Air bag restraint system control method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of vehicles, in particular to a safety airbag control method and device, electronic equipment and a storage medium, and the method comprises the steps that the current collision working condition of a vehicle is recognized, and current collision data corresponding to the current collision working condition in a preset sampling interval is obtained, and according to the current collision data, a corresponding target safety air bag explosion algorithm is matched, and when the current collision data meets a preset ignition condition, the vehicle is controlled to ignite a safety air bag. Therefore, the problems that a collision control algorithm adopted in the related technology is difficult to distinguish the characteristics of different collision working conditions and meet the provided RTTF, meanwhile, the complexity of a prediction model is increased through a large number of auxiliary algorithms, and therefore the robustness of a point explosion algorithm is reduced are solved, and through model training and characteristic screening, the accuracy of the point explosion algorithm is improved. And an air bag ignition algorithm of the support vector machine is constructed, so that the characteristics of different working conditions are effectively distinguished, and the robustness of the air bag ignition algorithm is improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a control method, device, electronic device and storage medium for an airbag. Background Art

[0002] With the widespread use of vehicles, the safety performance of vehicles has become one of the important factors for users to choose. Among them, the car airbag is a passive safety device in the car's passive safety system, providing a supplementary safety measure and is often used as the last resort in collision accidents. Since it is directly related to the user's life safety, it is particularly important whether the airbag can be detonated at the right time.

[0003] In the related art, when a vehicle has a frontal collision accident, such as Figure 1 As shown, most of them receive acceleration signals from front collision sensors and built-in sensors through airbag controllers for analysis, and send ignition commands to restraint system devices such as airbags and seat belt warnings without user intervention, while triggering the fuel and power-off devices to cut off the fuel system circuit and the electrical system circuit, thereby achieving the purpose of protecting personnel and system safety.

[0004] However, the collision control algorithm used in the related technology is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF (Required Time To Fire). At the same time, a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the ignition algorithm, which urgently needs to be solved. Summary of the invention

[0005] The present application provides an airbag control method, device, electronic device and storage medium to solve the problems that the collision control algorithm used in the related art is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF, and a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the detonation algorithm.

[0006] The first aspect of the present application provides a method for controlling an airbag, comprising the following steps:

[0007] Identify the current collision condition of the vehicle;

[0008] Acquire current collision data corresponding to the current collision condition within a preset sampling interval, match a corresponding target airbag detonation algorithm according to the current collision data, and determine whether the current collision data meets a preset detonation condition through the target airbag detonation algorithm;

[0009] If the current collision data meets the preset ignition condition, the vehicle is controlled to ignite the airbag.

[0010] Optionally, before matching the corresponding target airbag detonation algorithm according to the current collision data, the method further includes:

[0011] Obtaining a preset ignition condition corresponding to each preset collision condition of the vehicle;

[0012] Obtaining a detonation training data set of the airbag according to a preset ignition condition corresponding to each preset collision condition, inputting the detonation training data set of the airbag into a preset prediction detonation model of the airbag to obtain a current support vector, and determining feature data corresponding to each preset collision condition based on the current support vector;

[0013] An airbag detonation algorithm corresponding to each preset collision condition is constructed according to the characteristic data corresponding to each preset collision condition.

[0014] Optionally, obtaining the detonation training data set of the airbag according to the preset ignition condition corresponding to each preset collision condition includes:

[0015] Acquire a collision data matrix corresponding to each of the preset collision working conditions, and simulate and obtain a mirror collision data matrix corresponding to each of the preset collision working conditions according to the collision data matrix corresponding to each of the preset collision working conditions;

[0016] Obtaining the mirror collision data corresponding to each preset collision condition according to the mirror collision data matrix corresponding to each preset collision condition, and determining the scaling ratio of the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition based on the preset ignition condition corresponding to each preset collision condition;

[0017] Scaling the collision data corresponding to each preset collision condition and the mirror collision data based on the scaling ratio of the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition, outputting a collision data matrix set corresponding to each preset collision condition, and traversing the collision data matrix set corresponding to each preset collision condition, and outputting first filtered data and second filtered data corresponding to each preset collision condition;

[0018] Obtaining a target vector of the first filter data corresponding to each preset collision condition at a target time, determining the on / off state of the airbag detonation algorithm corresponding to each preset collision condition according to the target vector, and when the airbag detonation algorithm is on, obtaining the target filter data of the second filter data of the preset collision condition corresponding to the on state of the airbag detonation algorithm corresponding to each preset collision condition at the target time, and calculating the collision feature data at the target time according to the target filter data and the collision data corresponding to each preset collision condition,

[0019] Determine the characteristic vector of the collision characteristic data, and when generating the collision characteristic data sample at the target moment, obtain the collision characteristic data training sample at the target moment according to the collision characteristic data sample and the characteristic vector of the collision characteristic data, and traverse the collision characteristic data training sample and all running moments of the airbag detonation algorithm corresponding to each preset collision condition to obtain the airbag detonation training data set corresponding to each preset collision condition.

[0020] Optionally, before obtaining the collision feature data training sample at the target moment according to the collision feature data sample and the feature vector of the collision feature data, the method further includes:

[0021] If the collision feature data sample is not generated at the target moment, the airbag detonation algorithm corresponding to each preset collision condition is controlled to enter the next moment of the target moment, and the step of obtaining the target vector of the first filtered data corresponding to each preset collision condition at the target moment is continued.

[0022] Optionally, determining the current feature data corresponding to each preset collision condition based on the current support vector includes:

[0023] Based on each of the preset collision conditions, a detonation condition and a non-detonation condition of the vehicle are obtained, and a first label sample and first feature data belonging to the detonation condition are obtained in the support vector, and a second label sample and second feature data belonging to the non-detonation condition are obtained in the support vector;

[0024] The first characteristic data and the second characteristic data are checked to obtain a difference value between the first characteristic data and the second characteristic data, and when the difference value is less than a preset threshold, current characteristic data corresponding to each preset collision condition is determined according to the difference value.

[0025] According to the airbag control method of the embodiment of the present application, the current collision condition of the vehicle is identified, the current collision data corresponding to the current collision condition within the preset sampling interval is obtained, the corresponding target airbag detonation algorithm is matched according to the current collision data, and when the current collision data meets the preset ignition condition, the vehicle is controlled to detonate the airbag. Thus, the problem that the collision control algorithm used in the related art is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF, and a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the detonation algorithm, etc., is solved. Through model training and feature screening, the airbag detonation algorithm of the support vector machine is constructed, so as to effectively distinguish the characteristics of different conditions and improve the robustness of the airbag detonation algorithm.

[0026] A second aspect of the present application provides an airbag control device, including:

[0027] An identification module, used for identifying a current collision condition of the vehicle;

[0028] A judgment module, used to obtain current collision data corresponding to the current collision condition within a preset sampling interval, match the corresponding target airbag detonation algorithm according to the current collision data, and judge whether the current collision data meets the preset ignition condition through the target airbag detonation algorithm;

[0029] The control module is used for controlling the vehicle to ignite the airbag if the current collision data meets the preset ignition condition.

[0030] Optionally, before matching the corresponding target airbag detonation algorithm according to the current collision data, the judgment module further includes:

[0031] An acquisition unit, used for acquiring a preset ignition condition corresponding to each preset collision condition of the vehicle;

[0032] a determination unit, configured to obtain a detonation training data set of the airbag according to a preset ignition condition corresponding to each preset collision condition, input the detonation training data set of the airbag into a preset prediction detonation model of the airbag to obtain a current support vector, and determine feature data corresponding to each preset collision condition based on the current support vector;

[0033] A construction unit is used to construct an airbag detonation algorithm corresponding to each preset collision condition according to the characteristic data corresponding to each preset collision condition.

[0034] Optionally, the determining unit includes:

[0035] A first acquisition subunit is used to acquire a collision data matrix corresponding to each preset collision working condition, and simulate the collision data matrix corresponding to each preset collision working condition to obtain a mirror collision data matrix corresponding to each preset collision working condition;

[0036] a second acquisition subunit, configured to obtain the mirror collision data corresponding to each preset collision condition according to the mirror collision data matrix corresponding to each preset collision condition, and determine a scaling ratio between the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition based on a preset ignition condition corresponding to each preset collision condition;

[0037] a first traversal subunit, configured to scale the collision data corresponding to each preset collision condition and the mirror collision data based on a scaling ratio of the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition, output a collision data matrix set corresponding to each preset collision condition, traverse the collision data matrix set corresponding to each preset collision condition, and output first filtered data and second filtered data corresponding to each preset collision condition;

[0038] a third acquisition subunit, configured to acquire a target vector of the first filter data corresponding to each preset collision condition at a target time, determine the on / off state of the airbag detonation algorithm corresponding to each preset collision condition according to the target vector, and acquire the target filter data of the second filter data of the preset collision condition corresponding to the airbag detonation algorithm corresponding to each preset collision condition in the on state at the target time when the airbag detonation algorithm is on, and calculate the collision feature data at the target time according to the target filter data and the collision data corresponding to each preset collision condition,

[0039] The second traversal subunit is used to determine the feature vector of the collision feature data, and when generating the collision feature data sample at the target moment, obtain the collision feature data training sample at the target moment according to the collision feature data sample and the feature vector of the collision feature data, and traverse the collision feature data training sample and all running moments of the airbag detonation algorithm corresponding to each preset collision condition to obtain the airbag detonation training data set corresponding to each preset collision condition.

[0040] Optionally, before obtaining the collision feature data training sample at the target moment according to the collision feature data sample and the feature vector of the collision feature data, the second traversal subunit further includes:

[0041] A control subcomponent is used to control the airbag detonation algorithm corresponding to each preset collision condition to enter the next moment of the target moment if the collision characteristic data sample is not generated at the target moment, and continue to execute the step of obtaining the target vector of the first filtered data corresponding to each preset collision condition at the target moment.

[0042] Optionally, the determining unit includes:

[0043] a fourth acquisition subunit, configured to acquire, based on each of the preset collision conditions, a detonation condition and a non-detonation condition of the vehicle, and acquire a first label sample in the support vector and first feature data belonging to the detonation condition, and acquire a second label sample in the support vector and second feature data belonging to the non-detonation condition;

[0044] A determination subunit is used to check the first characteristic data and the second characteristic data to obtain a difference value between the first characteristic data and the second characteristic data, and when the difference value is less than a preset threshold, determine the current characteristic data corresponding to each preset collision condition according to the difference value.

[0045] According to the airbag control device of the embodiment of the present application, the current collision condition of the vehicle is identified, the current collision data corresponding to the current collision condition within the preset sampling interval is obtained, the corresponding target airbag detonation algorithm is matched according to the current collision data, and the vehicle is controlled to detonate the airbag when the current collision data meets the preset ignition condition. Thus, the problem that the collision control algorithm used in the related art is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF, and a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the detonation algorithm, etc., is solved. Through model training and feature screening, the airbag detonation algorithm of the support vector machine is constructed, so as to effectively distinguish the characteristics of different conditions and improve the robustness of the airbag detonation algorithm.

[0046] A third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the airbag control method as described in the above embodiment.

[0047] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the airbag control method as described in the above embodiment.

[0048] A fifth aspect of the present application provides a computer program product, including a computer program, which is executed to implement the airbag control method described in the above embodiment.

[0049] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0051] Figure 1 A schematic diagram of the distribution of peripheral sensors and ECU (Electronic Control Unit) in the related technology;

[0052] Figure 2A schematic diagram of a FRB (Full-width Rigid Barrier, 100% overlap rigid barrier collision test) collision condition of the related technology;

[0053] Figure 3 A schematic diagram of a frontal collision condition with an offset in the related art;

[0054] Figure 4 A schematic diagram of a frontal drill card collision working condition of the related technology;

[0055] Figure 5 A flowchart of a method for controlling an airbag according to an embodiment of the present application;

[0056] Figure 6 A flow chart of prediction model training according to one embodiment of the present application;

[0057] Figure 7 An internal flow chart of a training data generation model according to one embodiment of the present application;

[0058] Figure 8 This is a feature screening flow chart according to an embodiment of the present application;

[0059] Fig. 9 is a block diagram of an airbag control device according to an embodiment of the present application;

[0060] Fig.10 Schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0061] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0062] The following describes the control method, device, electronic device and storage medium of the airbag of the embodiment of the present application with reference to the accompanying drawings. Aiming at the problem that the collision control algorithm used in the related art mentioned in the above background technology is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF, and a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the detonation algorithm, the present application provides a control method for the airbag, in which the current collision condition of the vehicle is identified, the current collision data corresponding to the current collision condition within a preset sampling interval is obtained, the corresponding target airbag detonation algorithm is matched according to the current collision data, and when the current collision data meets the preset ignition condition, the vehicle is controlled to detonate the airbag. Thus, the problem that the collision control algorithm used in the related art is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF, and a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the detonation algorithm, etc., is solved, and the airbag detonation algorithm of the support vector machine is constructed through model training and feature screening, so as to effectively distinguish the characteristics of different conditions and improve the robustness of the airbag detonation algorithm.

[0063] Specifically, before introducing the embodiments of the present application, the airbag detonation strategy in the related art is first introduced. Figure 1 As shown, the strategy currently adopted by most OEMs is that when a frontal collision occurs, the airbag controller receives acceleration signals from the front collision sensor and the built-in sensor for analysis, and sends ignition commands to restraint system devices such as airbags and seat belt warnings without user intervention, and at the same time triggers the fuel and power-off devices to cut off the fuel system circuit and the electrical system circuit, thereby achieving the purpose of protecting personnel and system safety.

[0064] Furthermore, in the process of developing the airbag controller, the parameters in the collision control algorithm need to be matched with the vehicle model. That is, in order to meet the RTTF of a specific vehicle model, the collision control algorithm used by the controller and the parameters therein, usually the threshold line, need to be adjusted. The features used by the currently commonly used collision control algorithms are relatively simple, and it is often difficult to distinguish different collision conditions.

[0065] For example, Figure 2 The 16FRB and 50FRB shown in the figure represent the conditions where the front of the vehicle collides with the rigid barrier at 16 km / h and 50 km / h respectively with a 100% overlap rate. Among them, 16FRB is generally considered to be a low-speed collision and does not need to trigger any restraint system. Figure 3 As shown in the figure, 64SORB (Small Overlap Rigid Barrier) means that the vehicle collides head-on with a rigid barrier at 64 km / h with an offset. Figure 4As shown, 35URT (Under Riding Truck, drilling into a moving truck) means the condition that the vehicle drills into a truck at 35km / h, among which 35URT is considered to be a medium-to-high-speed collision, and it is necessary to issue an ignition command to the restraint system within an appropriate time, for example, by calculating the integral value of the acceleration in the x direction of the sensor in the ECU, and combining whether the acceleration of the front collision sensor and the built-in sensor at the corresponding moment exceeds the threshold line to determine whether each ignition circuit needs to be triggered. However, within the time required for 35URT to detonate, the signal of 16FRB is often larger than that of 35URT. Relevant technicians need to select appropriate features from a large number of data features for calibration, thereby making the performance of the airbag controller more susceptible to the experience and level of relevant technicians, and also bringing a huge workload to relevant technicians, resulting in a longer time required for the calibration process; in addition, the use of a large number of auxiliary algorithms increases the overall complexity of the model. For example, the detonation algorithm in the ECU also includes a frontal collision auxiliary algorithm module, which uses the acceleration on the y-axis of the ECU acceleration sensor and the difference between the accelerations of the left and right front collision sensors for auxiliary judgment. A large number of auxiliary algorithms increase the overall complexity of the model, resulting in poor robustness of the detonation algorithm and increased consumption of computing resources.

[0066] Therefore, in the actual calibration process, simple logic often cannot fully meet the RTTF of each working condition. In order to allow the controller to distinguish as many working conditions as possible and meet the RTTF as much as possible, relevant technicians need to select appropriate features from a large number of features for calibration. This not only brings a huge workload to the relevant technicians, resulting in a long calibration process (usually more than 7 weeks), but also makes the airbag controller performance more susceptible to the experience and level of the relevant technicians.

[0067] On the other hand, collision control algorithms based on machine learning can alleviate the above-mentioned problems, but directly using machine learning models for airbag controllers also has problems with robustness and interpretability. Complex machine learning models can achieve good performance on training samples, but they do not necessarily have good generalization properties. When the nature of the collision signal changes slightly, the model's prediction results may vary greatly. In addition, compared with building a detonation algorithm by combining simple logic, the more complex calculation mechanism inside the machine learning model is not conducive to after-sales service explanation.

[0068] Therefore, based on the above-mentioned related issues, the embodiment of the present application establishes a feature screening technology for an airbag control algorithm based on a support vector machine, which integrates the traditional calibration process and the machine learning algorithm. It has a high degree of automation and a short development cycle, and can reduce the impact of the level and experience of relevant technical personnel on the performance of the airbag. In addition, the trained model is not directly used for airbag detonation, but is used for feature screening to obtain features that can effectively distinguish different working conditions, thereby compressing the process of relevant technical personnel screening features from the feature library. At the same time, by combining these features to construct a detonation algorithm with good robustness and interpretability, its specific implementation process will be described in detail in the following embodiments.

[0069] Specifically, Figure 5 A schematic flow chart of a method for controlling an airbag provided in an embodiment of the present application.

[0070] like Figure 5 As shown, the airbag control method includes the following steps:

[0071] In step S501 , the current collision condition of the vehicle is identified.

[0072] Specifically, the airbag control method involved in the embodiment of the present application will generate corresponding control strategies based on different vehicle models and different working conditions. Therefore, in order to facilitate control accuracy, the embodiment of the present application first needs to obtain collision data of multiple collision conditions for calibration (including front collision sensors, acceleration in the x-axis and y-axis directions in the ECU).

[0073] In step S502, current collision data corresponding to the current collision condition within a preset sampling interval is obtained, a corresponding target airbag detonation algorithm is matched according to the current collision data, and the target airbag detonation algorithm is used to determine whether the current collision data meets the preset ignition condition.

[0074] Optionally, before matching the corresponding target airbag detonation algorithm according to the current collision data, it also includes: obtaining the preset ignition conditions corresponding to each preset collision condition of the vehicle; obtaining an airbag detonation training data set according to the preset ignition conditions corresponding to each preset collision condition, inputting the airbag detonation training data set into a preset airbag prediction detonation model to obtain a current support vector, and determining feature data corresponding to each preset collision condition based on the current support vector; and constructing an airbag detonation algorithm corresponding to each preset collision condition according to the feature data corresponding to each preset collision condition.

[0075] Among them, the preset collision condition, the preset sampling interval and the preset ignition condition can be set by those skilled in the art according to the ignition requirements in the actual test process, or can be obtained through computer simulation, and are not specifically limited here.

[0076] Specifically, the preset sampling interval of the embodiment of the present application can be 200ms before the collision to 500ms after the collision, and the sampling frequency of the collision data is reduced to 2000Hz to adapt to the calculation frequency of the airbag ECU processor. In multiple collision conditions that may occur in the vehicle, the corresponding preset ignition conditions are obtained based on the identified current collision conditions of the vehicle, that is, the RTTF set by relevant technical personnel, and the corresponding current collision data, such as the left front collision sensor acceleration, the right front collision sensor acceleration, the x-axis acceleration in the ECU, and the y-axis acceleration in the ECU in the current collision condition of the vehicle. Based on the above current collision data, the data is collected within the preset sampling interval, and According to the preset ignition conditions corresponding to each preset collision condition, an airbag detonation training data set is obtained, and then the airbag detonation training data set is input into the preset airbag prediction detonation model to obtain a current support vector. The current support vector includes two categories, positive and negative. That is, it is necessary to construct an airbag detonation training data set corresponding to each preset collision condition based on the RTTF time point, and determine the feature data corresponding to each preset collision condition in the current support vector, so as to detonate the airbag based on the RTTF time (labeled as 1), and not to detonate the airbag based on the RTTF time (labeled as 0), so as to obtain the airbag detonation algorithm corresponding to each preset collision condition.

[0077] Optionally, a detonation training data set for an airbag is obtained according to a preset ignition condition corresponding to each preset collision condition, including: obtaining a collision data matrix corresponding to each preset collision condition, simulating a mirror collision data matrix corresponding to each preset collision condition according to the collision data matrix corresponding to each preset collision condition; obtaining mirror collision data corresponding to each preset collision condition according to the mirror collision data matrix corresponding to each preset collision condition, and determining a scaling ratio between the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition based on the preset ignition condition corresponding to each preset collision condition; scaling the collision data corresponding to the preset collision condition and the mirror collision data based on the scaling ratio between the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition, outputting a collision data matrix set corresponding to each preset collision condition, traversing the collision data matrix set corresponding to each preset collision condition, and outputting the first collision data matrix set corresponding to each preset collision condition. A filter data and a second filter data; obtaining a target vector of the first filter data corresponding to each preset collision condition at a target time, determining the on / off state of the airbag detonation algorithm corresponding to each preset collision condition according to the target vector, and when the airbag detonation algorithm is on, obtaining the target filter data of the second filter data of the preset collision condition corresponding to the airbag detonation algorithm corresponding to each preset collision condition that is in the on state at the target time, and calculating the collision feature data at the target time according to the target filter data and the collision data corresponding to each preset collision condition, determining the feature vector of the collision feature data, and when generating a collision feature data sample at the target time, obtaining a collision feature data training sample at the target time according to the collision feature data sample and the feature vector of the collision feature data, and traversing the collision feature data training sample and all running moments of the airbag detonation algorithm corresponding to each preset collision condition to obtain an airbag detonation training data set corresponding to each preset collision condition.

[0078] Optionally, before obtaining the collision feature data training sample at the target moment based on the collision feature data sample and the feature vector of the collision feature data, it also includes: if the collision feature data sample is not generated at the target moment, the airbag detonation algorithm corresponding to each preset collision condition is controlled to enter the next moment of the target moment, and the step of obtaining the target vector of the first filtered data corresponding to each preset collision condition at the target moment is continued.

[0079] Specifically, Figure 6 As shown, the embodiment of the present application first performs model training, and the model training process mainly includes a training data generation module and an SVM (Support Vector Machine) model training module, wherein, Figure 7As shown, the training data generation module includes a data mirroring module S1, a margin construction module S2, a low-pass filter module S3, an algorithm triggering and exiting module S4, a feature extraction module S5 and a label generation module S6. The above six modules perform relevant feature extraction and calculation to finally obtain the airbag detonation training data set corresponding to each preset collision condition.

[0080] Furthermore, the data mirroring module S1 in the embodiment of the present application does not include control conditions and parameters; the parameter of the margin construction module S2 is the scaling ratio r, which will change according to the RTTF, that is, the requirements of the preset ignition conditions given by relevant technical personnel. When RTTF requires detonation, the scaling ratio r can be 1, and when RTTF requires no detonation, the scaling ratio r should be greater than 1. For example, the value of the parameter scaling ratio r can be 1, 1.05, 1.1, 1.15, 1.2, 1.25 or 1.3. It should be noted that the above parameter scaling ratio can be adjusted according to the needs of relevant technical personnel. The purpose of scaling the parameters is to improve the robustness of the model. When the parameters are scaled within a certain range, the model can still maintain the properties in RTTF; the low-pass filter module S3 includes two second-order low-pass filters, and the cutoff frequencies of the control parameters are 60Hz and 14Hz respectively; the control parameters of the algorithm trigger and exit module S4 are the trigger threshold, the exit threshold and the duration required for exit. When the acceleration in the x direction of the front collision sensor or ECU is greater than the trigger threshold, the algorithm will be started. When the acceleration in the x direction of the front collision sensor and the ECU is lower than the exit threshold for a period of time greater than the duration, the exit algorithm will not perform feature extraction calculations; the feature extraction module S5 does not include control conditions and parameters; the control parameter of the label generation module S6 is the extension time t + With t - If the detonation interval required by RTTF is [t1, t2], the label generation module S6 will convert [max(t1-t - ,0),t2+t + ] time as the positive class, and output the airbag detonation training data set containing both positive and negative samples (labeled as 0 or 1).

[0081] Specifically, the embodiment of the present application first obtains collision data of multiple preset collision conditions for calibration (including front collision sensor, acceleration in the x-axis and y-axis directions in the ECU), the sampling interval of the collision data can be set to 200ms before the collision to 500ms after the collision, and the sampling frequency of the collision data is reduced to 2000Hz to adapt to the calculation frequency of the airbag ECU processor; at the same time, the preset ignition conditions given by relevant technicians are obtained. Since the collision data and the preset ignition conditions are the inputs required by relevant technicians when performing calibration work, the embodiment of the present application does not require the input of additional data and documents. Figure 6 and Figure 7 As shown, the collision data corresponding to each preset collision condition and the preset ignition condition corresponding to each preset collision condition are input into the training data generation module, and the collision data matrix S corresponding to each preset collision condition is obtained based on the collision data corresponding to each preset collision condition and the preset ignition condition corresponding to each preset collision condition. raw,k , and the collision data matrix S raw,k Input into the data mirror module S1 to obtain the mirror collision data matrix S corresponding to each preset collision condition mir，k , and the mirror collision data matrix S corresponding to each preset collision condition mir，k S mir,k =[a r,k ,a l，k , a x，k , -a y,k ], and output the mirror collision data corresponding to each preset collision condition

[0082] in, Represents the collision data matrix of the kth working condition, the behavior is the time point, and the columns are different signals, a l，k is the left front collision sensor acceleration, a r，k is the right front collision sensor acceleration, a x，k is the acceleration in the x-axis direction in the ECU, a y，k is the acceleration in the y-axis direction in the ECU, -a y，k for a y，k The opposite number of .

[0083] Secondly, the mirror collision data corresponding to each preset collision condition obtained above and the collision data corresponding to each preset collision condition are input into the residual construction module S2. The residual construction module S2 determines the scaling ratio r of the collision data corresponding to each preset collision condition according to the RTTF. For example, when the RTTF requires detonation, the scaling ratio r can be 1. When the RTTF requires no detonation, the scaling ratio r should be greater than 1. For example, the scaling ratio r can be 1, 1.05, 1.1, 1.15, 1.2, 1.25 or 1.3. Then, according to the ratio corresponding to each preset collision condition, Scale the collision data and output the collision data matrix set corresponding to each preset collision condition And the collision data matrix set corresponding to each preset collision condition output by the residual construction module S2 Traverse, and then use the low-pass filter S3 to filter the collision data set S, and output the first filtered data S corresponding to each preset collision condition trig The second filtering data S corresponding to each preset collision conditionfilt , where the first filtered data S trig The cut-off frequency is 14 Hz, and the second filter data S filt The cut-off frequency is 60Hz.

[0084] Again, the present application is an embodiment based on the first filtering data S corresponding to each preset collision condition obtained trig The second filtering data S corresponding to each preset collision condition filt Traverse the time points of the airbag detonation algorithm to obtain the first filtering data S corresponding to each preset collision condition trig At the current time t, that is, the target vector S at the target time t trig,t , where the target vector S trig，t is a row vector of length 4, and the target vector S trig，t The input is sent to the algorithm trigger and exit module S4 to determine the on / off state of the airbag detonation algorithm according to the target vector corresponding to each preset collision condition, wherein the target vector S trig，t Including a first target threshold, a second target threshold, a third target threshold and a fourth target threshold. If any one of the first target threshold, the second target threshold or the third target threshold is greater than or equal to the target start threshold, the detonation algorithm of the airbag is turned on; if the detonation algorithm of the airbag is in the on state at the target time t, and the first target threshold, the second target threshold and the third target threshold are all less than the target exit threshold within the preset continuous interval, the detonation algorithm of the airbag is turned off. At this time, the algorithm triggering and exiting module S4 outputs the on and off state of the detonation algorithm of the airbag at the target time t. If the detonation algorithm of the airbag is in the off state at the target time t, it is determined whether the detonation algorithm of the airbag runs to the termination time point. If the detonation algorithm of the airbag runs to the termination time point, the above-mentioned collision data matrix set and subsequent related steps are continued to be traversed. Otherwise, the detonation algorithm of the airbag is controlled to enter the next moment of the target time t for calculation; if the detonation algorithm of the airbag is in the on state at the target time t, the second filtering data S corresponding to each preset collision condition is obtained. filt The target filtered data S at the target time t filt,t , and the target filtering data S filt，t Input to feature extraction module S5, feature extraction module S5 uses S fiilt，t And the collision feature data at the target time t is calculated based on the vehicle acceleration and other data stored during the operation of the airbag detonation algorithm. At this time, the relevant technicians can decide the features to be calculated based on experience after balancing the computing resources and performance, that is, the collision feature data corresponding to each preset collision condition. The most extreme case is to calculate all the features in the feature library and record the feature vector of the calculated collision feature data as Among them, f 1，t is the first feature calculated at the target time t, f 2，t is the second feature calculated at the target time t, ..., f n，t is the nth feature calculated at the target time t, and the feature extraction module S5 outputs the feature vector x of the collision feature data corresponding to each preset collision condition. t .

[0085] Again, when the detonation algorithm of the airbag is in the on state at the target time t, the label generation module S6 will determine whether it is necessary to generate a collision feature data sample at the target time t based on the RTTF corresponding to each preset collision condition. For example, if the RTTF requires no detonation, the feature vector x of the collision feature data at all times corresponding to each preset collision condition is t It is necessary to synthesize collision feature data training samples, obtain the labels yt of the collision feature data training samples, and set the labels yt of the collision feature data training samples t =0; if RTTF requires the explosion to occur within the time interval [t1, t2] after the collision occurs, then the interval [max(t1-t - ,0),t2+t + ]Collision feature data training sample label y t Recorded as 1, the time is less than max(t1-t - ,0) of the collision feature data sample label y t It is recorded as 0, and the collision feature data at other moments are not synthesized into the collision feature data training samples. Then, the airbag detonation algorithm corresponding to each preset collision condition is controlled to enter the next moment of the target moment, and the step of obtaining the target vector of the first filtered data corresponding to each preset collision condition at the target moment is continued.

[0086] Finally, the feature vector x of the collision feature data generated by the feature extraction module S5 in the embodiment of the present application is t The label y of the collision feature data training sample generated by the label generation module S6 t Combine them to obtain the collision feature data training sample (x t ,y t ), and the collision feature data training sample (x t ,y t ) is added to the preset training set, and the collision feature data training samples and all running moments of the airbag detonation algorithm corresponding to each preset collision condition are traversed, and finally an airbag detonation training data set containing a large number of positive and negative (labeled as 1 or 0) samples is obtained.

[0087] Furthermore, if Figure 6As shown, in the embodiment of the present application, the detonation training data set of the airbag obtained above is input into the SVM model training module, and the SVM model with slack variables in the SVM model training module is used to construct a binary classification algorithm. The optimization problem can be expressed as:

[0088]

[0089] st0≤a n ≤C

[0090]

[0091] Among them, N is the total number of collision feature data samples, y n is the label of the nth collision feature data sample, C is a parameter given by relevant technical personnel, where the larger C is, the fewer samples that are allowed to be misclassified, and a n with a m is the parameter to be optimized, X m is the feature of the mth sample, x n is the feature of the nth sample, k(·,·) is the kernel function, and the Gaussian kernel function can be used here, with the following expression:

[0092]

[0093] γ is the scale parameter formulated by relevant technical personnel, and x and z are feature vectors. The above quadratic programming problem is solved using the sequential minimum optimization algorithm. If the optimal solution is a n > 0, then the corresponding collision feature data sample (X n ,y n ) is the support vector, and the SVM model training module outputs all support vectors Among them, N S is the number of support vectors, N f is the number of collision feature data.

[0094] In step S503, if the current collision data meets the preset ignition condition, the vehicle is controlled to ignite the airbag.

[0095] Optionally, current feature data corresponding to each preset collision condition is determined based on the current support vector, including: based on each preset collision condition, obtaining the detonation condition and the non-detonation condition of the vehicle, and obtaining a first label sample in the support vector and the first feature data belonging to the detonation condition, and obtaining a second label sample in the support vector and the second feature data belonging to the non-detonation condition; testing the first feature data and the second feature data to obtain a difference value between the first feature data and the second feature data, and when the difference value is less than a preset threshold, determining the current feature data corresponding to each preset collision condition according to the difference value.

[0096] The preset airbag prediction detonation model can be selected by those skilled in the art according to actual collision detection requirements and is not specifically limited here.

[0097] Specifically, after obtaining the airbag detonation training data set corresponding to each of the above-mentioned preset collision conditions, the embodiment of the present application needs to input the airbag detonation training data set corresponding to each preset collision condition into a preset airbag prediction detonation model, such as an SVM model training module, to obtain a current support vector corresponding to each preset collision condition, and perform feature screening on the feature data based on the current support vector to determine the feature data corresponding to each preset collision condition, so as to construct an airbag detonation algorithm corresponding to each preset collision condition according to the feature data corresponding to each preset collision condition.

[0098] Specifically, if Figure 8 As shown, in the feature screening process, it mainly includes a support vector screening module and a feature filtering module, wherein the support vector screening module does not contain control conditions and parameters; the parameter of the feature filtering module is the threshold of the p value, so as to judge whether the feature data corresponding to each preset collision condition has a difference according to the p value. When the calculated p value is less than the preset threshold (which can be a threshold set by calibration or relevant technical personnel), it is considered that the corresponding feature has a significant difference in the two types of support vectors, and the feature data corresponding to each preset collision condition is output.

[0099] Specifically, the input of the support vector screening module of the embodiment of the present application includes the support vector SV output by the SVM model training module, and the conflicting detonation conditions and non-detonation conditions given by relevant technical personnel, wherein the first feature data with a label of 1 in the support vector SV and belonging to the detonation condition is recorded as SV p , that is, the feature data (positive support vector) that needs to detonate the airbag, and the second feature data with a label of 0 and belonging to the non-detonation condition in SV is recorded as SV n , that is, the feature data (negative support vector) that does not need to detonate the airbag, and the support vector screening module outputs SV p With SV n At this time, the feature filtering module filters SV p and SV n Perform a statistical t test on the current feature data corresponding to each preset collision condition, and calculate the p value of the current feature data corresponding to each preset collision condition. The p value of the nth feature is recorded as p n , when p nIf the value is less than the preset threshold, it means that the positive support vector and the negative support vector have significant differences in the feature, and the current feature data corresponding to each preset collision condition is output, so as to construct an airbag detonation algorithm according to the current feature data corresponding to each preset collision condition, so as to control the vehicle to detonate the airbag when the current collision data corresponding to each preset collision condition meets the preset ignition condition corresponding to each preset collision condition.

[0100] In summary, based on the detailed discussion of the above embodiments of the present application, the following beneficial effects are achieved:

[0101] (1) The feature screening technology of the airbag control algorithm based on support vector machine proposed in this application is different from the traditional rule-based calibration algorithm. This technology can more efficiently screen out features that can effectively distinguish different working conditions from the training samples generated based on RTTF through machine learning algorithm;

[0102] (2) This application is different from the detonation algorithm based entirely on machine learning. After obtaining the prediction model, the prediction model is not directly used for airbag detonation. Instead, the prediction model is used to perform feature screening to obtain features that can effectively distinguish different working conditions, and the detonation algorithm is constructed by combining these features. At the same time, it has good robustness and interpretability.

[0103] According to the airbag control method of the embodiment of the present application, the current collision condition of the vehicle is identified, the current collision data corresponding to the current collision condition within the preset sampling interval is obtained, the corresponding target airbag detonation algorithm is matched according to the current collision data, and when the current collision data meets the preset ignition condition, the vehicle is controlled to detonate the airbag. Thus, the problem that the collision control algorithm used in the related art is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF, and a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the detonation algorithm, etc., is solved. Through model training and feature screening, the airbag detonation algorithm of the support vector machine is constructed, so as to effectively distinguish the characteristics of different conditions and improve the robustness of the airbag detonation algorithm.

[0104] Next, the control device for the airbag proposed in accordance with the embodiment of the present application will be described with reference to the accompanying drawings.

[0105] Fig. 9 It is a block diagram of a control device for an airbag according to an embodiment of the present application.

[0106] like Fig. 9 As shown, the airbag control device 10 includes: an identification module 100 , a judgment module 200 and a control module 300 .

[0107] Wherein, the identification module 100 is used to identify the current collision condition of the vehicle;

[0108] The judgment module 200 is used to obtain the current collision data corresponding to the current collision condition within the preset sampling interval, match the corresponding target airbag detonation algorithm according to the current collision data, and judge whether the current collision data meets the preset ignition condition through the target airbag detonation algorithm;

[0109] The control module 300 is used to control the vehicle to ignite the airbag if the current collision data meets the preset ignition condition.

[0110] Optionally, before matching the corresponding target airbag detonation algorithm according to the current collision data, the judgment module 200 further includes:

[0111] An acquisition unit, used for acquiring a preset ignition condition corresponding to each preset collision condition of the vehicle;

[0112] a determination unit, configured to obtain a detonation training data set of the airbag according to a preset ignition condition corresponding to each preset collision condition, input the detonation training data set of the airbag into a preset prediction detonation model of the airbag, obtain a current support vector, and determine feature data corresponding to each preset collision condition based on the current support vector;

[0113] The construction unit is used to construct an airbag detonation algorithm corresponding to each preset collision condition according to the characteristic data corresponding to each preset collision condition.

[0114] Optionally, the determining unit includes:

[0115] A first acquisition subunit is used to acquire a collision data matrix corresponding to each preset collision working condition, and simulate the collision data matrix corresponding to each preset collision working condition to obtain a mirror collision data matrix corresponding to each preset collision working condition;

[0116] A second acquisition subunit is used to obtain the mirror collision data corresponding to each preset collision condition according to the mirror collision data matrix corresponding to each preset collision condition, and determine the scaling ratio of the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition based on the preset ignition condition corresponding to each preset collision condition;

[0117] a first traversal subunit, for scaling the collision data corresponding to each preset collision condition and the mirror collision data based on a scaling ratio of the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition, outputting a collision data matrix set corresponding to each preset collision condition, and traversing the collision data matrix set corresponding to each preset collision condition, and outputting first filtered data and second filtered data corresponding to each preset collision condition;

[0118] a third acquisition subunit, for acquiring a target vector of the first filter data corresponding to each preset collision condition at a target time, determining the on / off state of the airbag detonation algorithm corresponding to each preset collision condition according to the target vector, and acquiring the target filter data of the second filter data of the preset collision condition corresponding to the airbag detonation algorithm corresponding to each preset collision condition in the on state at the target time when the airbag detonation algorithm is on, and calculating the collision feature data at the target time according to the target filter data and the collision data corresponding to each preset collision condition,

[0119] The second traversal subunit is used to determine the characteristic vector of the collision feature data, and when generating the collision feature data sample at the target moment, obtain the collision feature data training sample at the target moment according to the collision feature data sample and the characteristic vector of the collision feature data, and traverse all running moments of the collision feature data training sample and the airbag detonation algorithm corresponding to each preset collision condition to obtain the airbag detonation training data set corresponding to each preset collision condition.

[0120] Optionally, before obtaining the collision feature data training sample at the target time according to the collision feature data sample and the feature vector of the collision feature data, the second traversal subunit further includes:

[0121] The control subcomponent is used to control the airbag detonation algorithm corresponding to each preset collision condition to enter the next moment of the target moment if the collision feature data sample is not generated at the target moment, and continue to execute the step of obtaining the target vector of the first filtered data corresponding to each preset collision condition at the target moment.

[0122] Optionally, the determining unit includes:

[0123] a fourth acquisition subunit, configured to acquire, based on each preset collision condition, a detonation condition and a non-detonation condition of the vehicle, and acquire, in the support vector, first feature data of a first label sample belonging to the detonation condition, and acquire, in the support vector, second feature data of a second label sample belonging to the non-detonation condition;

[0124] The determination subunit is used to check the first characteristic data and the second characteristic data to obtain a difference value between the first characteristic data and the second characteristic data, and when the difference value is less than a preset threshold, determine the current characteristic data corresponding to each preset collision condition according to the difference value.

[0125] According to the airbag control device of the embodiment of the present application, the current collision condition of the vehicle is identified, the current collision data corresponding to the current collision condition within the preset sampling interval is obtained, the corresponding target airbag detonation algorithm is matched according to the current collision data, and the vehicle is controlled to detonate the airbag when the current collision data meets the preset ignition condition. Thus, the problem that the collision control algorithm used in the related art is difficult to distinguish the characteristics of different collision conditions and meet the provided RTTF, and a large number of auxiliary algorithms increase the complexity of the prediction model, thereby reducing the robustness of the detonation algorithm, etc., is solved. Through model training and feature screening, the airbag detonation algorithm of the support vector machine is constructed, so as to effectively distinguish the characteristics of different conditions and improve the robustness of the airbag detonation algorithm.

[0126] Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0127] A memory 1001 , a processor 1002 , and a computer program stored in the memory 1001 and executable on the processor 1002 .

[0128] When the processor 1002 executes the program, the airbag control method provided in the above embodiment is implemented.

[0129] Furthermore, the electronic device further comprises:

[0130] The communication interface 1003 is used for communication between the memory 1001 and the processor 1002 .

[0131] The memory 1001 is used to store computer programs that can be executed on the processor 1002 .

[0132] The memory 1001 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0133] If the memory 1001, the processor 1002 and the communication interface 1003 are implemented independently, the communication interface 1003, the memory 1001 and the processor 1002 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.10 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0134] Optionally, in a specific implementation, if the memory 1001, the processor 1002 and the communication interface 1003 are integrated on a chip, the memory 1001, the processor 1002 and the communication interface 1003 can communicate with each other through an internal interface.

[0135] The processor 1002 may be 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.

[0136] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned airbag control method is implemented.

[0137] This embodiment also provides a computer program product, including a computer program, which is executed to implement the airbag control method of the above embodiment.

[0138] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0139] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0140] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0141] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0142] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0143] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0144] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0145] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for controlling an airbag, characterized in that: The following steps are involved: Identify the current collision condition of the vehicle; Acquire current collision data corresponding to the current collision condition within a preset sampling interval, match a corresponding target airbag detonation algorithm according to the current collision data, and determine whether the current collision data meets a preset detonation condition through the target airbag detonation algorithm; If the current collision data meets the preset ignition condition, the vehicle is controlled to ignite the airbag.

2. The method according to claim 1, characterized in that Before matching the corresponding target airbag detonation algorithm according to the current collision data, the method further includes: Obtaining a preset ignition condition corresponding to each preset collision condition of the vehicle; Obtaining a detonation training data set of the airbag according to a preset ignition condition corresponding to each preset collision condition, inputting the detonation training data set of the airbag into a preset prediction detonation model of the airbag to obtain a current support vector, and determining feature data corresponding to each preset collision condition based on the current support vector; An airbag detonation algorithm corresponding to each preset collision condition is constructed according to the characteristic data corresponding to each preset collision condition.

3. The method according to claim 1 or 2, characterized in that: The step of obtaining the detonation training data set of the airbag according to the preset ignition condition corresponding to each preset collision condition includes: Acquire a collision data matrix corresponding to each of the preset collision working conditions, and simulate and obtain a mirror collision data matrix corresponding to each of the preset collision working conditions according to the collision data matrix corresponding to each of the preset collision working conditions; Obtaining the mirror collision data corresponding to each preset collision condition according to the mirror collision data matrix corresponding to each preset collision condition, and determining the scaling ratio of the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition based on the preset ignition condition corresponding to each preset collision condition; Scaling the collision data corresponding to each preset collision condition and the mirror collision data based on the scaling ratio of the collision data corresponding to each preset collision condition and the mirror collision data corresponding to each preset collision condition, outputting a collision data matrix set corresponding to each preset collision condition, and traversing the collision data matrix set corresponding to each preset collision condition, and outputting first filtered data and second filtered data corresponding to each preset collision condition; Obtaining a target vector of the first filter data corresponding to each preset collision condition at a target time, determining the on / off state of the airbag detonation algorithm corresponding to each preset collision condition according to the target vector, and when the airbag detonation algorithm is on, obtaining the target filter data of the second filter data of the preset collision condition corresponding to the on state of the airbag detonation algorithm corresponding to each preset collision condition at the target time, and calculating the collision feature data at the target time according to the target filter data and the collision data corresponding to each preset collision condition, Determine the characteristic vector of the collision characteristic data, and when generating the collision characteristic data sample at the target moment, obtain the collision characteristic data training sample at the target moment according to the collision characteristic data sample and the characteristic vector of the collision characteristic data, and traverse the collision characteristic data training sample and all running moments of the airbag detonation algorithm corresponding to each preset collision condition to obtain the airbag detonation training data set corresponding to each preset collision condition.

4. The method according to claim 3, characterized in that Before obtaining the collision feature data training sample at the target moment according to the collision feature data sample and the feature vector of the collision feature data, the method further includes: If the collision feature data sample is not generated at the target moment, the airbag detonation algorithm corresponding to each preset collision condition is controlled to enter the next moment of the target moment, and the step of obtaining the target vector of the first filtered data corresponding to each preset collision condition at the target moment is continued.

5. The method according to claim 2, characterized in that: The determining, based on the current support vector, current feature data corresponding to each preset collision condition includes: Based on each of the preset collision conditions, a detonation condition and a non-detonation condition of the vehicle are obtained, and a first label sample and first feature data belonging to the detonation condition are obtained in the support vector, and a second label sample and second feature data belonging to the non-detonation condition are obtained in the support vector; The first characteristic data and the second characteristic data are checked to obtain a difference value between the first characteristic data and the second characteristic data, and when the difference value is less than a preset threshold, current characteristic data corresponding to each preset collision condition is determined according to the difference value.

6. A control device for an airbag, characterized in that: include: An identification module, used for identifying a current collision condition of the vehicle; A judgment module, used to obtain current collision data corresponding to the current collision condition within a preset sampling interval, match the corresponding target airbag detonation algorithm according to the current collision data, and judge whether the current collision data meets the preset ignition condition through the target airbag detonation algorithm; The control module is used for controlling the vehicle to ignite the airbag if the current collision data meets the preset ignition condition.

7. The device according to claim 6, characterized in that Before matching the corresponding target airbag detonation algorithm according to the current collision data, the judgment module further includes: An acquisition unit, used for acquiring a preset ignition condition corresponding to each preset collision condition of the vehicle; a determination unit, configured to obtain a detonation training data set of the airbag according to a preset ignition condition corresponding to each preset collision condition, input the detonation training data set of the airbag into a preset prediction detonation model of the airbag to obtain a current support vector, and determine feature data corresponding to each preset collision condition based on the current support vector; A construction unit is used to construct an airbag detonation algorithm corresponding to each preset collision condition according to the characteristic data corresponding to each preset collision condition.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the airbag control method according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the airbag control method as described in any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that The computer program is executed to implement the airbag control method according to any one of claims 1 to 5.

Images