Vehicle control method and device, computer equipment and readable storage medium

Abstract

The invention relates to a vehicle control method and device, computer equipment and a readable storage medium. The method comprises the steps of obtaining protection effect description information of an air bag in a collision event in response to the collision event triggering the expansion of the vehicle air bag; under the condition that the protection effect description information represents that the airbag protection fails, optimizing the airbag control parameters of the collision event until the optimized airbag control parameters pass safety verification of a digital twin verification platform; the digital twin verification platform is used for simulating a new collision event so as to verify the protection effect corresponding to the optimized airbag control parameters; deploying the optimized airbag control parameters to a vehicle control unit; the vehicle control unit is used for controlling the air bag of the vehicle to execute the next unfolding action. By adopting the method, the adaptability of the airbag system can be improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle control technology, and in particular to a vehicle control method, apparatus, computer equipment, readable storage medium, and program product. Background Art

[0002] With the rapid development of automotive intelligence and safety technology, the performance and intelligence level of the airbag system, as an important component of vehicle passive safety, have received widespread attention.

[0003] Traditional airbag systems usually adopt a fixed control strategy, which triggers airbag deployment through preset trigger conditions and fixed control parameters when a collision occurs. However, this method often cannot adaptively adjust the airbag system's control strategy based on the occupant protection effect, and cannot cope with various complex collision scenarios, resulting in the problem of low adaptability of traditional airbag systems.

[0004] Therefore, the airbag system of traditional technology has the problem of low adaptability. Summary of the Invention

[0005] Based on this, it is necessary to provide a vehicle control method, device, computer equipment, readable storage medium and program product that can improve the adaptability of the airbag system in response to the above technical problems.

[0006] In a first aspect, the present application provides a vehicle control method, the method comprising:

[0007] In response to a collision event that triggers deployment of a vehicle airbag, obtaining descriptive information about the protective effect of the airbag in the collision event;

[0008] When the protection effect description information indicates that the airbag protection fails, optimizing the airbag control parameters for the collision event until the optimized airbag control parameters pass the safety verification of the digital twin verification platform; the digital twin verification platform is used to simulate a new collision event to verify the protection effect corresponding to the optimized airbag control parameters;

[0009] The optimized airbag control parameters are deployed to a vehicle control unit; the vehicle control unit is used to control the airbag of the vehicle to perform the next deployment action.

[0010] In one embodiment, the optimizing the airbag control parameters for the collision event includes:

[0011] Acquiring collision scene information of the collision event; the collision scene information is used to describe at least one of a collision type and an occupant status associated with the collision event;

[0012] According to the airbag parameter optimization model that matches the collision scene information, the airbag control parameters of the collision event are optimized to obtain the optimized airbag control parameters; the airbag parameter optimization model is configured with a reference weight corresponding to the collision scene information; the reference weight represents the reference degree of the protection effect description information in the process of optimizing the airbag control parameters of the collision event.

[0013] In one embodiment, the optimizing the airbag control parameters of the collision event according to the airbag parameter optimization model matching the collision scene information to obtain the optimized airbag control parameters includes:

[0014] Inputting the airbag control parameters of the collision event as state information of the current collision event into the state equation of the airbag parameter optimization model, and outputting state estimation information; the state estimation information is used to represent the airbag control parameters of the next collision event; the next collision event represents the collision event next to the current collision event;

[0015] Inputting the state estimation information into the observation equation in the airbag parameter optimization model and outputting state observation information; the state observation information is used to characterize the protection effect description information corresponding to the airbag control parameters of the next collision event;

[0016] The state estimation information is adjusted according to the state observation information and the reference weight configured by the airbag parameter optimization model to obtain adjusted state estimation information; the adjusted state estimation information is used to characterize the optimized airbag control parameters.

[0017] In one embodiment, the collision event occurs multiple times, and the method further includes:

[0018] Establishing a correlation between the airbag control parameters of the collision event and the protection effect description information of the collision event to obtain airbag deployment observation data of the collision event;

[0019] The airbag deployment observation data of each collision event is added to an observation sequence; the observation sequence is used to update the reference weight of the airbag parameter optimization model configuration.

[0020] In one embodiment, obtaining the descriptive information of the protective effect of the airbag in the collision event includes:

[0021] Acquiring airbag deployment feedback data of the airbag; the airbag deployment feedback data is feedback data of at least two dimensions collected by the vehicle after the airbag performs a deployment action according to the airbag control parameters of the collision event;

[0022] Describing information on the protective effect of the airbag is determined based on the airbag deployment feedback data.

[0023] In one embodiment, the airbag deployment feedback data includes at least a head injury index value, a chest compression index value, and a tibial torsion index value, and determining the airbag protection effect information of the airbag based on the airbag deployment feedback data includes:

[0024] Performing a weighted summation on the head injury index value, the chest compression index value, and the tibial torsion index value to obtain a quantitative value of the protective effect of the airbag;

[0025] The protection effect quantification value of the airbag is obtained as the protection effect description information.

[0026] In a second aspect, the present application further provides a vehicle control device, the device comprising:

[0027] a response module, configured to, in response to a collision event that triggers the deployment of a vehicle airbag, obtain descriptive information on the protective effect of the airbag in the collision event;

[0028] an optimization module, configured to optimize airbag control parameters for the collision event, if the protection effect description information indicates that the airbag protection has failed, until the optimized airbag control parameters pass safety verification on a digital twin verification platform; the digital twin verification platform is configured to simulate a new collision event to verify the protection effect corresponding to the optimized airbag control parameters;

[0029] A deployment module is used to deploy the optimized airbag control parameters to a vehicle control unit; the vehicle control unit is used to control the airbag of the vehicle to perform the next deployment action.

[0030] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0032] In a fifth aspect, the present application also provides a computer program product, comprising a computer program, which implements the steps of the above method when executed by a processor.

[0033] The above-mentioned vehicle control method, device, computer equipment, vehicle control unit, readable storage medium and program product obtain descriptive information of the airbag's protective effect in the collision event in response to a collision event that triggers the deployment of the vehicle's airbag. When the descriptive information of the protective effect indicates that the airbag protection has failed, the airbag control parameters of the collision event are optimized until the optimized airbag control parameters pass the safety verification of the digital twin verification platform; wherein the digital twin verification platform is used to simulate new collision events to verify the protective effect corresponding to the optimized airbag control parameters; the optimized airbag control parameters are deployed to the vehicle control unit; and the vehicle control unit is used to control the vehicle's airbag to perform the next deployment action. In this way, by obtaining the descriptive information of the airbag's protective effect in the collision event, it is possible to accurately determine whether the airbag effectively plays a protective role. When the airbag protection fails, the airbag control parameters are optimized, and the digital twin verification platform is used to simulate new collision events and perform safety verification on the optimized airbag control parameters, effectively ensuring that the optimized parameters can effectively improve the protective effect of the airbag, so that the airbag system can better adapt to different situations in subsequent collisions, and effectively improving the adaptability of the airbag system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 A diagram of an application environment of a vehicle control method according to an embodiment;

[0036] Figure 2 is a flow chart of a vehicle control method according to an embodiment;

[0037] Figure 3 is a flow chart for optimizing an airbag control parameter in one embodiment;

[0038] Figure 4 is another flow chart for optimizing airbag control parameters in one embodiment;

[0039] Figure 5 is a flow chart of another vehicle control method according to an embodiment;

[0040] Figure 6 is a flow chart of a vehicle control method according to another embodiment;

[0041] Figure 7 is a flow chart of another vehicle control method according to another embodiment;

[0042] Figure 8 is a structural block diagram of a vehicle control device in one embodiment;

[0043] Figure 9 is a structural block diagram of a vehicle control device in one embodiment;

[0044] Figure 10 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0046] The collection and processing of the relevant data in this application should be strictly in accordance with the requirements of relevant national laws and regulations when applied in practice, and the informed consent or separate consent of the personal information subject should be obtained. Subsequent data use and processing should be carried out within the scope of authorization of laws and regulations and the personal information subject.

[0047] The vehicle control method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the smart vehicle 102 communicates with the computer device 104 through the network. The data storage system can store data that the computer device 104 needs to process. The data storage system can be integrated on the computer device 104, or it can be placed on the cloud or other network servers. In actual applications, the smart vehicle 102 can respond to a collision event that triggers the deployment of the vehicle airbag and collect the description information of the protection effect of the airbag in the collision event; then, the smart vehicle 102 sends the description information of the protection effect to the computer device 104, and the computer device 104 obtains the description information of the protection effect of the airbag in the collision event; when the description information of the protection effect indicates that the airbag protection fails, the computer device 104 optimizes the airbag control parameters of the collision event until the optimized airbag control parameters pass the safety verification of the digital twin verification platform; the digital twin verification platform is used to simulate new collision events to verify the protection effect corresponding to the optimized airbag control parameters; the computer device 104 deploys the optimized airbag control parameters to the vehicle control unit; the vehicle control unit is used to control the vehicle's airbag to perform the next deployment action.

[0048] In an exemplary embodiment, Figure 2 As shown, a vehicle control method is provided, which is applied to Figure 1 The computer device in the embodiment is used as an example to illustrate the method, including the following steps S202 to S206.

[0049] Step S202 : in response to a collision event that triggers the deployment of the vehicle airbag, obtaining description information of the protective effect of the airbag in the collision event.

[0050] In a specific implementation, after the airbag performs the deployment action according to the airbag control parameters of the collision event, the computer device can obtain the airbag deployment feedback data of the airbag; specifically, after the airbag performs the deployment action according to the airbag control parameters of the collision event, the vehicle can collect airbag deployment feedback data of multiple dimensions through various sensors of the vehicle; and send the airbag deployment feedback data to the computer device.

[0051] Then, the computer device can fuse the airbag deployment feedback data of multiple dimensions to comprehensively evaluate the protective effect of the airbag and generate a description of the airbag's protective effect. Specifically, the computer device can input the airbag deployment feedback data of multiple dimensions into a protection effect scoring model, and the protection effect scoring model can be used to fuse the airbag deployment feedback data of multiple dimensions (for example, weighted summing the airbag deployment feedback data of multiple dimensions) to obtain a protection effect score as a description of the airbag's protective effect in a collision event. In practical applications, the protection effect score can be expressed as .

[0052] Step S204: When the protection effect description information indicates that the airbag protection fails, the airbag control parameters of the collision event are optimized until the optimized airbag control parameters pass the safety verification of the digital twin verification platform.

[0053] In a specific implementation, after obtaining the protection effect description information, the computer device can determine whether the airbag fails to protect in the collision event based on the protection effect description information; the protection effect description information is used as the protection effect score. For example, computer equipment can determine the protection effect score Is it greater than the preset scoring threshold? The scoring threshold can be set to 0.8 (out of 1.0). When the score is greater than the preset threshold, the strategy optimization process is triggered, that is, the airbag control parameters of the collision event are optimized until the optimized airbag control parameters pass the safety verification of the digital twin verification platform.

[0054] Specifically, such as Figure 3 As shown, a computer device can optimize airbag control parameters for a collision event using an airbag control strategy update network to obtain optimized airbag control parameters. The optimized airbag control parameters can then be safety verified using a digital twin verification platform. The airbag control strategy update network can include a neural network model or a mathematical model for optimizing the airbag control parameters.

[0055] Among them, the digital twin verification platform is used to simulate new collision events to verify the protection effect corresponding to the optimized airbag control parameters.

[0056] Specifically, the digital twin verification platform uses pre-configured dynamics analysis software to construct a parametric occupant model (including five percentiles of the human body) and simulate different collision scenarios (30-120 km / h). The digital twin verification platform then loads the optimized airbag control parameters into the prototype vehicle control unit (ACU), injects collision signals via the NI PXI platform, and verifies the response time (RT ≤ 2ms). The digital twin verification platform also generates a safety verification report, including a stress diagram of the airbag protection effect, to identify whether the optimized airbag control parameter combinations represent high-risk combinations.

[0057] If the airbag control parameters are marked as a high-risk parameter combination after optimization, the above-mentioned airbag control strategy update network needs to be updated again, and the airbag control parameters of the collision event need to be optimized through the airbag control strategy update network to obtain the optimized airbag control parameters.

[0058] If the optimized airbag control parameters are not marked as a high-risk parameter combination, the optimized airbag control parameters are determined to be the optimized airbag control parameters that have passed the safety verification of the digital twin verification platform.

[0059] Step S206: deploying the optimized airbag control parameters to the vehicle control unit.

[0060] The vehicle control unit is used to control the vehicle's airbag to perform the next deployment action. In practical applications, the vehicle control unit may be an airbag control unit (ACU).

[0061] In the specific implementation, after the optimized airbag control parameters pass the safety verification of the digital twin verification platform, the computer equipment can verify the safety of the digital twin verification platform, so that the vehicle control unit ACU can control the vehicle's airbag to perform the next deployment action according to the optimized airbag control parameters.

[0062] In a specific implementation, the ACU can respond to the next collision event that triggers airbag deployment by controlling the vehicle's airbags to perform the corresponding deployment action according to the optimized airbag control parameters. Specifically, the ACU can control the airbag's deployment angle, speed, force, and delay time according to the optimized airbag control parameters.

[0063] In the above-mentioned vehicle control method, by responding to a collision event that triggers the deployment of the vehicle airbag, a description of the protective effect of the airbag in the collision event is obtained. When the description of the protective effect indicates that the airbag protection has failed, the airbag control parameters of the collision event are optimized until the optimized airbag control parameters pass the safety verification of the digital twin verification platform; wherein the digital twin verification platform is used to simulate a new collision event to verify the protective effect corresponding to the optimized airbag control parameters; the optimized airbag control parameters are deployed to the vehicle control unit; the vehicle control unit is used to control the vehicle's airbag to perform the next deployment action; in this way, by obtaining the description of the protective effect of the airbag in the collision event, it is possible to accurately determine whether the airbag effectively plays a protective role, and when the airbag protection fails, the airbag control parameters are optimized, and the digital twin verification platform is used to simulate a new collision event to perform safety verification on the optimized airbag control parameters, effectively ensuring that the optimized parameters can effectively improve the protective effect of the airbag, so that the airbag system can better adapt to different situations in subsequent collisions, and effectively improving the adaptability of the airbag system.

[0064] In an exemplary embodiment, optimizing airbag control parameters for a collision event includes: obtaining collision scene information of the collision event; optimizing the airbag control parameters for the collision event based on an airbag parameter optimization model that matches the collision scene information to obtain optimized airbag control parameters.

[0065] The collision scene information is used to describe at least one of the collision type and occupant status associated with the collision event.

[0066] Among them, the airbag parameter optimization model is configured with reference weights corresponding to the collision scenario information.

[0067] The reference weight represents the reference degree of the protection effect description information in the process of optimizing the airbag control parameters of the collision event.

[0068] In practical applications, different collision scene information corresponds to different reference weights.

[0069] For example, the protection effect description information in certain collision scenarios often cannot truly reflect the protection effect of the airbag; in this case, the reference weight of the protection effect description information in the process of optimizing the airbag control parameters of the collision event can be lowered, that is, the reference degree of the protection effect description information in the process of optimizing the airbag control parameters of the collision event can be reduced.

[0070] In a specific implementation, while optimizing airbag control parameters for a collision event, the computer device may obtain collision scenario information for the collision event, such as at least one of the collision type and occupant status. Based on the collision scenario information, the computer device may then determine an airbag parameter optimization model whose model parameters match the collision scenario information, thereby optimizing the airbag control parameters for the collision event using the airbag parameter optimization model. The model parameters may refer to the aforementioned reference weights, representing the degree to which the protective effect description information is used as a reference in optimizing the airbag control parameters for the collision event.

[0071] In practical applications, the reference weights in the airbag parameter optimization model can often represent the airbag parameter optimization strategy; in different airbag parameter optimization strategies, the reference weights of the protection effect description information in the process of optimizing the airbag control parameters of collision events are different.

[0072] Specifically, the computer device can input the airbag control parameters of the collision event into the airbag parameter optimization model corresponding to the airbag parameter optimization strategy to trigger the airbag parameter optimization model to optimize the airbag control parameters of the collision event, and obtain the airbag control parameters of the next collision event (it should be noted that the next collision event is a hypothetical collision event), that is, the optimized airbag control parameters.

[0073] The technical solution of this embodiment obtains the collision scene information of the collision event and optimizes the airbag parameters according to the collision scene information. The airbag parameter optimization model is configured with a reference weight corresponding to the collision scene information. The airbag parameter optimization model can optimize the airbag control parameters of the collision event according to the reference weight and the protection effect description information to obtain the optimized airbag control parameters. In this way, the airbag control parameters can be adaptively optimized based on the collision scene, thereby effectively improving the adaptability of the airbag system.

[0074] In an exemplary embodiment, the airbag control parameters of the collision event are optimized according to an airbag parameter optimization model that matches the collision scene information to obtain optimized airbag control parameters, including: inputting the airbag control parameters of the collision event as state information of the current collision event into the state equation in the airbag parameter optimization model, and outputting state estimation information; the state estimation information is used to characterize the airbag control parameters of the next collision event; inputting the state estimation information into the observation equation in the airbag parameter optimization model, and outputting state observation information; the state observation information is used to characterize the protection effect description information corresponding to the airbag control parameters of the next collision event; adjusting the state estimation information according to the state observation information and the reference weight configured by the airbag parameter optimization model to obtain adjusted state estimation information; the adjusted state estimation information is used to characterize the optimized airbag control parameters.

[0075] The next collision event is the next collision event after the current collision event. It should be noted that the next collision event is a hypothetical collision event.

[0076] The airbag parameter optimization model may refer to a mathematical model based on a Federated Kalman Filter.

[0077] In the case where the airbag parameter optimization model refers to a mathematical model based on the federated Kalman filter, the reference weight of the protection effect description information configured by the airbag parameter optimization model in the process of optimizing the airbag control parameters of the collision event may refer to the covariance matrix of the airbag parameter optimization model; the covariance matrix may include the covariance matrix , covariance matrix ; Among them, the covariance matrix Used to reflect the uncertainty of the state equation of the airbag parameter optimization model; covariance matrix Used to reflect the uncertainty of the observation equation of the airbag parameter optimization model.

[0078] In practical applications, different collision scenario information (collision type or occupant status) is associated with different covariance matrices. , covariance matrix , as shown in Table 1, Table 1 exemplarily shows the preset covariance matrix combinations of different collision scenarios;

[0079]

[0080] Table 1

[0081] In a specific implementation, when the computer device optimizes the airbag control parameters of a collision event according to the airbag parameter optimization strategy that matches the collision scene information to obtain the optimized airbag control parameters, the computer device can input the airbag control parameters of the collision event as the state information of the current collision event into the state equation in the airbag parameter optimization model during the prediction process of the federal Kalman filter, and output state estimation information to characterize the airbag control parameters of the next collision event through the state estimation information; then, the computer device can input the state estimation information into the observation equation in the airbag parameter optimization model, and output state observation information to characterize the protection effect description information corresponding to the airbag control parameters of the next collision event through the state observation information.

[0082] During the updating process of the federal Kalman filter, the computer device can determine the reference weight of the state estimation information based on the above-mentioned covariance matrix; then, according to the reference weight of the state observation information, determine the state correction information of the state estimation information; for example, if the reference weight of the state observation information is lower, the degree of correction of the state estimation information is lower.

[0083] Then, the state estimation information may be adjusted according to the state correction information of the state estimation information to obtain adjusted state estimation information, so as to represent the optimized airbag control parameters through the adjusted state estimation information.

[0084] In practical applications, physical limits can be set for airbag deployment parameters (e.g., maximum force ≤ 8000N, angle deviation ≤ 30°) to prevent secondary injuries caused by the optimization process.

[0085] The technical solution of this embodiment is to input the airbag control parameters of the collision event as the state information of the current collision event into the state equation in the airbag parameter optimization model, output the state estimation information, input the state estimation information into the observation equation in the airbag parameter optimization model, output the state observation information, and then correct the state estimation information based on the state observation information to obtain the optimized airbag control parameters. The airbag control parameters can be effectively optimized based on multiple factors of predicted data and observed data, and the noise and uncertainty data in the process of optimizing the airbag control parameters can be effectively removed, providing a reliable basis for subsequent control decisions.

[0086] In an exemplary embodiment, there are multiple collision events, and the method also includes: establishing a correlation between the airbag control parameters of the collision event and the protection effect description information of the collision event to obtain the airbag deployment observation data of the collision event; adding the airbag deployment observation data of each collision event to the observation sequence; the observation sequence is used to update the model parameters of the airbag parameter optimization model.

[0087] In a specific implementation, there are multiple collision events, and the computer device can establish the airbag control parameters of the collision event. (in, For the Airbag control parameters of the secondary collision event, is the airbag deployment angle, The airbag deployment force, Airbag deployment delay time) and protective effect description information of the collision event, such as the protection effect score The airbag deployment observation data of each collision event is obtained by calculating the correlation relationship between them; the airbag deployment observation data of each collision event is added to the observation sequence; the observation sequence is used to update the reference weight of the airbag parameter optimization model configuration, such as the specific values ​​in the covariance matrix associated with the airbag parameter optimization model mentioned above, so as to realize the use of the airbag deployment observation data of each collision event to dynamically adjust the error in the reference weight of the airbag parameter optimization model configuration; for example, in a certain collision scenario, the airbag deployment observation data of each collision event shows that no matter how the airbag control parameters are set, the protection effect score is lower than the preset value and does not change; this shows that in this collision scenario, the correlation between the airbag control parameters and the protection effect score is low. At this time, the reference weight corresponding to the collision scenario (i.e., collision scenario information) configured by the airbag parameter optimization model can be lowered to reduce the reference degree of the protection effect description information in the process of optimizing the airbag control parameters of the collision event.

[0088] The technical solution of this embodiment obtains the airbag deployment observation data of the collision event by establishing a correlation between the airbag control parameters of the collision event and the protection effect description information of the collision event; the airbag deployment observation data of each collision event is added to the observation sequence to update the model parameters of the airbag parameter optimization model, so that the airbag parameter optimization model can be continuously adaptively updated according to the actual collision situation, so that the airbag parameter optimization model can continue to maintain good parameter optimization performance.

[0089] In an exemplary embodiment, the method also includes: obtaining historical collision data; using the historical collision data to train an airbag control strategy generation model; the airbag control strategy generation model is constructed based on a neural network based on reinforcement learning, and the reinforcement learning neural network is used to output an initial strategy library; the initial strategy library includes at least one candidate airbag control parameter, and protection effect description information corresponding to the candidate airbag control parameter; based on the initial strategy library, constructing an airbag parameter optimization model; the airbag parameter optimization model is constructed based on a mathematical model based on the federal Kalman filter.

[0090] Among them, historical collision data may include simulation scene collision data and real vehicle test collision data; among them, the ratio between the number of groups of simulation scene collision data and the number of groups of real vehicle test collision data can be 200:1; in actual applications, the simulation scene collision data in the historical collision data can be greater than 1,000 groups, and the real vehicle test collision data in the historical collision data can be greater than 1,000 groups.

[0091] In a specific implementation, a computer device can obtain historical collision data and use the historical collision data to train an airbag control strategy generation model to be trained, thereby obtaining a trained airbag control strategy generation model. The airbag control strategy generation model can utilize a neural network model based on a DQN (Deep Q-Network). The airbag control strategy generation model can map input environmental information (e.g., occupant status information and vehicle collision information) to corresponding airbag control parameters (e.g., airbag deployment angle and airbag deployment force). The airbag control strategy generation model is then trained using protection effect description information (e.g., protection effect score) predicted using the airbag control parameters.

[0092] Then, if Figure 4 As shown, after training the airbag control strategy generation model using historical collision data, the computer device can use the airbag control strategy generation model to input several sets of candidate airbag control parameters and corresponding protective effect descriptions to construct an initial strategy library. Using the initial strategy library, the computer device constructs a federated Kalman filter basic model, namely the aforementioned airbag parameter optimization model. This airbag parameter optimization model can be used to describe the mapping relationship between the airbag control parameters for the current collision event and the airbag control parameters for the next collision event, allowing real-time parameter optimization to be performed using this airbag parameter optimization model.

[0093] In addition, when a new collision scenario is detected (such as child passengers or pet interference), the computer device can activate the lightweight GAN network to generate adversarial samples and use the adversarial sample airbag control strategy to generate a model for fine-tuning.

[0094] The technical solution of this embodiment uses a reinforcement learning-based neural network to learn the complex relationship between airbag control parameters and occupant protection effectiveness in different collision scenarios from massive amounts of historical collision data. By analyzing the correlation between various collision types (such as frontal collision, side collision, and rollover collision), occupant characteristics (age, weight, height, and sitting posture), and airbag control parameters (trigger time, inflation volume, and deflation time) in the historical data, a personalized initial strategy library is generated for different scenarios. In subsequent control parameter optimization, the airbag parameter optimization model is used to rapidly optimize the airbag control parameters for the collision event in real time.

[0095] In an exemplary embodiment, obtaining descriptive information of the protective effect of an airbag in a collision event includes: obtaining airbag deployment feedback data of the airbag; and determining descriptive information of the protective effect of the airbag according to the airbag deployment feedback data.

[0096] The airbag deployment feedback data is feedback data in at least two dimensions collected by the vehicle after the airbag performs a deployment action according to the airbag control parameters of a collision event.

[0097] In a specific implementation, after the airbag performs a deployment action according to the airbag control parameters of the collision event, the computer device can obtain the airbag deployment feedback data of the airbag; wherein, the airbag deployment feedback data may include a biomechanical feedback dimension, a visual feedback dimension and a vehicle feedback dimension.

[0098] Specifically, airbag deployment feedback data in the biomechanical feedback dimension can include data collected by a flexible piezoresistive array (16×16 dot matrix, 1kPa resolution) integrated within the seat to monitor the occupant's chest pressure distribution and torso acceleration in real time after airbag deployment (sampling rate 1kHz).

[0099] Airbag deployment feedback data in the visual feedback dimension can include capturing the occupant's post-collision posture deviation (Δx, Δy, Δz) through a 3D ToF (3D Time of Flight) camera for comparison with a preset safety posture template (error threshold ±5cm);

[0100] The airbag deployment feedback data in the vehicle feedback dimension may include the airbag control parameters (deployment force, delay time) of the vehicle control unit during airbag deployment and the deformation of the vehicle body after the collision (which can be collected by the B-pillar deformation sensor) of the ECU (Electronic Control Unit).

[0101] Then, the computer device can fuse the airbag deployment feedback data of multiple feedback dimensions to comprehensively evaluate the protection effect of the airbag and generate descriptive information of the airbag protection effect.

[0102] The technical solution of this embodiment is that after the airbag executes the deployment action according to the airbag control parameters of the collision event, the airbag deployment feedback data of multiple feedback dimensions are integrated to achieve a comprehensive multi-dimensional evaluation of the occupant protection effect of the airbag, thereby improving the objectivity of the occupant protection effect of the airbag.

[0103] In an exemplary embodiment, the airbag protection effect information of the airbag is determined based on the airbag deployment feedback data, including: weighted summation of the head injury index value, the chest compression index value and the tibial torsion index value to obtain a quantitative value of the airbag protection effect; and obtaining the quantitative value of the airbag protection effect as protection effect description information.

[0104] Among them, the airbag deployment feedback data at least includes a head injury index value, a chest compression index value and a tibial torsion index value.

[0105] Among them, the head injury index value is used to measure the passenger's head injury, which can refer to HIC 36(head injury criterion); in practical applications, the head injury index value may refer to the integral of the acceleration of the passenger's head center of mass within a time window of 36ms.

[0106] Among them, the chest compression index value is used to measure the injury of the passenger's chest, which can be obtained through the maximum deformation of the flexible piezoresistive array integrated inside the seat.

[0107] Among them, the tibial torsion index value is used to measure the injury of the passenger's tibia and can be calculated through the vehicle's pedal force sensor.

[0108] In a specific implementation, when the computer device determines the airbag protection effect information of the airbag based on the airbag deployment feedback data, the computer device can perform weighted summation on the head injury index value, the chest compression index value and the tibial torsion index value to obtain a quantitative value of the protection effect of the airbag; specifically, the computer device can obtain the weights corresponding to the head injury index value, the chest compression index value and the tibial torsion index value; then, according to the weights corresponding to the head injury index value, the chest compression index value and the tibial torsion index value, the head injury index value, the chest compression index value and the tibial torsion index value are weightedly summed to obtain the protection effect score of the airbag. That is, the above-mentioned quantitative value of protection effect is used as the description information of the protection effect of the airbag. In actual application, the protection effect score It can be expressed as:

[0109] ;

[0110] in, It can refer to the head injury index value; It can refer to the chest compression index value; It can refer to the tibial torsion index value; It can refer to the weight corresponding to the head injury index value. In practical applications, Can be set to 0.6; It can refer to the weight corresponding to the chest compression index value. In practical applications, Can be set to 0.3; It can refer to the weight corresponding to the tibial torsion index value. In practical applications, Can be set to 0.1.

[0111] In actual applications, when the protection effect score is greater than a preset score threshold, it is determined that the protection effect description information indicates that the airbag protection has failed.

[0112] The technical solution of this embodiment obtains a quantitative value of the protective effect of the airbag by performing a weighted summation of the head injury index value, the chest compression index value and the tibial torsion index value. The quantitative value of the protective effect of the airbag is obtained as the protective effect description information, which can realize a multi-faceted evaluation of the occupant protection effect of the airbag based on the injury conditions of the three parts of the occupant's head, chest and tibia.

[0113] In another embodiment, Figure 5 As shown, another vehicle control method is provided, which is applied to Figure 1 The following steps are described using the computer device example in FIG.

[0114] Step S502 : In response to a collision event that triggers the deployment of a vehicle airbag, airbag deployment feedback data of the airbag is obtained.

[0115] Step S504: Determine the airbag protection effect description information based on the airbag deployment feedback data.

[0116] Step S506 : When the protection effect description information indicates that the airbag protection fails, collision scene information of the collision event is obtained; the collision scene information is used to describe at least one of the collision type and the occupant status associated with the collision event.

[0117] In step S508, the airbag control parameters of the collision event are optimized according to the airbag parameter optimization model that matches the collision scene information to obtain the optimized airbag control parameters, until the optimized airbag control parameters pass the safety verification of the digital twin verification platform.

[0118] Step S510, deploying the optimized airbag control parameters to the vehicle control unit; the vehicle control unit is used to control the vehicle's airbag to perform the next deployment action.

[0119] It should be noted that the specific limitations of the above steps can refer to the specific limitations of a vehicle control method above.

[0120] In an exemplary embodiment, Figure 6 As shown, a vehicle control method is provided, which is applied to Figure 1 Taking the vehicle control unit as an example, the method includes the following steps S602 to S606.

[0121] Step S602: Acquire vehicle occupant status information and vehicle collision information.

[0122] The occupant status information is obtained based on the vehicle's occupant image information. Specifically, the vehicle control unit (ACU) can collect occupant image information in real time through sensors such as in-vehicle cameras. The ACU then processes the occupant image information using computer vision algorithms and outputs occupant status information. This occupant status information can include occupant posture information, occupant position information, occupant movement trends, and occupant dynamic changes.

[0123] Vehicle collision information is predicted based on the vehicle's state information. The vehicle control unit (ACU) can obtain dynamic information such as the vehicle's acceleration, speed, steering angle, and rollover angle. Combined with data from the collision sensor, it determines the vehicle's collision risk and collision type to obtain vehicle collision information.

[0124] Specifically, the vehicle control unit ACU can collect IMU data (1000Hz sampling rate), the steering angle signal of the electronic stability program (ESP), and the roof lidar point cloud (detecting rollover tilt angle) to obtain the acceleration vector, the tilt angle change rate, and the ESP steering angle mutation amount; the vehicle control unit ACU is pre-deployed with a pre-trained multimodal collision classification model based on random forest; the vehicle control unit ACU can input the acceleration vector, the tilt angle change rate and the ESP steering angle mutation amount into the multimodal collision classification model to trigger the multimodal collision classification model to output the collision type prediction result and obtain vehicle collision information; in actual applications, the collision type may include at least one of a frontal collision, a side collision, and a rollover.

[0125] In addition, when the vehicle control unit ACU detects that the rollover angular velocity is greater than 30° / s and lasts for 200ms, the vehicle control unit ACU directly triggers the deployment of the vehicle's roof airbag.

[0126] Step S604: outputting airbag control parameters for the collision event based on the passenger status information and the vehicle collision information.

[0127] In a specific implementation, the vehicle control unit ACU is pre-deployed with a pre-trained airbag control neural network. The vehicle control unit ACU can input the occupant status information and vehicle collision information into the pre-trained airbag control neural network to trigger the pre-trained airbag control neural network to output the airbag control parameters of the collision event.

[0128] Step S606: Control the airbag of the vehicle to perform corresponding deployment action according to the airbag control parameters.

[0129] In a specific implementation, the vehicle control unit (ACU) can respond to a collision event that triggers airbag deployment by controlling the vehicle's airbags to perform corresponding deployment actions according to airbag control parameters. Specifically, the ACU can control the airbag deployment angle, deployment speed, deployment force, and deployment delay time according to the airbag control parameters.

[0130] The above-mentioned vehicle control method obtains the vehicle's occupant status information and vehicle collision information, and outputs the airbag control parameters of the collision event based on the occupant status information and vehicle collision information, and controls the vehicle's airbag to perform corresponding deployment actions based on the airbag control parameters; in this way, adaptive adjustment of the airbag can be achieved, the risk of occupant injury in a collision can be reduced, the intelligent and precise control of the airbag system can be realized, and the safety performance of the vehicle can be improved.

[0131] In an exemplary embodiment, airbag control parameters for a collision event are output based on occupant status information and vehicle collision information, including: determining an airbag deployment angle and an airbag deployment speed based on occupant posture information and occupant position information; determining an airbag deployment force and an airbag deployment delay time based on collision type information and collision severity information; and outputting airbag control parameters for a collision event based on the airbag deployment angle, airbag deployment speed, airbag deployment force, and airbag deployment delay time.

[0132] The occupant status information includes at least occupant posture information and occupant position information.

[0133] The vehicle collision information includes at least collision type information and collision severity information.

[0134] In a specific implementation, the ACU can determine the airbag deployment angle and speed based on the occupant posture and position information while outputting airbag control parameters for a collision event based on occupant status information and vehicle collision information. Specifically, the ACU can determine the projected distance between the occupant's head and the collision direction, as well as the lateral velocity of the occupant's chest, based on the occupant posture and position information. The ACU can then determine the airbag deployment angle based on the collision type information, the projected distance between the occupant's head and the collision direction, and the lateral velocity of the occupant's chest.

[0135] In addition, while the ACU outputs the airbag control parameters for the collision event based on the occupant status information and vehicle collision information, it also determines the airbag deployment force and airbag deployment delay time based on the collision type information and collision severity information. Specifically, the ACU can determine whether the collision type information is a frontal collision with an acceleration greater than 30g. If the collision type information is a frontal collision with an acceleration greater than 30g, the ACU can query the corresponding force parameters in a preconfigured data table based on the vehicle speed and occupant weight. The ACU then determines the airbag deployment force based on the force parameters and the safety factor that matches the occupant's level of dislocation. If the collision type information is a frontal collision with an acceleration greater than 30g, the ACU can also determine the airbag deployment delay time based on the occupant's distance from the vehicle.

[0136] In actual applications, the above-mentioned pre-configured data table can be constructed by combining the collision test database and the own trolley test data (covering the speed range of 50km / h-120km / h).

[0137] The technical solution of this embodiment determines the airbag deployment angle and airbag deployment speed based on the occupant posture information and the occupant position information, and determines the airbag deployment force and airbag deployment delay time based on the collision type information and the collision severity information, and outputs the airbag control parameters of the collision event based on the airbag deployment angle, airbag deployment speed, airbag deployment force and airbag deployment delay time; in this way, it can be applicable to a variety of collision scenarios and occupant postures, meet the safety requirements under different working conditions, effectively utilize the vehicle's occupant status information and vehicle collision information, and intelligently and adaptively control the airbag deployment actions, such as the airbag deployment angle, airbag deployment speed, airbag deployment force and airbag deployment delay time, thereby effectively improving the adaptability of the airbag system.

[0138] In another embodiment, Figure 7 As shown, a vehicle control method is provided, which is applied to Figure 1 Taking the vehicle control unit in the intelligent vehicle as an example, the method includes the following steps:

[0139] Step S702 , obtaining the vehicle's occupant status information and vehicle collision information; the occupant status information at least includes occupant posture information and occupant position information, and the vehicle collision information at least includes collision type information and collision severity information.

[0140] Step S704 : determining the airbag deployment angle and airbag deployment speed based on the occupant posture information and the occupant position information.

[0141] Step S706 : determining the airbag deployment force and airbag deployment delay time according to the collision type information and the collision severity information.

[0142] Step S708: Outputting airbag control parameters for the collision event based on the airbag deployment angle, airbag deployment speed, airbag deployment force, and airbag deployment delay time.

[0143] It should be noted that the specific limitations of the above steps can refer to the specific limitations of a vehicle control method above.

[0144] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0145] Based on the same inventive concept, embodiments of the present application further provide a vehicle control device for implementing the aforementioned vehicle control method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more vehicle control device embodiments provided below can be found in the above-described limitations of the vehicle control method and will not be further elaborated here.

[0146] In an exemplary embodiment, Figure 8 As shown, a vehicle control device is provided, comprising:

[0147] A response module 810 is configured to, in response to a collision event that triggers deployment of a vehicle airbag, obtain information describing a protective effect of the airbag in the collision event;

[0148] an optimization module 820 configured to optimize the airbag control parameters for the collision event, if the protection effect description information indicates that the airbag protection has failed, until the optimized airbag control parameters pass safety verification on a digital twin verification platform; the digital twin verification platform is configured to simulate a new collision event to verify the protection effect corresponding to the optimized airbag control parameters;

[0149] The deployment module 830 is used to deploy the optimized airbag control parameters to a vehicle control unit; the vehicle control unit is used to control the airbag of the vehicle to perform the next deployment action.

[0150] In one embodiment, the optimization module 820 is specifically used to obtain collision scene information of the collision event; the collision scene information is used to describe at least one of the collision type or occupant status associated with the collision event; the airbag control parameters of the collision event are optimized according to the airbag parameter optimization strategy matching the collision scene information to obtain the optimized airbag control parameters; the airbag parameter optimization strategy is used to determine the degree of correction of the protection effect description information to the optimized airbag control parameters.

[0151] In one embodiment, the airbag parameter optimization strategy includes a covariance matrix of an airbag parameter optimization model, and the optimization module 820 is specifically used to input the airbag control parameters of the collision event as the state information of the current collision event into the state equation in the airbag parameter optimization model, and output state estimation information; the state estimation information is used to characterize the airbag control parameters of the next collision event; the state estimation information is input into the observation equation in the airbag parameter optimization model, and output state observation information; the state observation information is used to characterize the protection effect corresponding to the airbag control parameters of the next collision event; according to the covariance matrix and the state observation information, the state estimation information is corrected to obtain the optimized airbag control parameters.

[0152] In one embodiment, the optimization module 820 is specifically used to determine the Kalman gain of the airbag parameter optimization model based on the covariance matrix; determine the state correction information of the state estimation information based on the Kalman gain and the state observation information; use the state correction information to adjust the state estimation information to obtain adjusted state estimation information; the adjusted state estimation information is used to characterize the optimized airbag control parameters.

[0153] In one embodiment, the collision event occurs multiple times, and the device is further used to establish a correlation between the airbag control parameters of the collision event and the protection effect description information of the collision event to obtain the airbag deployment observation data of the collision event; the airbag deployment observation data of each collision event is added to the observation sequence; the observation sequence is used to update the model parameters of the airbag parameter optimization model.

[0154] In one embodiment, the device is also used to obtain historical collision data; the historical collision data is used to train an airbag control strategy generation model; the airbag control strategy generation model is constructed based on a neural network of reinforcement learning, and the reinforcement learning neural network is used to output an initial strategy library; the initial strategy library includes at least one candidate airbag control parameter, and protection effect description information corresponding to the candidate airbag control parameter; based on the initial strategy library, the airbag parameter optimization model is constructed; the airbag parameter optimization model is constructed based on a mathematical model of federal Kalman filtering.

[0155] In one embodiment, the response module 810 is specifically used to obtain airbag deployment feedback data of the airbag; the airbag deployment feedback data is feedback data of at least two dimensions collected by the vehicle after the airbag performs the deployment action according to the airbag control parameters of the collision event; based on the airbag deployment feedback data, the protection effect description information of the airbag is determined.

[0156] In one embodiment, the airbag deployment feedback data includes at least a head injury index value, a chest compression index value, and a tibial torsion index value. The response module 810 is specifically used to perform weighted summation on the head injury index value, the chest compression index value, and the tibial torsion index value to obtain a quantitative value of the protective effect of the airbag; obtaining the quantitative value of the protective effect of the airbag is the protective effect description information.

[0157] In an exemplary embodiment, Figure 9 As shown, a vehicle control device is provided, which is applied to a vehicle control unit, including:

[0158] An acquisition module 910 is configured to acquire the occupant status information and vehicle collision information of the vehicle; the occupant status information is obtained based on the occupant image information of the vehicle; and the vehicle collision information is obtained based on the vehicle status information prediction of the vehicle;

[0159] an output module 920, configured to output airbag control parameters for the collision event based on the occupant status information and the vehicle collision information;

[0160] The control module 930 is configured to control the airbag of the vehicle to perform a corresponding deployment action according to the airbag control parameters.

[0161] In one embodiment, the occupant status information includes at least occupant posture information and occupant position information, and the vehicle collision information includes at least collision type information and collision severity information. The output module 920 is used to determine the airbag deployment angle and airbag deployment speed based on the occupant posture information and occupant position information; determine the airbag deployment force and airbag deployment delay time based on the collision type information and collision severity information; and output the airbag control parameters of the collision event based on the airbag deployment angle, the airbag deployment speed, the airbag deployment force, and the airbag deployment delay time.

[0162] Each module in the aforementioned vehicle control device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor within a computer device in the form of hardware, or may be stored in a computer device memory in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0163] In an exemplary embodiment, a computer device or a vehicle control unit is provided, the internal structure of which can be as follows: Figure 10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a vehicle control method is implemented.

[0164] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0165] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs the steps of the vehicle control method described above. The steps of the vehicle control method described herein may be steps of the vehicle control method described in each of the above embodiments.

[0166] In one embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the vehicle control method described above. The steps of the vehicle control method described here may be the steps of the vehicle control method described in each of the above embodiments.

[0167] In one embodiment, a computer program product is provided, comprising a computer program. When the computer program is executed by a processor, the processor executes the steps of the vehicle control method described above. The steps of the vehicle control method described here may be the steps of the vehicle control method described in each of the above embodiments.

[0168] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0169] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magneto resistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0170] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0171] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A vehicle control method, characterized in that: The method comprises: In response to a collision event that triggers deployment of a vehicle airbag, obtaining descriptive information about the protective effect of the airbag in the collision event; When the protection effect description information indicates that the airbag protection fails, optimizing the airbag control parameters for the collision event until the optimized airbag control parameters pass safety verification on a digital twin verification platform; the digital twin verification platform is used to simulate new collision events to verify the protection effect corresponding to the optimized airbag control parameters; The optimized airbag control parameters are deployed to a vehicle control unit; the vehicle control unit is used to control the airbag of the vehicle to perform the next deployment action.

2. The method according to claim 1, characterized in that The optimizing of the airbag control parameters for the collision event includes: Acquiring collision scene information of the collision event; the collision scene information is used to describe at least one of a collision type and an occupant status associated with the collision event; According to the airbag parameter optimization model that matches the collision scene information, the airbag control parameters of the collision event are optimized to obtain the optimized airbag control parameters; the airbag parameter optimization model is configured with a reference weight corresponding to the collision scene information; the reference weight represents the reference degree of the protection effect description information in the process of optimizing the airbag control parameters of the collision event.

3. The method according to claim 2, characterized in that Optimizing the airbag control parameters of the collision event according to the airbag parameter optimization model matched with the collision scene information to obtain the optimized airbag control parameters includes: Inputting the airbag control parameters of the collision event as state information of the current collision event into the state equation of the airbag parameter optimization model, and outputting state estimation information; the state estimation information is used to characterize the airbag control parameters of the next collision event; the next collision event is the collision event next to the current collision event; Inputting the state estimation information into the observation equation in the airbag parameter optimization model and outputting state observation information; the state observation information is used to characterize the protection effect description information corresponding to the airbag control parameters of the next collision event; The state estimation information is adjusted according to the state observation information and the reference weight configured by the airbag parameter optimization model to obtain adjusted state estimation information; the adjusted state estimation information is used to characterize the optimized airbag control parameters.

4. The method according to claim 3, characterized in that The collision event occurs multiple times, and the method further includes: Establishing a correlation between the airbag control parameters of the collision event and the protection effect description information of the collision event to obtain airbag deployment observation data of the collision event; The airbag deployment observation data of each collision event is added to an observation sequence; the observation sequence is used to update the reference weight of the airbag parameter optimization model configuration.

5. The method according to claim 1, characterized in that The obtaining of the descriptive information of the protective effect of the airbag in the collision event includes: Acquiring airbag deployment feedback data of the airbag; the airbag deployment feedback data is feedback data of at least two dimensions collected by the vehicle after the airbag performs a deployment action according to the airbag control parameters of the collision event; Describing information on the protective effect of the airbag is determined based on the airbag deployment feedback data.

6. The method according to claim 5, characterized in that The airbag deployment feedback data includes at least a head injury index value, a chest compression index value, and a tibial torsion index value. Determining the airbag protection effect information of the airbag based on the airbag deployment feedback data includes: Performing a weighted summation on the head injury index value, the chest compression index value, and the tibial torsion index value to obtain a quantitative value of the protective effect of the airbag; The protection effect quantification value of the airbag is obtained as the protection effect description information.

7. A vehicle control device, characterized in that: The device comprises: a response module, configured to, in response to a collision event that triggers the deployment of a vehicle airbag, obtain descriptive information on the protective effect of the airbag in the collision event; an optimization module, configured to optimize airbag control parameters for the collision event, if the protection effect description information indicates that the airbag protection has failed, until the optimized airbag control parameters pass safety verification on a digital twin verification platform; the digital twin verification platform is configured to simulate a new collision event to verify the protection effect corresponding to the optimized airbag control parameters; A deployment module is used to deploy the optimized airbag control parameters to a vehicle control unit; the vehicle control unit is used to control the airbag of the vehicle to perform the next deployment action.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images