, such as a computer or a server, etc. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In this case of a distributed scenario, one of the multiple devices can only execute one or more steps of the method of the embodiment of the present application, and these multiple devices will interact with each other to complete the above method.

[0074] It should be noted that some embodiments of the present application have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order from that in the above embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0075] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present application also provides a child seat selection evaluation device, as Figure 3 shown, including: A data and model acquisition module 10, configured to acquire a child seat selection data set, where the child seat selection data set at least includes collision pulse data of multiple vehicles, seat measurement data of multiple seats, and multiple injury result data, and the injury result is calculated based on the collision pulse data, the standard working condition simulation model of the vehicle, and the seat simulation model of the child seat.

[0076] A data correction module 20, configured to correct the child seat selection data set to obtain a standardized data set.

[0077] A feature extraction module 30, configured to perform key feature parameterization on the standardized data set, and screen out waveform features that are significantly correlated with the target variable in the standardized data set through the Pearson correlation coefficient to obtain key feature data.

[0078] The model training module 40 is used to train a child seat selection model pre-constructed by using the random forest regression algorithm with the key feature data, including: performing ensemble learning on the input key feature data through multiple decision trees.

[0079] The evaluation module 50 inputs the collision pulse data of the target vehicle into the trained child seat selection model to obtain a child seat suitable for the target vehicle.

[0080] The update module 60 is used to update the child seat selection model by using the updated child seat selection data set in response to the update of the child seat selection data set.

[0081] The device provided by the embodiment of the present disclosure enables the verification and evaluation of child seats without performing simulations and tests on each product one by one through training the model. Only by inputting the pulse data of vehicle collisions, the selection of child seats in the early development process of new vehicles can be quickly carried out, greatly saving manpower, time, and costs, and accelerating the vehicle model R & D process. Moreover, the device provided by the embodiment of the present disclosure can dynamically adapt to the update of regulations. After the data related to child seats provided in the regulations is updated, only the child seat selection data set needs to be updated, and then the child seat selection evaluation scheme for all R & D vehicles can be quickly provided.

[0082] In some embodiments, the standard condition simulation model is calculated based on the rear seat data, front seat data, cockpit data, and safety configuration data of multiple vehicles.

[0083] In some embodiments, the seat measurement data includes the size data of the child seat and the installation data of the child seat, and the injury result data includes the 3ms synthetic acceleration data of the head of the child seat, the head injury data of the child seat, the neck force data of the child seat, and the chest compression displacement data of the child seat.

[0084] In some embodiments, the above data correction module is specifically used for: Using the interpolation method to repair the incomplete data in the collision pulse data.

[0085] In some embodiments, the above data correction module is specifically used for: Processing the outliers, outliers, and missing values in the seat measurement data and the injury result data, performing denoising processing and data correction processing on the seat measurement data and the injury result data, unifying the data units of the seat measurement data, and unifying the data units of the injury result data.

[0086] In some embodiments, the waveform features include waveform activation duration, signal intensity integration, root mean square power, dominant harmonic amplitude, spectral extreme points, frequency band energy integration, power spectrum centroid, and spectral randomness measure.

[0087] In some embodiments, the above-mentioned evaluation module is specifically configured to: Input the collision pulse data of the target vehicle into the child seat selection model to obtain the injury result data of multiple child seats to be selected on the target vehicle respectively; calculate the selection scores of multiple child seats to be selected on the target vehicle according to the injury result data and the target regulations; sort the multiple selection scores, and use the child seat with the highest selection score as the child seat suitable for the target vehicle.

[0088] For the convenience of description, when describing the above device, various modules are described separately according to their functions. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0089] The device in the above embodiment is used to implement the corresponding child seat selection and evaluation method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0090] Figure 4 The structural schematic diagram of an electronic device provided by an embodiment of the present disclosure is shown.

[0091] As Figure 4 shown, the electronic device may include a processor 1101 and a memory 1102 storing computer program instructions.

[0092] Specifically, the above-mentioned processor 1101 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0093] Memory 1102 may include a mass storage for information or instructions. By way of example and not limitation, memory 1102 may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 1102 may include removable or non-removable (or fixed) media. Where appropriate, memory 1102 may be internal or external to the integrated gateway device. In a particular embodiment, memory 1102 is a non-volatile solid state memory. In a particular embodiment, memory 1102 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0094] Processor 1101 reads and executes the computer program instructions stored in memory 1102 to perform the steps of the child seat selection evaluation method provided by the embodiments of the present disclosure.

[0095] In one example, the electronic device may further include a transceiver 1103 and a bus 1104. Among them, as Figure 4 shown, processor 1101, memory 1102, and transceiver 1103 are connected through bus 1104 and complete communication with each other.

[0096] Bus 1104 includes hardware, software, or both. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side BUS (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable bus or a combination of two or more of these. Where appropriate, bus 1104 may include one or more buses. Although embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0097] The following are embodiments of a computer-readable storage medium provided by embodiments of the present disclosure. The computer-readable storage medium and the child seat selection and evaluation method of the above embodiments belong to the same inventive concept. Details not described in detail in the embodiments of the computer-readable storage medium may refer to the embodiments of the child seat selection and evaluation method.

[0098] This embodiment provides a storage medium containing computer-executable instructions that, when executed by a computer processor, are used to execute a child seat selection and evaluation method.

[0099] Of course, the computer-executable instructions of a storage medium provided by embodiments of the present disclosure are not limited to the above method operations and may also execute related operations in the child seat selection and evaluation method provided by any embodiment of the present disclosure.

[0100] Through the above description of the embodiments, those skilled in the art can clearly understand that the present disclosure can be implemented by means of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present disclosure, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk, or optical disc of a computer, etc., including several instructions for causing a computer cloud platform (which can be a personal computer, server, or network cloud platform, etc.) to execute the child seat selection and evaluation method provided by each embodiment of the present disclosure.

[0101] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising a..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the above elements.

[0102] The above are only specific embodiments of the present disclosure, enabling those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments described above, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating the selection of a child seat, characterized in that, Including: Obtain a child seat selection data set, where the child seat selection data set at least includes collision pulse data of multiple vehicles, seat measurement data of multiple seats, and multiple injury result data. Among them, the injury result data is calculated based on the collision pulse data, a standard driving condition simulation model of the vehicle, and a seat simulation model of the child seat; Perform data correction on the child seat selection data set to obtain a standardized data set; Perform key feature parameterization on the standardized data set, and screen out waveform features significantly related to the target variable in the standardized data set through the Pearson correlation coefficient to obtain key feature data; Use the key feature data to train a pre-constructed child seat selection model using the random forest regression algorithm, including: performing ensemble learning on the input key feature data through multiple decision trees; Input the collision pulse data of the target vehicle into the trained child seat selection model to obtain a child seat suitable for the target vehicle; In response to the update of the child seat selection data set, use the updated child seat selection data set to update the child seat selection model.

2. The method according to claim 1, wherein The standard driving condition simulation model is calculated based on the rear seat data, front seat data, cockpit data, and safety configuration data of multiple vehicles.

3. The method according to claim 1, wherein The seat measurement data includes the size data of the child seat and the installation data of the child seat. The injury result data includes the head 3ms synthetic acceleration data of the child seat, the head injury data of the child seat, the neck force data of the child seat, and the chest compression displacement data of the child seat.

4. The method according to claim 1, wherein The performing data correction on the child seat selection data set to obtain a standardized data set includes: Use the interpolation method to repair the incomplete data in the collision pulse data.

5. The method according to claim 1, characterized in that, The performing data correction on the child seat selection data set to obtain a standardized data set includes: Process the outliers, outliers, and missing values in the seat measurement data and the injury result data, perform denoising processing and data correction processing on the seat measurement data and the injury result data, and unify the data units of the seat measurement data and the data units of the injury result data.

6. The method according to claim 1, wherein The waveform features include waveform activation duration, signal intensity integral, root mean square power, dominant harmonic amplitude, spectral extreme points, band energy integral, power spectrum centroid, and spectral randomness measure.

7. The method according to claim 1, characterized in that The inputting the collision pulse data of the target vehicle into the trained child seat selection model to obtain a child seat suitable for the target vehicle includes: Input the collision pulse data of the target vehicle into the child seat selection model to respectively obtain the injury result data of multiple candidate child seats on the target vehicle; Calculate the selection scores of multiple candidate child seats on the target vehicle according to the injury result data and the target regulations; Sort the multiple selection scores, and use the child seat with the highest selection score as the child seat suitable for the target vehicle.

8. A child seat selection and evaluation device, characterized in that, Including: A data and model acquisition module for acquiring a child seat selection data set, where the child seat selection data set at least includes collision pulse data of multiple vehicles, seat measurement data of multiple seats, and multiple injury result data, and the injury result data is calculated based on the collision pulse data, a standard driving condition simulation model of the vehicle, and a seat simulation model of the child seat; A data correction module for correcting the child seat selection data set to obtain a standardized data set; A feature extraction module for parameterizing key features of the standardized data set, and screening out waveform features significantly related to the target variable in the standardized data set through the Pearson correlation coefficient to obtain key feature data; A model training module for training a pre-constructed child seat selection model using the random forest regression algorithm with the key feature data, including: performing ensemble learning on the input key feature data through multiple decision trees; An evaluation module for inputting the collision pulse data of the target vehicle into the trained child seat selection model to obtain a child seat suitable for the target vehicle; An update module for updating the child seat selection model with the updated child seat selection data set in response to the update of the child seat selection data set.

9. An electronic device, characterized in that, Comprising: A processor; A memory for storing executable instructions; wherein, the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The storage medium stores a computer program, and when the computer program is executed by the processor, the processor is caused to implement the method according to any one of claims 1 to 7.

Images