Airbag control method, computer readable storage medium, computer program product and vehicle
By using target datasets and multilayer sensors in the automotive airbag control system, combined with preset thresholds and feature parameters, the problem of misjudgment caused by single threshold judgment is solved, and more accurate airbag control is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-29
- Publication Date
- 2026-04-10
AI Technical Summary
In existing technologies, automotive airbag control systems determine whether to trigger based on a single threshold, which has poor anti-interference capabilities and is prone to misjudgment under special collision conditions.
The selected values in the target dataset, the preset first threshold, and the preset second threshold are used to control the airbag's action. Combined with the preset model and feature parameters, the system makes accurate judgments through neural network models such as multilayer perceptrons, decision trees, or random forests.
It improves the accuracy of airbag detection, reduces misjudgments, and enhances safety under complex collision conditions.
Smart Images
Figure CN121822345A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of vehicles, in particular to an airbag control method, a computer readable storage medium, a computer program product and a vehicle. BACKGROUND
[0002] The airbag control system of an automobile is one of the most important active safety devices of a modern vehicle, and plays an irreplaceable role in vehicle safety protection. In the related art, a single threshold is used to determine whether the airbag needs to be triggered. The single threshold determination method has poor anti-interference ability, and in the face of special collision conditions, the determination method is prone to misjudgment. SUMMARY
[0003] The present application aims to at least solve one of the above technical problems in the art. To this end, the first object of the present application is to provide an airbag control method, comprising:
[0004] determining a selected value in a target data set, the target data set being target sensor data within a preset time length;
[0005] controlling the airbag to perform an action according to the selected value, a preset first threshold and a preset second threshold; wherein the preset first threshold is smaller than the preset second threshold.
[0006] By controlling the airbag to perform an action according to the selected value in the target data set and the preset first threshold and the preset second threshold, the determination of the airbag to perform an action is more accurate than the single threshold determination method in the related art, and the misjudgment is reduced.
[0007] According to some embodiments of the present application, determining a selected value in a target data set comprises:
[0008] determining the maximum value or the minimum value in the target data set as the selected value.
[0009] According to some embodiments of the present application, controlling the airbag to perform an action according to the selected value, a preset first threshold and a preset second threshold comprises:
[0010] controlling the airbag to perform a first action when the selected value is less than the preset first threshold; and / or
[0011] controlling the airbag to perform a second action when the selected value is greater than the preset second threshold; and / or
[0012] controlling the airbag to perform the first action or the second action according to the target data set and a preset model when the selected value is greater than the preset first threshold and less than the preset second threshold.
[0013] According to some embodiments of the present application, the controlling the airbag to perform the first action or the second action according to the target data set and the preset model comprises:
[0014] According to the target data set, a sub-data set is determined, the sub-data set being a data set greater than a preset first threshold and less than a preset second threshold in the target data set;
[0015] According to the sub-data set and the preset model, the airbag is controlled to perform the first action or the second action.
[0016] According to some embodiments of the present application, the controlling the airbag to perform the first action or the second action according to the sub-data set and the preset model comprises:
[0017] According to the preset characteristic parameter, a characteristic parameter value of the sub-data set is calculated;
[0018] The characteristic parameter value of the sub-data set is input into the preset model to determine an output state value, the output state value being used to control the airbag to perform the first action or the second action.
[0019] According to some embodiments of the present application, the output state value comprises a first state value and a second state value,
[0020] In a case where the preset model outputs the first state value, the airbag is controlled to perform the first action;
[0021] In a case where the preset model outputs the second state value, the airbag is controlled to perform the second action.
[0022] According to some embodiments of the present application, the preset characteristic parameter is obtained according to the following method:
[0023] Any data in a test data set is labeled to obtain a test labeled data set,
[0024] According to the test labeled data set, a test sub-data set is determined, the test sub-data set being data greater than a preset first threshold and less than a preset second threshold in the test labeled data set;
[0025] According to the test sub-data set, a preset characteristic parameter is determined;
[0026] The test data set is target sensor data in a preset time length in a standard crash test phase, and the target sensor is a.
[0027] According to some embodiments of the present application, the preset characteristic parameter is determined according to the test sub-data set, comprising:
[0028] b alternative characteristic parameters of the test sub-data set are calculated;
[0029] An F value of analysis of variance of the a*b alternative characteristic parameters is calculated to obtain a*b F values of analysis of variance.
[0030] The a*b variance analysis F values are arranged from large to small, and the preset characteristic parameters corresponding to the first s variance analysis F values are selected as the candidate characteristic parameters.
[0031] According to some embodiments of the present application, the candidate characteristic parameters are at least two of the following data: mean, standard deviation, unbiased kurtosis, unbiased skewness, maximum value, minimum value, median, zero-crossing rate, energy, signal amplitude change rate, signal entropy, integral, absolute value integral, total energy in frequency domain, main frequency, frequency center.
[0032] According to some embodiments of the present application, the preset model is obtained according to the following method:
[0033] According to the preset characteristic parameters, the characteristic parameter values of the test sub-data set are calculated.
[0034] The characteristic parameter values of the test sub-data set are input into the preset neural network model, and an output state value is obtained to train the preset model, and the output state value is used to control the airbag to perform an action.
[0035] According to some embodiments of the present application, the preset neural network model is one of a multilayer perceptron, a decision tree, and a random forest.
[0036] According to some embodiments of the present application, the preset first threshold and the preset second threshold are obtained according to the following method:
[0037] Any data in the test data set is labeled to obtain a test labeled data set, and the test data set is target sensor data within a preset time length in a standard crash test phase;
[0038] According to the test labeled data set, the preset first threshold and the preset second threshold are calculated.
[0039] According to some embodiments of the present application, according to the test labeled data set, the preset first threshold and the preset second threshold are calculated, including:
[0040] According to the first test labeled data set, the preset first threshold is calculated, and the first test labeled data set is related to the second action of the airbag.
[0041] According to the second test labeled data set, the preset second threshold is calculated, and the second test labeled data set is related to the first action of the airbag.
[0042] According to some embodiments of the present application, according to the first test labeled data set, the preset first threshold is calculated, including calculating the preset first threshold according to the following formula:
[0043] T low = α · min(max(x1), max(x2),..., max(xn ));
[0044] wherein T low is a preset first threshold value, x1, x2,... x n comprise the first test labeled data set, the first test labeled data set is obtained when the airbag is controlled to perform the second action in the standard crash test stage, a is a constant greater than 0 and less than 1, and n is the number of crashes.
[0045] According to some embodiments of the present application, the preset second threshold value is calculated according to the second test labeled data set, including calculating the preset second threshold value according to the following formula:
[0046]
[0047] wherein T height is a preset second threshold value, y1, y2,... y m comprise the second test labeled data set, the second test labeled data set is obtained when the airbag is controlled to perform the first action in the standard crash test stage, a is a constant greater than 0 and less than 1, and n is the number of crashes.
[0048] According to some embodiments of the present application, the target sensor includes at least one of an airbag ECU x-direction acceleration sensor, an airbag ECU y-direction acceleration sensor, a left front impact sensor, a right front impact sensor, a left B-pillar acceleration sensor, a right B-pillar acceleration sensor, a left door pressure sensor, and a right door pressure sensor.
[0049] According to some embodiments of the present application, the preset time length is 20 ms.
[0050] Another aspect of an embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the airbag control method described above.
[0051] Another aspect of an embodiment of the present application provides a computer program product including a computer program, the computer program being executed by a processor to implement the airbag control method described above.
[0052] Another aspect of an embodiment of the present application provides a vehicle including the computer readable storage medium described above, or including the computer program product described above. BRIEF DESCRIPTION OF DRAWINGS
[0053] Figure 1 is a flowchart of an airbag control method according to an embodiment of the present application;
[0054] Figure 2 is a use flowchart of an airbag control method according to an embodiment of the present application;
[0055] Figure 3 This is a flowchart illustrating the determination of preset characteristic parameters in an airbag control method according to an embodiment of the present invention. Detailed Implementation
[0056] Embodiments of the present invention are described in detail below. Examples of these embodiments are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.
[0057] The following description, with reference to the accompanying drawings, illustrates an airbag control method according to an embodiment of the present invention, including:
[0058] S1: Determine the selected value in the target dataset, which is the target sensor data within a preset time period;
[0059] S2: Control the airbag to perform actions based on the selected value, the preset first threshold and the preset second threshold; wherein, the preset first threshold is less than the preset second threshold.
[0060] like Figure 1 Specifically, when a vehicle collision occurs, any sensor installed in the vehicle will collect multiple time-domain data within a preset time period. These multiple time-domain data constitute a target dataset. Therefore, when the vehicle has multiple sensors, each sensor has a target dataset. For a target sensor, a selected value is selected from the target dataset, and the selected value is compared with a preset first threshold and a preset second threshold to further control the airbag's activation. In some embodiments, the airbag activation includes either the airbag deploying or not deploying. According to some embodiments of the present invention, determining the selected value in the target dataset includes:
[0061] S10: Determine the maximum or minimum value in the target dataset as the selected value.
[0062] Specifically, in some embodiments, by selecting the maximum value in the target dataset as the selected value, the maximum value in the target dataset is compared with a preset first threshold and a preset second threshold, thereby further controlling the airbag's action, i.e., controlling the airbag to deploy or controlling the airbag not to deploy. In some other embodiments, the selected value can also be other values selected from the target dataset depending on the subsequent judgment method. Since multiple sensors are installed in the vehicle, each sensor corresponds to a target dataset and can select a selected value from the target dataset. According to some embodiments of the present invention, controlling the airbag's action includes controlling the airbag to perform a first action or controlling the airbag to perform a second action. Controlling the airbag's action based on the selected value, the preset first threshold, and the preset second threshold includes:
[0063] S20: If the selected value is less than a preset first threshold, control the airbag to perform a first action; and / or
[0064] S21: If the selected value is greater than the preset second threshold, control the airbag to perform the second action; and / or
[0065] S22: When the selected value is greater than the preset first threshold and less than the preset second threshold, control the airbag to perform actions according to the target dataset and the preset model.
[0066] like Figure 2 As shown, specifically, controlling the airbag to perform the first action (preventing airbag deployment) and the second action (deploying airbag) involves controlling the airbag to deploy. For any sensor in the vehicle, the selected value is the maximum value in the target dataset. When the vehicle has multiple sensors, each sensor corresponds to a target dataset and a maximum value within that dataset. Each sensor also corresponds to a set of preset first and second thresholds. Therefore, when the selected values in all sensor target datasets are less than their respective preset first thresholds, the airbag performs the first action. When the selected values in all sensor target datasets are greater than their respective preset second thresholds, the airbag performs the second action. When the selected value in any target dataset is greater than the corresponding preset first threshold and less than the corresponding preset second threshold, the airbag is controlled to perform either the first or second action based on the target dataset and the preset model.
[0067] According to some embodiments of the present invention, controlling the airbag to perform actions based on a target dataset and a preset model includes:
[0068] S220: determining a sub-data set according to the target data set, the sub-data set being a data set greater than a preset first threshold and less than a preset second threshold in the target data set;
[0069] S221: controlling the airbag to perform an action according to the sub-data set and a preset model.
[0070] Specifically, the target data set has too much data, and the data less than the preset first threshold corresponds to the airbag not firing, and the data greater than the preset second threshold corresponds to the airbag firing, so only the data greater than the preset first threshold and less than the preset second threshold needs to be selected as the sub-data set, and the airbag can be controlled to perform an action according to the sub-data set and the preset model. The data in the sub-data set can be processed in multiple ways to obtain different types of processed data.
[0071] According to some embodiments of the present application, controlling the airbag to perform an action according to the sub-data set and a preset model includes:
[0072] S2210: calculating a feature parameter value of the sub-data set according to a preset feature parameter;
[0073] S2211: inputting the feature parameter value of the sub-data set into the preset model to determine an output state value, the output state value being used to control the airbag to perform an action.
[0074] The output state value includes a first state value and a second state value,
[0075] In the case that the preset model outputs the first state value, the airbag is controlled to perform the first action;
[0076] In the case that the preset model outputs the second state value, the airbag is controlled to perform the second action.
[0077] Specifically, since the target data set has a large amount of data, a preset model is selected, wherein the preset model is a neural network model, the feature parameter value of the sub-data set is input into the neural network model, the neural network model outputs a first state value or a second state value, in some embodiments, the first state value is 1 and the second state value is 0, when the first state value is output, the airbag is controlled to fire, and when the second state value is output, the airbag is controlled not to fire. In some embodiments, the target sensor is provided with multiple target sensors, the preset feature parameter is determined by a pre-process, and for multiple target sensors, the number of preset feature parameters is determined to be s, but the s preset feature parameters are not completely calculated from the target data set obtained by all target sensors in the pre-determination process, and even in some embodiments, the s preset feature parameters are calculated from the target data set obtained by the same target sensor in the pre-determination process.
[0078] According to some embodiments of the present application, the preset first threshold value and the preset second threshold value are obtained according to the following method:
[0079] S3: Label any data in the test data set to obtain a test labeled data set, the test data set being target sensor data within a preset time length in a standard crash test phase;
[0080] According to the test labeled data set, the preset first threshold value and the preset second threshold value are calculated.
[0081] As shown in Figure 3 , specifically, the preset model needs to be trained by a standard crash test, and before the model training, the preset first threshold value and the preset second threshold value need to be determined, so the standard crash test needs to be performed first, and in the standard crash test process, the data collected by the target sensor when the safety airbag performs the second action (i.e. the safety airbag is detonated) is labeled;
[0082] and the data collected by the target sensor when the safety airbag performs the first action (i.e. the safety airbag is not detonated) is labeled;
[0083] According to the data labeled by the above two processes, a test labeled data set is obtained, and the preset first threshold value and the preset second threshold value are calculated according to the test labeled data set.
[0084] In some implementations, the target sensor is provided with a plurality of target sensors, each target sensor corresponding to a set of preset first threshold values and preset second threshold values, and for each target sensor, the preset first threshold value and the preset second threshold value are calculated in the same way as described above.
[0085] According to some embodiments of the present application, the test labeled data set includes a first test labeled data set and a second test labeled data set, and according to the test labeled data set, the preset first threshold value and the preset second threshold value are calculated, including:
[0086] According to the first test labeled data set, the preset first threshold value is calculated, the first test labeled data set being related to the second action of the safety airbag;
[0087] According to the second test labeled data set, the preset second threshold value is calculated, the second test labeled data set being related to the first action of the safety airbag.
[0088] Specifically, in the crash test, for any given sensor, the sensor data corresponding to the airbag's second action (airbag deployment) is labeled as the first test labeled data. Multiple first test labeled data sets within a preset time period constitute the first test labeled dataset. Similarly, the sensor data corresponding to the airbag's first action (airbag non-deployment) is labeled as the second test labeled data. Multiple second test labeled data sets within a preset time period constitute the second test labeled dataset. For a vehicle equipped with multiple target sensors, each target sensor can obtain a set of first test labeled datasets and a set of second test labeled datasets. Furthermore, each target sensor can calculate a set of preset first thresholds and preset second thresholds.
[0089] According to some embodiments of the present invention, calculating a preset first threshold based on a first experimental labeled dataset includes calculating the preset first threshold according to the following formula:
[0090] T low =α·min(max(x1),max(x2),...,max(x n (1)
[0091] Among them, T low Let x1, x2, ..., x be the preset first threshold. n The first experimental annotation dataset is formed, which is obtained when the airbag is controlled to perform the second action during the standard crash test phase. α is a constant greater than 0 and less than 1, and n is the number of crashes.
[0092] Specifically, in some embodiments, α = 0.9, and in some other embodiments, α can be set to other values, depending on the preset first threshold and the preset second threshold and the situation of controlling the airbag deployment in the standard crash test.
[0093] According to some embodiments of the present invention, calculating a preset second threshold based on a second experimental labeled dataset includes calculating the preset second threshold according to the following formula:
[0094]
[0095] Among them, T height For the preset second threshold, y1, y2, ..., y m The second experimental annotation dataset is formed, which is obtained when the airbag is controlled to perform the first action during the standard crash test phase. α is a constant greater than 0 and less than 1, and n is the number of crashes.
[0096] Specifically, in some embodiments, a = 0.9, and in some other embodiments, a can also be set to other values, depending on the preset first threshold and the preset second threshold and the situation of controlling the airbag to be triggered in the standard collision test.
[0097] When the situation of T height > occurs, the value of a is reduced and the preset first threshold and the preset second threshold are recalculated. height B
[0098] According to some embodiments of the present application, the target sensor is provided with a, and the preset feature parameter is obtained according to the following method:
[0099] The test sub-data set is determined according to the test labeled data set, and the test sub-data set is the data greater than the preset first threshold and less than the preset second threshold in the test labeled data set;
[0100] The b alternative feature parameters of the test sub-data set are calculated;
[0101] The analysis of variance F value of the a*b alternative feature parameters is calculated, and a*b analysis of variance F values are obtained;
[0102] The a*b analysis of variance F values are arranged from large to small, and the alternative feature parameters corresponding to the first s analysis of variance F values are selected as the preset feature parameters.
[0103] Specifically, in some embodiments, the number of target sensors is 8, and the alternative feature parameters of each target sensor are 16, so all the alternative feature parameters are 16*8 = 128, the analysis of variance F value of the 128-dimensional feature is calculated, and the calculation formula is as follows:
[0104]
[0105] Among them, MS B is the inter-group mean square, indicating the variance between different categories; MS W is the intra-group mean square, indicating the variance within each category.
[0106] The larger the analysis of variance F value is, the greater the correlation between the alternative feature parameter and the control of the airbag to execute the action is, and the more helpful the alternative feature parameter is to the determination of the airbag to execute the action. The analysis of variance F values are sorted in descending order, and the first 10-dimensional alternative feature parameters with the highest correlation are taken as the final preset feature parameters.
[0107] According to some embodiments of the present application, the alternative feature parameter is at least two of the mean, the standard deviation, the unbiased kurtosis, the unbiased skewness, the maximum value, the minimum value, the median, the zero-crossing rate, the energy, the signal amplitude change rate, the signal entropy, the integral, the absolute value integral, the total energy in the frequency domain, the main frequency, and the frequency center.
[0108] Specifically, according to the order of mean, standard deviation, unbiased kurtosis, unbiased skewness, maximum value, minimum value, median, zero-crossing rate, energy, signal amplitude change rate, signal entropy, integral, absolute value integral, total energy in frequency domain, main frequency, frequency center, the calculation formulas of the above characteristic parameters are arranged as follows:
[0109]
[0110] max(x) = max(x1, x2,..., xn) (10) n
[0111] min(x) = min(x1, x2,..., xn) (11) n
[0112]
[0113]
[0114] According to some embodiments of the present application, the preset model is obtained according to the following method:
[0115] According to the preset characteristic parameters, the characteristic parameter values of the test sub-data set are calculated;
[0116] The characteristic parameter values of the test sub-data set are input into the preset neural network model, and an output state value is obtained to train the preset model, and the output state value is used to control the airbag to perform an action.
[0117] According to some embodiments of the present application, the preset neural network model is one of a multi-layer perceptron, a decision tree, and a random forest.
[0118] Specifically, the preset neural network model can also have more choices, such as a convolutional neural network, etc., and the specific preset neural network model is not limited here.
[0119] According to some embodiments of the present application, the target sensor includes at least one of an airbag ECU x-direction acceleration sensor, an airbag ECU y-direction acceleration sensor, a left front impact sensor, a right front impact sensor, a left B-pillar acceleration sensor, a right B-pillar acceleration sensor, a left door pressure sensor, and a right door pressure sensor.
[0120] Specifically, the above sensors are all sensors with relatively strong relevance to whether the airbag is detonated, so the above sensor data is input into the preset model as a target data set. In some other embodiments, other sensors related to whether the airbag is detonated can also be provided on the vehicle and data collected by the other sensors can be obtained, and the specific sensor setting position and setting number are not limited here.
[0121] According to some embodiments of the present application, the preset time length is 20 ms.
[0122] Specifically, in some embodiments, the sensor data within 20 ms is already sufficient to implement the above-mentioned airbag control method, and in some other embodiments, the preset time length can be adjusted as long as sufficient data can be collected within the preset time length to implement the above-mentioned method.
[0123] Another aspect of the embodiments of the present application provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the above-mentioned airbag control method.
[0124] Another aspect of the embodiments of the present application provides a computer program product comprising a computer program, the computer program being executed by a processor to implement the above-mentioned airbag control method.
[0125] Another aspect of the embodiments of the present application provides a vehicle comprising the above-mentioned computer readable storage medium or comprising the above-mentioned computer program product.
[0126] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.
[0127] In addition, the terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include any of the features. In the description of the present application, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise specifically limited.
[0128] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0129] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.
[0130] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in any embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0131] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.
Claims
1. A method for controlling an airbag, characterized in that, include: Determine selected values in the target dataset, where the target dataset is target sensor data within a preset time period; The airbag is controlled to perform actions based on the selected value, a preset first threshold, and a preset second threshold; wherein the preset first threshold is less than the preset second threshold.
2. The airbag control method according to claim 1, characterized in that, Determining the selected values in the target dataset includes: The maximum or minimum value in the target dataset is determined to be the selected value.
3. The airbag control method according to claim 1, characterized in that, The step of controlling the airbag to perform actions based on the selected value, a preset first threshold, and a preset second threshold includes: If the selected value is less than the preset first threshold, control the airbag to perform a first action; and / or If the selected value is greater than the preset second threshold, control the airbag to perform a second action; and / or When the selected value is greater than the preset first threshold and less than the preset second threshold, the airbag is controlled to perform the first action or the second action according to the target dataset and the preset model.
4. The airbag control method according to claim 3, characterized in that, The step of controlling the airbag to perform the first action or the second action according to the target dataset and the preset model includes: A subset of data is determined based on the target dataset, wherein the subset of data is a set of data in the target dataset that is greater than the preset first threshold and less than the preset second threshold; The airbag is controlled to perform either the first action or the second action based on the subset of data and the preset model.
5. The airbag control method according to claim 4, characterized in that, The step of controlling the airbag to perform the first action or the second action based on the subset dataset and the preset model includes: Calculate the feature parameter values of the subset dataset based on preset feature parameters; The feature parameter values of the subset are input into the preset model to determine the output state value, which is used to control the airbag to perform the first action or the second action.
6. The airbag control method according to claim 5, characterized in that, The output state value includes a first state value and a second state value. When the preset model outputs the first state value, the airbag is controlled to perform the first action; When the preset model outputs the second state value, the airbag is controlled to perform the second action.
7. The airbag control method according to claim 5, characterized in that, The preset feature parameters are obtained according to the following method: Label any data point in the experimental dataset to obtain the experimental labeled dataset. A test subset is determined based on the test labeled dataset, wherein the test subset is the data in the test labeled dataset that is greater than the preset first threshold and less than the preset second threshold; The preset feature parameters are determined based on the experimental subset dataset; The test dataset consists of target sensor data from the standard crash test phase within the preset time period, and there are a target sensors.
8. The airbag control method according to claim 7, characterized in that, Determining the preset feature parameters based on the experimental subset dataset includes: Calculate b candidate feature parameters for the experimental subset dataset; Calculate the F-values of the analysis of variance for a*b candidate feature parameters to obtain a*b F-values of the analysis of variance; Arrange the a*b variance analysis F values from largest to smallest, and select the candidate feature parameters corresponding to the first s variance analysis F values as the preset feature parameters.
9. The airbag control method according to claim 8, characterized in that, The candidate feature parameters are at least two of the following data: mean, standard deviation, unbiased kurtosis, unbiasedness, maximum value, minimum value, median, zero-crossing rate, energy, signal amplitude change rate, signal entropy, integral, absolute value integral, total energy in the frequency domain, dominant frequency, and frequency center.
10. The airbag control method according to claim 7, characterized in that, The preset model is obtained according to the following method: Calculate the feature parameter values of the experimental subset dataset based on the preset feature parameters; The feature parameter values of the experimental subset are input into a preset neural network model and the output state value is obtained to train the preset model. The output state value is used to control the airbag to perform actions.
11. The airbag control method according to claim 10, characterized in that, The preset neural network model is one of the following: multilayer perceptron, decision tree, and random forest.
12. The airbag control method according to any one of claims 1-11, characterized in that, The preset first threshold and the preset second threshold are obtained according to the following method: Any data in the test dataset is labeled to obtain the test labeled dataset, which is the target sensor data of the standard crash test phase within the preset time period; Based on the experimental labeled dataset, calculate the preset first threshold and the preset second threshold.
13. The airbag control method according to claim 12, characterized in that, The step of calculating the preset first threshold and the preset second threshold based on the experimental labeled dataset includes: The preset first threshold is calculated based on the first test annotation dataset, wherein the first test annotation dataset is related to the second action of the airbag; The preset second threshold is calculated based on the second test annotation dataset, which is related to the first action of the airbag.
14. The airbag control method according to claim 13, characterized in that, The step of calculating the preset first threshold based on the first experimental labeled dataset includes calculating the preset first threshold according to the following formula: T low =α·min(max(x1),max(x2),...,max(x n )); Among them, T low For a preset first threshold, x1, x2, ..., x n The first test annotation dataset is formed by controlling the airbag to perform the second action during the standard crash test phase, where α is a constant greater than 0 and less than 1, and n is the number of crashes.
15. The airbag control method according to claim 13, characterized in that, The step of calculating the preset second threshold based on the second experimental labeled dataset includes calculating the preset second threshold according to the following formula: Among them, T height For the preset second threshold, y1, y2, ..., y m The second experimental annotation dataset is formed by controlling the airbag to perform the first action during the standard crash test phase, where α is a constant greater than 0 and less than 1, and n is the number of crashes.
16. The airbag control method according to any one of claims 1-11, characterized in that, The target sensor includes at least one of the following: airbag ECU x-direction acceleration sensor, airbag ECU y-direction acceleration sensor, left front collision sensor, right front collision sensor, left B-pillar acceleration sensor, right B-pillar acceleration sensor, left door pressure sensor, and right door pressure sensor.
17. The airbag control method according to any one of claims 1-11, characterized in that, The preset duration is 20ms.
18. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the airbag control method as described in any one of claims 1-17.
19. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the airbag control method as described in any one of claims 1-17.
20. A vehicle, characterized in that, It includes the computer-readable storage medium of claim 18, or the computer program product of claim 19.