Vehicle control method and device, vehicle and storage medium

Abstract

The invention relates to a vehicle control method and device, a vehicle and a storage medium, and the method comprises the steps: determining a steering intention state of the vehicle based on an obtained running state parameter of the vehicle; under the condition that the steering intention state is in any one of a preparation state, an activation state and a stable state, determining a target prediction value and a user type parameter of the transverse acceleration; determining the predicted lateral displacement of the user according to the target predicted value of the lateral acceleration, the acceleration threshold, the quality parameter, the standard quality parameter and the statistical value of the response factor; and according to the predicted lateral displacement of the user, the corresponding target supporting pressure is determined, so that a seat air bag of the vehicle is controlled to be inflated to the target supporting pressure. According to the invention, the adaptability and comfort of a user can be improved, and more accurate active lateral support is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle control, in particular to a vehicle control method and device, a vehicle and a storage medium. BACKGROUND

[0002] With the continuous development of automobile technology, vehicle active safety and driving comfort have become the research focus. In the driving process, the lateral force caused by the turning operation of the vehicle may cause the lateral displacement of the passenger, affecting the riding experience and safety.

[0003] Therefore, how to provide timely and effective lateral support when the vehicle turns to improve the riding comfort of the user is a problem to be solved at present. SUMMARY

[0004] The present application provides a vehicle control method and device, a vehicle and a storage medium, which can improve the riding comfort of the user in the reasonable turning process.

[0005] The present application provides a vehicle control method, which comprises: determining the turning intention state of the vehicle based on the obtained running state parameters of the vehicle; wherein the turning intention state comprises no intention state, preparation state, activation state, stable state and recovery state; determining the target prediction value of the lateral acceleration and the user type parameter in the case that the turning intention state is in any one of the preparation state, the activation state and the stable state; wherein the user type parameter comprises the acceleration threshold value corresponding to the user type, the mass parameter corresponding to the user type and the response factor corresponding to the user type; determining the predicted lateral displacement of the user according to the statistical value between the target prediction value of the lateral acceleration, the acceleration threshold value, the mass parameter, the standard mass parameter and the response factor; determining the corresponding target support pressure according to the predicted lateral displacement of the user, so as to control the airbag of the seat of the vehicle to inflate to the target support pressure.

[0006] According to the above technical means, by collecting the running state parameters of the vehicle, early identification of the turning intention of the vehicle is realized, so that active support is started before the passenger has not obviously felt the centrifugal force; after the true turning intention is identified, individualized response prediction is further combined with the type difference of the passenger. Finally, the adaptive support strategy is generated according to the prediction result, and the final support degree is optimized through the driving mode correction coefficient. In this way, compared with the support not in time caused by the lag response in the related art, the present application can identify the turning intention in advance and perform lateral support in advance, so as to improve the adaptability and comfort of the user, and realize more accurate active lateral support.

[0007] In some embodiments, the running state parameters include a steering angle, a steering angle velocity, a lateral acceleration, a yaw rate and a driving speed; based on the acquired running state parameters of the vehicle, a steering intention state of the vehicle is determined, including: The steering angle, the steering angle velocity, the lateral acceleration, the yaw rate and the driving speed are input into a steering intention state machine to determine a steering angle confidence, a steering angle velocity confidence, a lateral acceleration confidence, a yaw rate confidence and a signal consistency confidence; The steering angle confidence, the steering angle velocity confidence, the lateral acceleration confidence, the yaw rate confidence and the signal consistency confidence are subjected to a weighted fusion process to determine a total confidence; The steering intention state of the vehicle is determined according to the total confidence.

[0008] According to the above technical means, by inputting multiple key running state parameters into the state machine, different dimension confidences are calculated respectively, and the total confidence is obtained by weighted fusion, and then the steering intention state of the current vehicle is determined according to the preset condition. This multi-signal fusion method significantly improves the recognition accuracy of the true turning intention, and avoids the misjudgment or omission caused by a single signal source.

[0009] In some embodiments, the steering intention state of the vehicle is determined according to the total confidence, including: In the case that the total confidence is less than a first threshold value, the steering intention of the vehicle is determined as a no-intention state; In the case that the total confidence is greater than the first threshold value and less than a second threshold value, and the duration is greater than a first time threshold value, the steering intention of the vehicle is determined as a preparation state; In the case that the total confidence is greater than the second threshold value and less than a third threshold value, the steering intention of the vehicle is determined as an activation state; In the case that the total confidence is greater than the third threshold value and the duration is greater than a second time threshold value, the steering intention of the vehicle is determined as a stable state; In the case that the total confidence is less than a fourth threshold value and greater than the first threshold value, the steering intention of the vehicle is determined as a recovery state; wherein the fourth threshold value is less than the first threshold value, the first threshold value is less than the second threshold value, the second threshold value is less than the third threshold value, and the first time threshold value is less than the second time threshold value.

[0010] According to the above technical means, by setting multiple dynamic thresholds and duration conditions, different turning intention states can be accurately distinguished. For example, when the total confidence is low, it is judged as no intention state to prevent false triggering; when the total confidence is high and maintained for a certain time, it enters the preparation state to prepare for the subsequent support action; and in the high confidence state, it enters the active or stable state to start the support mechanism. In addition, the introduction of the recovery state helps to smoothly reduce the support strength after the turn is completed, improving the ride experience. This state division method ensures reasonable response in different driving scenarios, improving the practicality of the overall control logic.

[0011] In some embodiments, a target prediction value of lateral acceleration is determined, including: determining a first prediction value of the lateral acceleration at the second time based on statistical values between the steering wheel angle, the steering transmission ratio, the wheelbase, the driving speed and the steering gradient at the first time; the first time is before the second time; determining a second prediction value of the lateral acceleration at the second time based on the lateral acceleration, the first derivative of the lateral acceleration and the second derivative of the lateral acceleration at the first time; determining a third prediction value of the lateral acceleration at the second time by fitting processing one or more historical lateral accelerations; determining the target prediction value of the lateral acceleration by weighted fusion processing of the first prediction value, the second prediction value and the third prediction value.

[0012] According to the above technical means, by using three prediction methods and weighted fusion of their results, the change trend of lateral acceleration can be more comprehensively captured, and the prediction accuracy can be improved. This method not only considers the current motion state, but also combines historical data and physical models, making the prediction more reliable and suitable for different complex road conditions and driving behaviors, thereby improving the adaptability to the real turning process.

[0013] In some embodiments, a corresponding target support pressure is determined according to the predicted lateral displacement of the user, including: in the case that the predicted lateral displacement is less than a first displacement threshold, determining the target support pressure as a first support pressure; in the case that the predicted lateral displacement is greater than the first displacement threshold and less than a second displacement threshold, determining the target support pressure as a second support pressure; in the case that the predicted lateral displacement is greater than the second displacement threshold, determining the target support pressure as a third support pressure; wherein the first support pressure is less than the second support pressure, and the second support pressure is less than the third support pressure.

[0014] According to the above technical means, by dividing the predicted lateral displacement into different intervals and setting corresponding support pressure levels for each interval, the support intensity can be flexibly adjusted according to the displacement of the occupant, ensuring sufficient support effect and avoiding discomfort caused by excessive support. This way embodies the idea of hierarchical control, making the support strategy more targeted and comfortable.

[0015] In some embodiments, after determining the target support pressure corresponding to the predicted lateral displacement, the above method further comprises: obtaining the driving mode of the vehicle in the current driving process; wherein the driving mode includes a comfort mode, a standard mode and a sports mode; determining a correction coefficient corresponding to the driving mode according to the driving mode of the vehicle; determining a fourth support pressure based on the correction coefficient corresponding to the driving mode and the target support pressure.

[0016] According to the above technical means, by introducing the driving mode correction coefficient, the support strategy can be adjusted according to the driver's preference, making it more suitable for actual needs. For example, in the comfort mode, the support intensity is small to improve the ride comfort; in the sports mode, the support intensity is large to enhance the control stability. This design fully considers the needs of different driving scenarios, improving the flexibility and user experience of the system.

[0017] In some embodiments, according to the driving mode of the vehicle, the correction coefficient corresponding to the driving mode is determined, comprising: in the case of the driving mode being the comfort mode, the correction coefficient corresponding to the driving mode is determined as a first correction coefficient; in the case of the driving mode being the standard mode, the correction coefficient corresponding to the driving mode is determined as a second correction coefficient; in the case of the driving mode being the sports mode, the correction coefficient corresponding to the driving mode is determined as a third correction coefficient; wherein the first correction coefficient is less than the second correction coefficient, and the second correction coefficient is less than the third correction coefficient.

[0018] According to the above technical means, by assigning different correction coefficients to different driving modes, individualized adjustment can be made according to the driving style while maintaining the basic support logic, providing the best support effect in different driving environments and meeting diverse needs.

[0019] The embodiment of the application provides a vehicle control device, which comprises: a first determination unit configured to determine a steering intention state of the vehicle based on the obtained operating state parameters of the vehicle and the steering intention state machine; wherein the steering intention state includes a no intention state, a preparation state, an activation state, a stable state and a recovery state; The second determining unit is configured to determine a user type parameter and a target prediction value of the lateral acceleration when the steering intention state is in any one of the preparation state, the activation state and the stabilization state, wherein the user type parameter comprises an acceleration threshold corresponding to a user type, a mass parameter corresponding to the user type and a response factor corresponding to the user type. The third determining unit is configured to determine a predicted lateral displacement of the user according to a statistical value between the target prediction value of the lateral acceleration, the acceleration threshold, the mass parameter, a standard mass parameter and the response factor. The fourth determining unit is configured to determine a corresponding target support pressure according to the predicted lateral displacement of the user, so as to control the air bag of the seat of the vehicle to inflate to the target support pressure.

[0020] The embodiments of the present application provide a vehicle, comprising a processor and a memory, the memory storing a computer program capable of running on the processor, and the processor implements the steps in any one of the above methods when executing the computer program.

[0021] The embodiments of the present application provide a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps in any one of the above methods.

[0022] The embodiments of the present application provide a computer program product, comprising a computer program or instructions, and the computer program or instructions are executed by a processor to implement the steps in any one of the above methods.

[0023] The beneficial effects of the present application are as follows: (1) By collecting the running state parameters of the vehicle, early identification of the turning intention of the vehicle is realized, so that active support is started before the passengers feel the centrifugal force obviously; after the real turning intention is identified, individualized response prediction is further combined with the passenger type difference. Finally, an adaptive support strategy is generated according to the prediction result, and the final support intensity is optimized through the driving mode correction coefficient. In this way, compared with the related art in which the support is not timely due to the lag response, the turning intention can be identified in advance and lateral support can be performed in advance, so that the adaptability and comfort of the user can be improved, and more accurate active lateral support is realized.

[0024] (2) By inputting multiple key running state parameters into the state machine, different dimensions of confidence are calculated respectively, and the total confidence is obtained by weighted fusion, and then the steering intention state of the current vehicle is determined according to the preset condition. This multi-signal fusion method significantly improves the identification accuracy of the real turning intention, and avoids the misjudgment or omission caused by a single signal source.

[0025] (3) By setting multiple dynamic thresholds and duration conditions, different turning intention states can be accurately distinguished. For example, when the total confidence is low, it is judged as no intention state to prevent false triggering; when the total confidence rises and maintains for a certain time, it enters the preparation state to prepare for subsequent support actions; and in high confidence, it enters the active or stable state to start the support mechanism. In addition, the introduction of the recovery state helps to smoothly reduce the support intensity after the turn is completed, improving the ride experience. This state division method ensures reasonable response in different driving scenarios, improving the practicality of the overall control logic.

[0026] (4) By using three prediction methods and weighting the results, the trend of lateral acceleration can be more comprehensively captured, improving prediction accuracy. This method not only considers the current motion state, but also combines historical data and physical models, making the prediction more reliable and suitable for different complex road conditions and driving behaviors, thereby improving the adaptability to real turning processes.

[0027] (5) By dividing the predicted lateral displacement into different intervals and setting corresponding support pressure levels for each interval, the support intensity can be flexibly adjusted according to the displacement of the occupant, ensuring sufficient support effect while avoiding discomfort caused by excessive support. This way embodies the idea of hierarchical control, making the support strategy more targeted and comfortable.

[0028] (6) By introducing a driving mode correction coefficient, the support strategy can be adjusted according to the driver's preferences to better meet actual needs. For example, in comfort mode, the support intensity is smaller to improve ride comfort; in sports mode, the support intensity is larger to enhance control stability. This design fully considers the needs of different driving scenarios, improving the flexibility and user experience of the system.

[0029] (7) By assigning different correction coefficients to different driving modes, individual adjustments can be made according to driving styles while maintaining the basic support logic, providing the best support effect in different driving environments and meeting diverse needs. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 Flowchart of a vehicle control method provided by an embodiment of the present application Figure 1 ; Figure 2 Flowchart of a vehicle control method provided by an embodiment of the present application Figure 2 ; Figure 3 Schematic diagram of the relationship between dynamic threshold and speed provided by an embodiment of the present application Figure 4Flowchart of a vehicle control method provided for an embodiment of the present application Figure 3 ; Figure 5 Overall flowchart of a vehicle control method provided for an embodiment of the present application Figure 6 Overall architecture diagram of a vehicle control method provided for an embodiment of the present application Figure 7a Diagram of the relationship between time and vehicle speed provided for an embodiment of the present application Figure 7b Diagram of the relationship between time and predicted lateral acceleration provided for an embodiment of the present application Figure 7c Diagram of the relationship between time and steering wheel rotation angle provided for an embodiment of the present application Figure 7d Diagram of the relationship between time and steering wheel rotation angular velocity provided for an embodiment of the present application Figure 7e Diagram of the relationship between time and yaw rate provided for an embodiment of the present application Figure 8 Composition structure diagram of a vehicle control device provided for an embodiment of the present application Figure 9 Hardware entity diagram of a vehicle provided for an embodiment of the present application DETAILED DESCRIPTION

[0031] Other advantages and effects of the present application can be easily understood by those skilled in the art from the description of the present application. The present application can also be implemented or applied in other different specific embodiments, and the details in the description can be modified or changed based on different views and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, and are not intended to limit the protection scope of the present application.

[0032] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present application in a schematic manner, and only show the components related to the present application in the diagrams, not the number, shape and size of the components when actually implemented. The actual implementation of each component can be randomly changed in terms of shape, number and proportion, and the layout pattern of the components can also be more complex.

[0033] In the following description, “some embodiments” are described, which describe a subset of all possible embodiments, but it can be understood that “some embodiments” can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0034] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0035] In this embodiment, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, object A and / or object B can represent three situations: object A exists alone, object A and object B exist simultaneously, and object B exists alone.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0037] With the rapid development of the automotive industry and the increasing complexity of road traffic environments, active safety technologies in automobiles have become an important means of improving driving safety and comfort. Turning is one of the most common driving maneuvers in daily driving. When a vehicle turns, the driver and passengers inevitably experience lateral displacement due to centrifugal force. This phenomenon is particularly noticeable during high-speed turns, emergency maneuvers, or navigating winding mountain roads, and may affect driving comfort and safety.

[0038] Currently, automotive lateral support technologies mainly fall into two categories: traditional passive lateral support and sensor-triggered active support. Traditional passive lateral support primarily relies on mechanical structures such as reinforced seat side wings and reinforced seat belts. These devices can only provide fixed restraint forces and cannot be adjusted according to the actual cornering intensity, making it difficult to provide optimal support in different driving scenarios.

[0039] Active support systems based on lateral acceleration are currently the mainstream technology. These systems detect the vehicle's lateral movement using lateral acceleration sensors; when the detected lateral acceleration exceeds a preset threshold, they activate the seat side wing inflation mechanism or other support mechanisms. Another type of active support system can be controlled based on steering angle signals. It determines the cornering intensity by monitoring changes in the steering wheel angle, and activates the support device when the steering angle exceeds a set value.

[0040] However, the methods in these technologies may have certain limitations in practical applications.

[0041] First, there is an inherent time lag defect. Most related technologies employ a passive response mode: "signal detected → threshold exceeded → response activated → support action executed." This mode suffers from an insurmountable time delay. By the time the system detects lateral acceleration or steering angle signals and determines that support is needed, the occupants have already begun to undergo lateral displacement under centrifugal force. Considering factors such as sensor signal processing time, controller decision time, and actuator response time, the entire process typically takes 200-500 milliseconds. During this time, the occupants have already experienced significant lateral displacement, and providing support at this point misses the optimal window, failing to effectively prevent the initial displacement of the occupants.

[0042] Secondly, there are limitations to using a single signal source. Related technologies primarily rely on a single sensor signal as the trigger condition, such as using only a lateral acceleration sensor or only a steering angle sensor. This design is susceptible to signal noise, sensor malfunction, or special road conditions. For example, when driving on a bumpy road, the lateral acceleration sensor may detect a momentary peak in lateral acceleration, causing the system to trigger falsely; while on a slippery road, even with a large steering angle, the actual lateral acceleration may be small due to tire slippage, leading to missed triggers. A single signal source cannot comprehensively reflect the vehicle's true motion state and the driver's operational intentions.

[0043] Secondly, there is a lack of predictive capabilities. Most related technologies can only respond based on sensor signals at the current moment and cannot predict the development trend of the turning process. The system cannot distinguish between brief lane correction actions and continuous turning processes, nor can it predict changes in the intensity of the turn. This leads to a mismatch between support strategies and actual needs; either over-support affects comfort, or insufficient support fails to provide effective protection.

[0044] Finally, there is a lack of personalized adaptability. Most systems in related technologies use fixed trigger thresholds and support parameters, failing to consider individual differences among occupants. Occupants of different body types, ages, and physical conditions exhibit significant differences in their sensitivity to and tolerance to lateral forces. Lighter occupants will experience greater displacement under the same lateral acceleration, requiring earlier support intervention; while heavier occupants may require stronger support. The "one-size-fits-all" approach of related technologies cannot provide the most suitable support experience for different occupants.

[0045] Therefore, how to provide timely and effective lateral support when the vehicle is turning, thereby improving the user's ride comfort, is an urgent problem to be solved.

[0046] Based on this, embodiments of this application provide a vehicle control method, the method comprising: determining the vehicle's steering intention state based on acquired vehicle operating state parameters; wherein the steering intention state includes an inactive state, a ready state, an active state, a stable state, and a recovery state; when the steering intention state is in any of the ready state, active state, and stable state, determining a target predicted value of lateral acceleration and user type parameters; wherein the user type parameters include an acceleration threshold corresponding to the user type, a mass parameter corresponding to the user type, and a response factor corresponding to the user type; determining the user's predicted lateral displacement based on the statistical values ​​between the target predicted value of lateral acceleration, the acceleration threshold, the mass parameter, the standard mass parameter, and the response factor; and determining the corresponding target support pressure based on the user's predicted lateral displacement, so as to control the vehicle's seat airbags to inflate to the target support pressure.

[0047] In this way, by collecting vehicle operating parameters, early identification of the vehicle's turning intention is achieved, allowing active support to begin before occupants noticeably feel centrifugal force. After identifying the actual turning intention, personalized response predictions are further made based on occupant type differences. Finally, an adapted support strategy is generated based on the prediction results, and the final support force is optimized using a driving mode correction coefficient. Thus, compared to the untimely support caused by delayed response in related technologies, this application can identify turning intentions in advance and provide lateral support ahead of time, thereby improving user adaptability and comfort and achieving more precise active lateral support.

[0048] The technical solutions in the embodiments of this application will now be clearly and completely described with reference to the accompanying drawings.

[0049] It should be noted that the vehicle control method provided in the embodiments of this application can be executed by an on-board controller in the vehicle. The on-board controller can be an Electronic Control Unit (ECU) or other computing device with data processing capabilities. In other words, the vehicle control method provided in the embodiments of this application can be executed by an on-board controller or collaboratively by other distributed control systems.

[0050] Figure 1 A flowchart illustrating a vehicle control method provided in this application embodiment. Figure 1 ,like Figure 1 As shown, it may include S101 to S104, wherein: S101, based on the acquired vehicle operating status parameters, determine the vehicle's steering intention state. The steering intention state includes no intention state, ready state, active state, stable state, and recovery state.

[0051] Here, the operating status parameters may include, but are not limited to, signal parameters such as steering angle, vehicle speed, lateral acceleration, and yaw rate.

[0052] The "unintentional" state refers to both the initial and final states, indicating that the vehicle is traveling straight or making only minor directional corrections. In this state, all sensor signals fluctuate within normal noise levels, the steering angle is typically less than 5 degrees, and lateral acceleration is close to zero. The unintentional state serves as a clear benchmark, preventing the misinterpretation of normal straight-line vibrations, directional corrections caused by road irregularities, and unconscious driver actions as turning intentions. This state is also the system's safety state; the system will return to this state if any anomalies or uncertainties are detected.

[0053] The ready state is activated when the driver intends to turn and begins to turn the steering wheel. At this point, the physical characteristics are: the steering angle and steering angular velocity begin to increase, but due to vehicle inertia, the delay in tire-ground interaction, and the response time of the suspension system, the lateral acceleration has not yet changed significantly. The key significance of setting this state is to provide a longer prediction time window, typically 200-300 milliseconds, which is precisely the time required for the airbag to inflate. The ready state is designed with ergonomic factors in mind; it typically takes 200-500 milliseconds from the driver's intention to turn to the vehicle's noticeable lateral movement, and this time window is the golden period for the system to predict and prepare. Simultaneously, the ready state can distinguish between a genuine turning intention and a brief directional correction, avoiding frequent false triggers and providing preparation time for subsequent support actions, allowing the system to adjust pressure and pre-set valves in advance.

[0054] The active state indicates that lateral acceleration begins to increase significantly, and the vehicle has begun to produce actual lateral movement. This is a critical moment when the support system begins to function. The physical characteristics of this state include: a rapid increase in lateral acceleration, typically exceeding 1.5 m / s²; the vehicle body begins to yaw, with a significant increase in yaw rate; and occupants begin to feel centrifugal force, with a tendency to lean outwards. The active state is the core operating state of the entire system. In this state, the airbags begin to inflate, providing lateral support to the occupants. The name "active" accurately reflects the system's operating state, meaning that all support functions are activated and begin to work.

[0055] A steady state indicates that the vehicle has entered a stable circular motion, and all dynamic parameters tend to stabilize. Characteristics of this state include: the steering angle remains relatively constant with a rate of change close to zero; lateral acceleration reaches its peak and remains stable; the vehicle is in a quasi-steady-state turn, such as on a long curve on a highway. The steady state setting considers scenarios involving prolonged turns, where a vehicle may maintain a stable turning state for several seconds or even longer on mountain roads or long highway curves. In this situation, the system needs to continuously provide stable support while monitoring changes in various parameters, ready to adjust the support strategy as needed.

[0056] The recovery state is activated when the driver begins to straighten the steering wheel and the vehicle prepares to exit the turn. This state is characterized by: the steering angle beginning to decrease, returning to zero; lateral acceleration beginning to decrease; but the vehicle still exhibits some lateral movement due to inertia. The recovery state design reflects the system's foresight; although the driver has begun to straighten the steering wheel, the vehicle and occupants are still subjected to lateral forces. Immediately removing the support would cause discomfort to the occupants, therefore the system incorporates a recovery state to gradually reduce the support force, achieving a smooth transition.

[0057] In some embodiments, the electronic control unit can acquire vehicle operating status parameters from vehicle sensors via the CAN bus. Furthermore, the acquired operating status parameters can be preprocessed, and the current turning intention state of the vehicle can be determined based on the preprocessed operating status parameters.

[0058] For example, when the driver begins to turn the steering wheel but the lateral acceleration has not yet changed significantly, the electronic control unit can determine that the vehicle's steering intention is in a ready state; when the lateral acceleration begins to rise and exceeds a certain threshold, the electronic control unit can determine that the vehicle's steering intention is in an active state. This state recognition method can capture the driver's turning intention in advance, providing a basis for the generation of subsequent support strategies.

[0059] It is understandable that preprocessing the operating state parameters may include performing Kalman filtering on the operating state parameters to obtain the vehicle's operating state parameters.

[0060] It should be noted that the process noise covariance matrix Q of a traditional Kalman filter is fixed, but the vehicle's motion state exhibits significant time-varying characteristics. During smooth driving, the signal noise is relatively small, requiring a strong smoothing effect; during sharp turns or emergency avoidance, the signal changes drastically, requiring stronger tracking capabilities. Therefore, this application employs an adaptive mechanism to dynamically adjust the Q value. The core of the adaptive mechanism lies in the dynamic adjustment of the process noise covariance matrix Q.

[0061] In some embodiments, performing Kalman filtering on the operating state parameters to obtain the vehicle's operating state parameters may include the following steps: Based on the observations and observation matrix at the third time point, the prediction error at the third time point is determined; Based on the observation matrix, the prediction covariance matrix at the third time step, and the observation noise covariance matrix, the prediction variance at the third time step is determined. Based on the prediction error and prediction variance at the third time step, the process noise covariance matrix in the Kalman filtering process is adjusted to obtain the adjusted process noise covariance matrix. The initial operating state parameters are processed by Kalman filtering based on the adjusted process noise covariance matrix to obtain the vehicle's operating state parameters.

[0062] For example, the formula for calculating prediction error can be expressed as: (1) in, For the first Prediction error at time, For the first The observed value at time, For the observation matrix, For based on Time information for the first The predicted value of the state at any given time.

[0063] For example, the formula for calculating the prediction variance can be expressed as: (2) in, For the first The prediction variance at time point For the first The prediction covariance matrix at time 1, R is the transpose of the observation matrix, and R is the observation noise covariance matrix.

[0064] For example, when When this occurs, it indicates that the system is changing rapidly or that there is strong interference. In this case, the Q value of the process noise covariance matrix is ​​increased by 20%, i.e. This is to improve the filter's ability to track rapidly changing conditions. When When this indicates that the system is relatively stable, the Q value is reduced by 5%, i.e. This is done to enhance smoothing and reduce the impact of noise. It's understandable that the asymmetric adjustment strategy stems from a more pressing consideration of improving tracking capabilities than enhancing smoothing performance.

[0065] In some embodiments, since the sampling frequencies of the various sensor signals are different, a sliding window-based time synchronization method can be used. The synchronization window is set to 20 milliseconds (of course, it can also be adjusted according to the actual scenario, and this application does not limit it). The selection of this time window is based on the following considerations: firstly, to ensure effective sampling of low-frequency signals (such as vehicle speed), and secondly, to maintain the real-time requirements of the system. For signals that require time alignment, a linear interpolation formula can be used for time alignment.

[0066] For example, the linear interpolation formula can be expressed as: (3) in, The synchronized signal value. and These are the two sampling times before and after the signal. The target synchronization time is given. It's understandable that the choice of linear interpolation represents a trade-off between computational complexity and accuracy.

[0067] The following embodiments of this application will detail several ways to determine the vehicle's steering intention state based on the acquired vehicle operating state parameters.

[0068] In one possible implementation, various sensors can be used to acquire vehicle operating state parameters such as steering angle, vehicle speed, lateral acceleration, and yaw rate. These operating state parameters are then input into a pre-trained neural network model, which can ultimately output the vehicle's steering intention state.

[0069] In another possible implementation, vehicle operating state parameters such as steering angle, vehicle speed, lateral acceleration, and yaw rate can be acquired through multiple sensors. These operating state parameters are then input into a multi-state machine, and the confidence level of each operating state parameter is calculated. Based on the confidence level, the vehicle's steering intention state can be determined.

[0070] S102, when the steering intention state is in any of the ready state, active state, or stable state, determine the target predicted value of the lateral acceleration and the user type parameter.

[0071] The user type parameter refers to a set of personalized parameters set according to the differences in occupant body type. Different types of occupants (such as light, standard, and heavy) have different response characteristics. For example, the user type parameter may include the acceleration threshold corresponding to the user type, the mass parameter corresponding to the user type, and the response factor corresponding to the user type.

[0072] Here, the target predicted value of lateral acceleration refers to the estimated value of lateral acceleration at a future moment, calculated based on current and historical data. This target predicted value can be used to anticipate the lateral movement trend of a vehicle, thus providing input for support strategies.

[0073] It is understandable that users can be categorized into multiple types based on their weight characteristics. For example, this could include light, standard, and heavy body types. For instance, the weight range for light body types could be less than 60 kg, including children and smaller adults; the weight range for standard body types could be 60-80 kg, including average adults; and the weight range for heavy body types could be greater than 80 kg, including larger adults.

[0074] It's also understandable that the parameter differences between different types of occupants manifest in three aspects: acceleration threshold (also known as trigger threshold), mass parameters, and response factor. The acceleration threshold for lightweight occupants is 2.5. The standard body size is 3.0. The weight is 3.5. This difference reflects the varying sensitivities of occupants of different body types to lateral acceleration. Mass parameters affect the calculation of lateral displacement; lighter occupants are more likely to experience lateral displacement under the same lateral acceleration. The response factor reflects the sensitivity of different body types to support; smaller occupants are more sensitive to support.

[0075] It should be noted that the following embodiments of this application will detail several implementation methods for determining the target predicted value of the vehicle's lateral acceleration.

[0076] In one possible implementation (denoted as Method A), the target predicted value of the vehicle's lateral acceleration can be based on the vehicle's kinematic model. First, the front wheel steering angle is calculated based on the steering angle and steering gear ratio; further, the turning radius is calculated based on Ackermann steering geometry; finally, the theoretical lateral acceleration is calculated using the centripetal acceleration formula. Considering the understeer characteristics of actual vehicles, the system introduces a correction factor, the understeer gradient. It can be set to 0.002. This value is maliciously obtained based on test data from a large number of vehicles. The correction formula is solved iteratively, and it usually converges in 3-5 iterations.

[0077] In some embodiments, the front wheel steering angle of a vehicle can be determined based on the proportional relationship between the vehicle's steering wheel angle and its steering gear ratio. For example, the formula for the front wheel steering angle can be expressed as: (4) in, This refers to the vehicle's steering angle; This refers to the vehicle's steering gear ratio; This indicates the steering angle of the vehicle's front wheels.

[0078] In some embodiments, the turning radius of a vehicle can be determined based on the proportional relationship between the vehicle's wheelbase and the sine of the vehicle's front wheel steering angle. For example, the turning radius of a vehicle can be expressed as: (5) Where L is the wheelbase of the vehicle; This indicates the vehicle's turning radius.

[0079] In some embodiments, the theoretical lateral acceleration of a vehicle can be determined based on the proportional relationship between the vehicle's speed and its turning radius. For example, the theoretical lateral acceleration can be expressed as: (6) in, Indicates the vehicle's speed; This represents the vehicle's theoretical lateral acceleration.

[0080] In some embodiments, a correction factor may be introduced to take into account the understeer characteristics of the actual vehicle. Insufficient turning gradient The value is set to 0.002. This value is based on test data from a large number of vehicles. The correction formula is solved iteratively, and it usually converges in 3-5 iterations.

[0081] For example, the correction formula can be expressed as: (7) in, Let n represent the lateral acceleration of the vehicle at time n. This indicates insufficient steering gradient (also known as steering gradient). It can be expressed as the predicted lateral acceleration of the vehicle at time n+1.

[0082] In another possible implementation (here referred to as Method B), the predicted value of the lateral acceleration at future times can be determined based on transient prediction by calculating the lateral acceleration at the first moment, the first derivative of the lateral acceleration, and the second derivative of the lateral acceleration.

[0083] For example, the formula for determining the predicted lateral acceleration based on transient prediction can be expressed as: (8) in, Let represent the lateral acceleration of the vehicle at time t; Let represent the first derivative of the vehicle's lateral acceleration at time t; This represents the second derivative of the vehicle's lateral acceleration at time t; T is the prediction time domain, which can be set to 0.5 seconds or other values.

[0084] In another possible implementation, determining the target predicted value of the lateral acceleration may include the following steps: S1021, based on the statistical values ​​between the steering wheel angle, steering ratio, wheelbase, driving speed and steering gradient at the first moment, determine the first predicted value of the lateral acceleration at the second moment; the first moment is before the second moment.

[0085] It should be noted that the first predicted values ​​of the lateral velocity and speed at the second moment can be calculated using the above method A, which will not be elaborated here for the sake of simplicity.

[0086] S1022, based on the lateral acceleration at the first moment, the first derivative of the lateral acceleration, and the second derivative of the lateral acceleration, determine the second predicted value of the lateral acceleration at the second moment.

[0087] It should be noted that the second predicted values ​​of the lateral velocity at the second moment can be calculated according to method B above, which will not be elaborated here for the sake of simplicity.

[0088] S1023, Fit one or more historical lateral accelerations to determine a third predicted value of the lateral acceleration at the second moment.

[0089] As can be understood, fitting is a mathematical modeling method that analyzes a set of historical lateral acceleration samples to find a curve or function that best represents the changing patterns of these data. In this embodiment, the system selects several historical lateral acceleration samples and uses the least squares method or other fitting algorithms to construct a function model describing the changing trend of lateral acceleration.

[0090] Historical lateral acceleration refers to lateral acceleration data recorded over a past period, reflecting the lateral movement of a vehicle under different operating conditions. Analyzing this historical data reveals patterns and regularities in lateral acceleration changes, providing a basis for future predictions.

[0091] In practical applications, historical lateral acceleration data from recent periods can be fitted to predict the lateral acceleration value for a future period. This method is particularly suitable for scenarios where lateral acceleration changes follow a certain regularity, such as continuous turning or uniform circular motion.

[0092] S1024, perform weighted fusion processing on the first predicted value, the second predicted value and the third predicted value to determine the target predicted value of lateral acceleration.

[0093] Weighted fusion processing is a method that integrates multiple prediction results. By assigning different weights to each prediction result, weighted fusion processing combines these prediction results into a final prediction value. In this embodiment, the results of the prediction method in S1021 (also known as steady-state prediction), the prediction method in S1022 (also known as transient prediction), and the prediction method in S1023 (also known as extrapolation prediction) can be weighted and averaged to obtain a more accurate and stable lateral acceleration prediction value.

[0094] For example, the weights of various prediction methods can be dynamically adjusted based on their reliability in specific scenarios. For instance, when a vehicle enters a stable turning state, the weight of the steady-state prediction method might be increased, while in cases of sharp turns or sudden lane changes, the weight of the transient prediction method might be even higher. The adaptive adjustment mechanism implemented by the system ensures that the prediction results always closely approximate the actual situation. For example, the weight coefficient for steady-state prediction could be 0.4, the weight coefficient for transient prediction could be 0.4, and the weight coefficient for extrapolation prediction could be 0.2.

[0095] By combining and integrating the three prediction methods described above, the development trend of lateral acceleration can be predicted in advance within a time window of 200-500 milliseconds, thus providing sufficient preparation time for the active support of the airbag system. Compared with traditional passive response control methods, the predictive control strategy adopted in this embodiment can significantly improve the system's response speed and control accuracy, and effectively enhance occupant safety and comfort.

[0096] In summary, by employing a combination of multi-dimensional data acquisition and modeling, dynamic prediction methods, and weighted fusion processing, this embodiment of the application can achieve high-precision prediction of lateral acceleration. Based on this prediction result, this embodiment of the application can activate the active lateral support system in advance. Due to the early activation of the active lateral support system, this embodiment of the application can significantly reduce the lateral displacement of occupants during cornering and improve the overall riding experience and safety.

[0097] S103 determines the user's predicted lateral displacement based on the statistical values ​​between the target predicted value of lateral acceleration, acceleration threshold, mass parameters, standard mass parameters, and response factors.

[0098] Here, predicted lateral displacement refers to the amount of lateral body movement that an occupant may experience given a lateral acceleration, which is a fundamental parameter that determines the airbag inflation pressure.

[0099] In one possible implementation, the above-mentioned step S103, "determining the user's predicted lateral displacement based on the statistical values ​​between the target predicted value of lateral acceleration, the acceleration threshold, the mass parameter, the standard mass parameter, and the response factor," may include the following steps: The lateral acceleration difference is determined based on the target predicted value of lateral acceleration and the acceleration threshold; the mass ratio is determined based on the proportional relationship between the mass parameter and the standard mass parameter; and the user's predicted lateral displacement is determined based on the response factor, the lateral acceleration difference, and the mass ratio.

[0100] For example, the formula for predicting lateral displacement can be expressed as: (9) in, It can represent the response factor; It can represent the predicted lateral displacement; It can represent the target predicted value of lateral acceleration; This can be expressed as an acceleration threshold; It can be expressed as a quality parameter; It can be expressed as a standard quality parameter.

[0101] In another possible implementation, the target predicted value of lateral acceleration, acceleration threshold, mass parameter, standard mass parameter and response factor can be input into a pre-trained neural network model, thereby directly outputting the predicted lateral displacement obtained by the model.

[0102] S104 determines the corresponding target support pressure based on the user's predicted lateral displacement, so as to control the vehicle's seat airbags to inflate to the target support pressure.

[0103] Here, the target support pressure refers to the pressure value that the airbag should reach, calculated based on the predicted lateral displacement, used to control the inflation level of the seat airbag, thereby providing appropriate lateral support for the occupant.

[0104] In one possible implementation, a functional relationship can be constructed based on the user's predicted lateral displacement and the target support pressure. After determining the user's predicted lateral displacement, the predicted lateral displacement can be substituted into the functional relationship to obtain the target support pressure corresponding to the predicted lateral displacement.

[0105] It should be noted that the functional relationship can be obtained by researchers through fitting a large amount of experimental data, and it can be a linear relationship or a non-linear relationship.

[0106] In another possible implementation, a mapping table between predicted lateral displacement and target support pressure can be pre-built. That is, the predicted lateral displacement can be divided into multiple different intervals, with different target support pressures corresponding to different lateral displacement intervals.

[0107] For example, the above-mentioned S104 "determining the corresponding target support pressure based on the user's predicted lateral displacement" may further include the following steps: S1041, if the predicted lateral displacement is less than the first displacement threshold, the target support pressure is determined as the first support pressure.

[0108] The first displacement threshold is a pre-set numerical boundary used to determine whether lateral displacement has reached the minimum trigger condition. Ergonomic factors can be considered when setting the first displacement threshold to ensure that the airbag begins to provide adequate support before the user experiences significant lateral movement. For example, the first displacement threshold can be set to 8mm, 10mm, or 15mm.

[0109] The first support pressure refers to the initial support force applied when the predicted lateral displacement is below the first displacement threshold. The first support pressure is relatively low, typically between 20-30 kPa, designed to prevent slight shaking by the user while avoiding discomfort caused by excessive support. The first support pressure is suitable for scenarios such as slight vehicle turns and bumpy road surfaces.

[0110] By predicting that a user will experience a slight lateral displacement and activating low-strength support in advance, unnecessary swaying can be effectively reduced while ensuring the system's responsiveness.

[0111] S1042, if the predicted lateral displacement is greater than the first displacement threshold and less than the second displacement threshold, the target support pressure is determined as the second support pressure.

[0112] The second displacement threshold is a pre-set second judgment point. If the value of the second displacement threshold is higher than the first displacement threshold, it indicates that the user has experienced a relatively significant lateral displacement, but it has not yet reached the level requiring high-strength support. For example, the second displacement threshold may be between 15mm and 40mm.

[0113] The second support pressure refers to the moderate support force applied when the predicted lateral displacement falls between the first and second displacement thresholds. The second support pressure is typically between 30-45 kPa and is suitable for normal turning or minor emergency avoidance scenarios. At this point, the user has begun to feel centrifugal force but has not yet experienced significant body shift; the system needs to strengthen the support to prevent further displacement.

[0114] By introducing a second support pressure, the support strength can be dynamically adjusted according to different degrees of lateral displacement. It avoids excessive intervention during slight displacement and does not react lagly during significant displacement. This method of introducing a second support pressure and dynamically adjusting the support strength according to different degrees of lateral displacement allows for a more precise match to the user's actual needs, thereby enhancing the support effect and improving overall riding safety and comfort.

[0115] S1043, if the predicted lateral displacement is greater than the second displacement threshold, the target support pressure is determined as the third support pressure.

[0116] It's understandable that the second support level is lower than the third support level.

[0117] Here, the third support pressure refers to the high-intensity support applied when the predicted lateral displacement exceeds the second displacement threshold. The third support pressure is typically between 45-60 kPa and is suitable for high-risk driving scenarios such as sharp turns, high-speed driving, or emergency avoidance. When the predicted lateral displacement exceeds the second displacement threshold, the user has already experienced a significant lateral displacement, and the system needs to immediately apply high-intensity support to prevent the user's body from sliding out of the seat or hitting the door.

[0118] Understandably, the vehicle also includes an airbag execution and pressure control module, which translates support strategies into actual airbag movements. The inflation and deflation of the vehicle's seat airbags can be controlled via rapid solenoid valves, achieving precise adjustment of the target support pressure. The vehicle's seat airbags are positioned at key support locations in the seat (such as the backrest and seat cushion sides) to provide effective lateral support. The vehicle also features air tanks and compressors to ensure that the seat airbags can inflate quickly, meeting the needs of predictive control.

[0119] In practice, by using airbag actuators and pressure control modules, precise control can be achieved over the inflation process of the vehicle's seat airbags, thereby providing appropriate lateral support to occupants at the optimal time and improving the overall riding experience and safety.

[0120] In this embodiment, by dividing the predicted lateral displacement into multiple intervals and assigning a corresponding support pressure level to each interval, the support force can be automatically adjusted according to different levels of lateral displacement, thereby achieving a personalized support strategy. Using this method, the response accuracy and personalized adaptability of the support system can be improved, effectively preventing lateral displacement of the user and thus enhancing vehicle safety and ride comfort.

[0121] The vehicle control method provided in this application collects and processes vehicle operating status parameters through a multi-sensor signal fusion processing module, enabling early identification of the vehicle's turning intention. This allows for active support to begin before occupants noticeably experience centrifugal force. After identifying the actual turning intention, personalized response prediction is further performed based on occupant type differences. Finally, an adapted support strategy is generated based on the prediction results, and the final support force is optimized using a driving mode correction coefficient. Thus, compared to the delayed support caused by lag in related technologies, this application can identify turning intentions in advance and provide lateral support earlier, thereby improving user adaptability and comfort and achieving more precise active lateral support.

[0122] In some embodiments, Figure 2 A flowchart illustrating a vehicle control method provided in this application embodiment. Figure 2 ,like Figure 3 As shown, the above-mentioned S101 "determining the vehicle's steering intention state based on the acquired vehicle operating state parameters" may include the following steps: S1011, input the steering angle, steering angular velocity, lateral acceleration, yaw rate and driving speed into the steering intention state machine, and determine the confidence level of steering angle, steering angular velocity, lateral acceleration, yaw rate and signal consistency.

[0123] Here, the steering intention state machine is a multi-state system used to identify the driver's turning intention. Based on the comprehensive analysis of signals from multiple sensors, the steering intention state machine can determine whether the vehicle is currently in different stages such as turning preparation, activation, stabilization, or recovery. The steering intention state machine typically includes five states: no intention, preparation, activation, stabilization, and recovery.

[0124] In some embodiments, the steering angle confidence score can be determined based on the vehicle's current steering angle and a dynamic threshold. For example, the formula for calculating the steering angle confidence score can be expressed as: (10) in, Indicates the confidence level of the steering angle; Indicates the current steering angle; This indicates a dynamic threshold.

[0125] It should be noted that the dynamic threshold can be adaptively adjusted according to the vehicle's speed. For example, the formula for calculating the dynamic threshold can be expressed as: (11) in, Indicates the basic threshold. Indicates the speed influence coefficient. Current vehicle speed To achieve the minimum trigger speed, For reference speed, To adjust the index.

[0126] For example, Figure 3 This is a schematic diagram illustrating the relationship between dynamic threshold and speed provided in an embodiment of this application; as shown below. Figure 4 As shown, the dynamic threshold value is limited to a range of 3. Up to 15 ;exist , In this case, with the current vehicle speed As the speed increases from 0 to 120 km / h, the dynamic threshold changes from 5... Increased to 15 Understandable. ; The speed influence coefficient can control the magnitude of the threshold change with speed; The system will not work if the trigger speed is below the minimum. This corresponds to the typical driving speed on urban expressways; This causes the threshold to change non-linearly.

[0127] In some embodiments, the steering angular velocity confidence score can be determined based on the vehicle's current steering angular velocity and a first preset threshold. For example, the formula for calculating the steering angular velocity confidence score can be expressed as: (12) in, Indicates the confidence level of the steering angular velocity; Indicates the current steering angular velocity; This indicates the first preset threshold.

[0128] In some embodiments, the lateral acceleration confidence score can be determined based on the vehicle's lateral acceleration and a second preset threshold. For example, the formula for calculating the lateral acceleration confidence score can be expressed as: (13) in, Indicates the confidence level of lateral acceleration; Indicates the current lateral acceleration; This indicates the second preset threshold.

[0129] In some embodiments, the yaw rate confidence level can be determined based on the vehicle's current yaw rate and a third preset threshold. For example, the formula for calculating the yaw rate confidence level can be expressed as: (14) in, Indicates the confidence level of the yaw rate; Indicates the current yaw rate; This indicates the third preset threshold.

[0130] In some embodiments, signal consistency confidence can assess the directional consistency of each signal. Under normal circumstances, when turning right, the steering angle and yaw rate are both positive, and the consistency confidence is 1; when turning left, both are negative, and the consistency confidence is also 1. If inconsistency occurs, it may be due to sensor failure or special operating conditions, in which case the consistency confidence can drop to 0.5.

[0131] S1012 performs weighted fusion processing on the confidence scores of steering angle, steering angular velocity, lateral acceleration, yaw rate, and signal consistency to determine the total confidence score.

[0132] For example, the formula for calculating the total confidence level can be expressed as: (15) in, These are the weighting factors for steering angle confidence, steering angular velocity confidence, lateral acceleration confidence, yaw rate confidence, and signal consistency confidence, respectively.

[0133] It is understandable that the weighting reflects the importance and reliability of each signal in the judgment. For example, the steering angle, as a signal directly controlled by the driver, has the highest weight, while lateral acceleration, as a direct manifestation of vehicle response, has the second highest weight. For instance, the weighting factors for steering angle confidence, steering angular velocity confidence, lateral acceleration confidence, yaw rate confidence, and signal consistency confidence can be 0.3, 0.2, 0.25, 0.15, and 0.1, respectively.

[0134] S1013, determine the vehicle's steering intention state based on the total confidence level.

[0135] In some embodiments, after determining the total confidence level, the vehicle's steering intention state can be determined based on the range of the total confidence level and the duration of the vehicle's current state.

[0136] In this embodiment, steering angle, steering angular velocity, lateral acceleration, yaw rate, and driving speed can be input into the steering intention state machine to determine the confidence level of each dimension. The confidence levels of each dimension are then weighted and fused to obtain a total confidence level. The vehicle's steering intention state is determined based on the total confidence level. Using this method, the accuracy and robustness of turning intention recognition can be improved, allowing for earlier detection of actual turning behavior. This enables the early triggering of active support strategies, enhancing occupant safety and comfort.

[0137] In some embodiments, the above-described S1013 "determining the vehicle's steering intention state based on total confidence" may include the following steps: If the total confidence level is less than the first threshold, the vehicle's steering intention is determined to be in an unintentional state. If the total confidence level is greater than the first threshold and less than the second threshold, and the duration is greater than the first time threshold, the vehicle's steering intention is determined to be in a ready state. If the total confidence level is greater than the second threshold and less than the third threshold, the vehicle's steering intention is determined to be active. If the total confidence level is greater than the third threshold and the duration is greater than the second time threshold, the vehicle's steering intention is determined to be in a stable state. If the total confidence level is less than the fourth threshold but greater than the first threshold, the vehicle's steering intention is determined to be in a recovery state.

[0138] Among them, the fourth threshold is less than the first threshold, the first threshold is less than the second threshold, the second threshold is less than the third threshold, and the first time threshold is less than the second time threshold.

[0139] It is understandable that the first, second, third, and fourth thresholds can be used to define the transition boundaries between various states. The selection of the first, second, third, and fourth thresholds can be based on a large amount of real-vehicle test data and physical model analysis to ensure that the vehicle's driving state can be accurately identified under different driving conditions.

[0140] For example, the first threshold can be 0.3, which is a relatively low value, used to capture turning intentions as early as possible; the second threshold can be 0.6, indicating that multiple signals show clear turning characteristics; the third threshold can be 0.8, used to determine whether the vehicle has entered a stable turn; and the fourth threshold can be 0.2, used to identify the recovery state.

[0141] It should be noted that the setting of the first threshold, the second threshold, the third threshold and the fourth threshold fully considers the needs under different driving scenarios and can be flexibly adjusted according to different driving scenarios. This application embodiment does not limit this.

[0142] The first and second time thresholds can be used to determine the duration required for certain state transitions. For example, when transitioning from an unintentional state to a ready state, it is required that the total confidence score is not only greater than the first threshold but less than the second threshold, but also that the total confidence score remains above the first threshold for a certain period of time to ensure the stability of the turning maneuver. The setting of the time thresholds takes into account the vehicle's dynamic characteristics and the physiological responses of the occupants, ensuring that the system can make reasonable judgments and avoids false triggering of support strategies due to brief signal fluctuations.

[0143] As can be understood, the unintentional state serves as the initial state, used to confirm whether the vehicle has truly entered the turning intention stage; the preparation state provides early warning, creating conditions for the system to preload support strategies; the activation state is the critical moment when the support system formally intervenes; the stable state ensures the continuous effectiveness of support force during long-term turning; and the recovery state is used to achieve a smooth release of support force, improving ride comfort. There are clear logical relationships between these five states, collectively forming a closed-loop feedback mechanism to ensure that the system can make reasonable judgments and responses under various complex road conditions.

[0144] In this embodiment, a five-state machine and a multi-dimensional confidence calculation method are introduced to achieve intelligent recognition of a vehicle's turning intention. By introducing the five-state machine and the multi-dimensional confidence calculation method, the vehicle's driving state can be accurately determined, thereby enabling the generation of reasonable support strategies and ultimately improving passenger safety and ride comfort.

[0145] In some embodiments, Figure 3 A flowchart illustrating a vehicle control method provided in this application embodiment. Figure 3 ,like Figure 5 As shown, after determining the target support pressure corresponding to the predicted lateral displacement in S104 above, the above method may further include the following steps: S401, obtain the driving mode of the vehicle during the current driving process.

[0146] Here, driving mode refers to a driving state selected by the user through vehicle settings, used to define the vehicle's power output characteristics, suspension stiffness, steering feedback, and the behavior logic of assistance systems. Driving modes can include, but are not limited to, Comfort mode, Standard mode, and Sport mode.

[0147] In Comfort mode, the vehicle is tuned for smoothness and fuel efficiency, reducing power response to improve ride comfort; Standard mode is the default setting, balancing performance and energy consumption; Sport mode enhances power output and handling response, suitable for aggressive driving scenarios.

[0148] S402, determine the correction factor corresponding to the driving mode based on the vehicle's driving mode.

[0149] Here, the correction factor is a numerical parameter set based on different driving modes, used to adjust the final output value of the target support pressure. The role of the correction factor is to make a weighted adjustment to the target support pressure, thereby adapting to the actual needs of the current driving scenario. For example, in comfort mode, the correction factor may be less than 1 to reduce the support intensity and avoid excessive interference with the occupant's natural movements; in sport mode, the correction factor may be greater than 1 to enhance the support effect and ensure that the occupant can maintain a stable seating position during sharp turns.

[0150] It should be noted that the correction factor can be set based on a large amount of real vehicle test data and ergonomic research to ensure that it can provide the best support effect in different driving modes.

[0151] In one possible implementation, the above-mentioned S402 "determining the correction coefficient corresponding to the driving mode based on the vehicle's driving mode" may include the following steps: When the driving mode is in comfort mode, the correction factor corresponding to the driving mode is determined as the first correction factor; When the driving mode is set to standard mode, the correction factor corresponding to the driving mode is determined to be the second correction factor. When the driving mode is Sport mode, the correction factor corresponding to the driving mode is determined to be the third correction factor; wherein, the first correction factor is less than the second correction factor, and the second correction factor is less than the third correction factor.

[0152] It's understandable that in Comfort mode, occupants prioritize ride comfort, resulting in a smaller correction coefficient and lower support. Conversely, in Sport mode, occupants may prefer stronger support, leading to a larger correction coefficient and higher support. This support strategy design allows users to flexibly adjust the support strategy according to their preferences and driving environment, achieving a personalized experience. For example, the first correction coefficient can be 0.8; the second correction coefficient can be 1; and the third correction coefficient can be 1.2.

[0153] S303, based on the correction coefficient corresponding to the driving mode and the target support pressure, determines the fourth support pressure.

[0154] Here, the fourth support pressure refers to the actual support pressure that should be applied to the airbag after considering the driving mode correction factor. For example, the formula for calculating the fourth support pressure can be expressed as: (16) in, This can be represented as the fourth support / resistance level; It can represent the target support pressure; It can represent the correction coefficient corresponding to the driving mode.

[0155] Furthermore, the target support pressure obtained from the original prediction is multiplied by the driving mode correction coefficient to generate a fourth support pressure value adapted to the current driving scenario, and this value is used to control the inflation and deflation of the airbag. The control system transmits the fourth support pressure to the execution module, which drives the airbag to respond quickly to ensure that the support action is completed before the occupant feels the lateral force.

[0156] It should be noted that the airbag arrangement takes into account a balance between ergonomics and support effectiveness. The four airbags are located at key support positions of the seat: the backrest airbag provides lateral support for the lumbar region and shoulders. Rapid inflation ensures the effectiveness of predictive control, while slow deflation avoids discomfort caused by sudden pressure changes.

[0157] In some embodiments, the seat airbags of a vehicle can be triggered earlier when the support pressure is determined, in order to address the lag problem of traditional passive systems.

[0158] As can be understood, advance trigger time refers to the time interval between prediction and the rise in lateral acceleration to the execution of the support action. Advance trigger time can take into account the hardware response delay of the airbag system (such as valve response time and airbag inflation time) to ensure that the airbag inflates before the occupant feels the centrifugal force, thereby achieving true predictive support.

[0159] For example, the formula for calculating the advance trigger time can be expressed as: (17) in, Valve response time Inflation time for the airbag.

[0160] Understandably, after determining the support pressure, the vehicle's airbags can be inflated at an earlier trigger time to ensure that the airbags are fully inflated before the occupants feel the lateral force, thereby improving user comfort.

[0161] In this embodiment, the current driving mode is obtained, and a corresponding correction coefficient is determined based on the driving mode. Then, the fourth support pressure is calculated in conjunction with the target support pressure. By adjusting the output mode of the support pressure using this method, the output of the support pressure can be made more targeted and adaptable, thereby meeting the needs of different driving scenarios and further improving the comfort and safety of passengers during vehicle turning.

[0162] In some embodiments of this application, Figure 5 This is a schematic diagram of the overall process of a vehicle control method provided in an embodiment of this application, such as... Figure 6 As shown, the method includes S501 to S511, wherein: S501 inputs the steering angle, steering angular velocity, lateral acceleration, yaw rate, and driving speed into the steering intention state machine to determine the confidence levels of the steering angle, steering angular velocity, lateral acceleration, yaw rate, and signal consistency.

[0163] S502 performs weighted fusion processing on the confidence scores of steering angle, steering angular velocity, lateral acceleration, yaw rate, and signal consistency to determine the total confidence score.

[0164] S503 determines the vehicle's steering intention state based on the total confidence level.

[0165] S504, determine whether the turning intention state is in any of the ready, active, and stable states. If so, execute S505.

[0166] S505, determine the target predicted value of lateral acceleration and user type parameters.

[0167] S506 determines the user's predicted lateral displacement based on the statistical values ​​between the target predicted value of lateral acceleration, acceleration threshold, mass parameters, standard mass parameters, and response factors.

[0168] S507, determine whether the predicted lateral displacement is less than the first displacement threshold. If yes, execute S508; otherwise, execute S509.

[0169] S508, the target support pressure is determined as the first support pressure.

[0170] S509, determine whether the predicted lateral displacement is less than the second displacement threshold. If yes, execute S510; otherwise, execute S511.

[0171] S510, the target support pressure is determined as the second support pressure.

[0172] S511, the target support pressure is determined to be the third support pressure.

[0173] In some embodiments of this application, Figure 6 A schematic diagram of the overall architecture of a vehicle control method provided in this application embodiment is shown below. Figure 7a As shown, the overall architecture may include: a signal input module 601, a core processing module 602, and an output module 603. Wherein: The signal input module 601 may include: a CAN bus input unit 6011, a system sensor input unit 6012, an occupant type input unit 6013, and a user input unit 6014.

[0174] The CAN bus input unit 6011 can receive information such as steering angle, vehicle speed, lateral acceleration, and yaw rate; the system sensor input unit 6012 can acquire information such as air tank pressure and compressor temperature to ensure normal air tank pressure and prevent compressor overheating; the occupant type input unit 6013 can acquire occupant size (body type, height, etc.); and the user input unit 6014 can acquire system switch signals and driving modes. The occupant type input unit 6013 can be input by the user, calculated by a seat pressure sensor, or identified by an in-cabin sensing system; the system switch signals and driving modes can be obtained through user settings in the cockpit.

[0175] The core processing module 602 may include: a multi-sampling rate signal fusion processing unit 6021, a turning intention recognition and status judgment unit 6022, a lateral acceleration prediction calculation unit 6023, a personalized occupant response prediction unit 6024, an intelligent support strategy generation unit 6025, an airbag execution and pressure control unit 6026, and a system management unit 6027.

[0176] Among them, the multi-sampling rate signal fusion processing unit 6021 serves as the data input basis and is responsible for processing multiple signals from different sensors.

[0177] The turning intention recognition and state judgment unit 6022 adopts a five-state state machine architecture, based on in-depth physical analysis of the turning process and extensive experimental verification.

[0178] The lateral acceleration prediction unit 6023 employs three prediction methods for parallel computation, which can be combined to obtain the predicted value of lateral acceleration based on the advantages of each method in different scenarios.

[0179] The personalized occupant response prediction unit 6024 considers the significant differences in response among occupants of different body types under the same lateral acceleration, as a uniform support strategy cannot meet the needs of everyone. Occupants are divided into three categories: light body type (<60kg) including children and smaller adults; standard body type (60-80kg) including average adults; and heavy body type (>80kg) including larger adults.

[0180] The intelligent support strategy generation unit 6025 embodies the concept of hierarchical control. The three-level support strategy is designed based on extensive subjective evaluation experiments: the off setting (≤15mm) is suitable for slight turns, requiring no additional support (or only slight support); the medium setting (15-40mm) is suitable for normal turns, providing comfortable lateral support; and the strong setting (>40mm) is suitable for sharp turns or emergency avoidance, providing maximum support. This hierarchical strategy avoids the discomfort caused by excessive support while ensuring sufficient support is provided when needed.

[0181] The airbag actuation and pressure control unit 6026 translates the control strategy into physical actions. The hardware system is designed with reliability and response speed in mind. It consists of an air storage system, a compressor, and a valve system. All valves are fast-acting solenoid valves with a switching time of less than 20 milliseconds, and pulse width modulation (PWM) control enables precise pressure regulation.

[0182] The System Management Unit 6027 monitors the overall system's operational status, performance data, and fault codes. Through real-time data monitoring, the System Management module ensures coordinated operation of all submodules and can promptly trigger diagnostic procedures when anomalies are detected. Furthermore, this module records system status logs and event logs, providing data support for subsequent analysis and system optimization.

[0183] The output module 603 may include functions such as a five-valve control unit 6031, a compressor control unit 6032, a diagnostic output unit 6033, and a user feedback unit 6034, allowing users to provide feedback to optimize system performance. The user feedback module can display system status, performance data, and event logs. Specifically, the five-valve control unit 6031 also includes driver's left airbag, driver's right airbag, passenger's left airbag, and passenger's right airbag; the compressor control unit 6032 also includes a compressor switch and a vent valve; the diagnostic output unit 6033 also includes fault code output, system status output, performance data output, and event log output; and the user feedback unit 6034 also includes audible feedback and displayed information feedback.

[0184] The following describes the application of the vehicle control method provided in the embodiments of this application in a real-world scenario.

[0185] This application provides a method for predictively activating lateral support 200-500 milliseconds before occupants leave their positions during vehicle turning. By identifying the driver's turning intention in advance and predicting the development trend of lateral acceleration, active support is provided before the occupants are subjected to centrifugal force.

[0186] Multi-sampling-rate signal fusion processing: Acquire signals from multiple vehicle sensors, perform noise suppression and smoothing through an adaptive Kalman filter, and use a time synchronization method to align signals with different sampling frequencies to a unified time reference; Turning Intent Recognition and State Judgment: Intelligent recognition of the driver's turning intent is achieved through a multi-state machine, multi-dimensional confidence is calculated and the total confidence is calculated through weighted fusion, and state transition is performed based on dynamic thresholds; Lateral acceleration prediction calculation: Three prediction methods, namely steady-state prediction, transient prediction and sensor extrapolation, are used in parallel to calculate and obtain accurate lateral acceleration prediction values ​​through adaptive weight fusion. Personalized occupant response prediction: Calculate personalized lateral displacement prediction values ​​based on occupant type identification results, taking into account the response characteristics and trigger threshold differences of different occupants; Intelligent support strategy generation: Based on the predicted lateral displacement, a graded support strategy is generated, which is then adjusted in combination with the driving mode to achieve predictive control by calculating the advance trigger time. Airbag execution and pressure control: Executes support strategies to achieve rapid-response airbag inflation and deflation control, providing precise lateral support force output.

[0187] For example, Figure 7a This is a schematic diagram illustrating the relationship between time and vehicle speed, provided for an embodiment of this application; the Y-axis represents vehicle speed (km / h). Figure 7b This displays the vehicle's speed change curve during a 35-second test. The vehicle's acceleration, deceleration, or constant speed driving can be observed. The smoothness or fluctuation of the speed curve can reflect the driver's throttle / brake operation or cruise control performance.

[0188] For example, Figure 7b A schematic diagram illustrating the relationship between time and predicted lateral acceleration provided for embodiments of this application; Figure 7c This displays the expected or calculated lateral (left-right) acceleration changes of the vehicle while turning. The peak of the curve corresponds to moments of sharp turns or rapid steering wheel movements.

[0189] For example, Figure 7c A schematic diagram illustrating the relationship between time and steering wheel angle provided in this application embodiment; Figure 7d It records the steering wheel rotation amplitude of the driver (or autonomous driving system). Positive and negative values ​​typically represent turning left and right, respectively. The shape of the curve directly reflects the steering command: gentle changes represent smooth steering, while rapid, large changes represent emergency obstacle avoidance or aggressive driving.

[0190] For example, Figure 7d A schematic diagram illustrating the relationship between time and steering wheel angular velocity provided in an embodiment of this application; Figure 7e This displays the speed of steering wheel rotation, i.e., the rate of change of angle. A high angular velocity indicates that the driver is turning the steering wheel rapidly, indicating aggressive steering. A low angular velocity indicates gentle steering. This parameter is crucial for judging the urgency of the driver's intentions.

[0191] For example, Figure 7e A schematic diagram illustrating the relationship between time and yaw rate provided in this application embodiment; Figure 8This displays the angular velocity of the vehicle's rotation about its vertical axis (the axis passing through the center of the roof and chassis). This is a core physical quantity for measuring a vehicle's steering response and stability. It directly reflects the vehicle's tendency to "fishtail" or "yaw." Under normal circumstances, it should have a definite correspondence with steering wheel angle and vehicle speed. If the actual measured yaw rate deviates too much from the value predicted based on the model, the electronic stability program intervenes, correcting the vehicle's attitude by braking individual wheels to prevent loss of control.

[0192] In some implementations, the processing procedure of a multi-state machine is described in detail. For example, state nodes may include: NONE (no intention state): the initial state, indicating that the system has no intention to turn; PREPARING (ready state): the state where the system is preparing to turn; ACTIVE (active state): the state where the system is actively turning; STEDAY (stable state): the state where the system is in a stable turning state; RECOVERING (recovering state): the state where the system is recovering.

[0193] The state machine can record the vehicle's previous state and calculate the duration of that state. If the current state is an unintentional state, and the confidence level is >0.3, it switches to the PREPARING state, records the start time, and increments the state transition counter by 1. Otherwise, it remains in the NONE state.

[0194] If the current state is PREPARING, and the confidence level is >0.6, switch to ACTIVE state and increment the state transition counter by 1; if the confidence level is <0.2 or the duration is >2000ms, switch to NONE state and increment the state transition counter by 1; otherwise, remain in PREPARING state.

[0195] If the current state is ACTIVE, and the confidence level is >0.8 and the duration is >500ms, switch to the STEDAY state and increment the state transition counter by 1. If the confidence level is <0.4, switch to the RECOVERING state and increment the state transition counter by 1; otherwise, remain in the ACTIVE state.

[0196] If the current state is STEDAY, and the confidence level is less than 0.4, switch to the RECOVERING state and increment the state transition counter by 1; otherwise, maintain the STEDAY state.

[0197] If the current state is RECOVERING, switch to NONE state if the confidence level is <0.2 or the duration is >3000ms, and increment the state transition counter by 1; switch to ACTIVE state if the confidence level is >0.5, and increment the state transition counter by 1; otherwise, remain in RECOVERING state.

[0198] Furthermore, after each state transition, the state duration can be updated. Then, it can be determined whether the state has changed. If the state changes, the turning intention detection flag (ACTIVE||STEDAY) can be updated; if the state does not change, the process ends or waits for the next state determination.

[0199] Based on the above embodiments, this application also provides a vehicle control device. Figure 8 This is a schematic diagram of the composition structure of a vehicle control device provided in an embodiment of this application, as shown below. Figure 9 As shown, the vehicle control device 800 includes a first determining unit 801, a second determining unit 802, a third determining unit 803, and a fourth determining unit 804, wherein: The first determining unit 801 is used to determine the steering intention state of the vehicle based on the acquired vehicle operating state parameters; wherein the steering intention state includes no intention state, ready state, active state, stable state and recovery state. The second determining unit 802 is used to determine the target predicted value of lateral acceleration and user type parameters when the steering intention state is in any of the ready state, active state, and stable state; wherein, the user type parameters include the acceleration threshold corresponding to the user type, the mass parameter corresponding to the user type, and the response factor corresponding to the user type. The third determining unit 803 is used to determine the user's predicted lateral displacement based on the target predicted value of lateral acceleration, acceleration threshold, mass parameter, standard mass parameter and response factor statistical value; The fourth determining unit 804 is used to determine the corresponding target support pressure based on the user's predicted lateral displacement, so as to control the vehicle's seat airbag to inflate to the target support pressure.

[0200] In some embodiments of this application, the operating state parameters include steering angle, steering angular velocity, lateral acceleration, yaw rate, and driving speed; the first determining unit 801 is further configured to input the steering angle, steering angular velocity, lateral acceleration, yaw rate, and driving speed into the steering intention state machine, determine the steering angle confidence, steering angular velocity confidence, lateral acceleration confidence, yaw rate confidence, and signal consistency confidence; perform weighted fusion processing on the steering angle confidence, steering angular velocity confidence, lateral acceleration confidence, yaw rate confidence, and signal consistency confidence to determine the total confidence; and determine the vehicle's steering intention state based on the total confidence.

[0201] In some embodiments of this application, the first determining unit 801 is further configured to: determine that the vehicle's steering intention is in a state of no intention when the total confidence level is less than a first threshold; determine that the vehicle's steering intention is in a state of preparation when the total confidence level is greater than the first threshold and less than a second threshold, and the duration is greater than a first time threshold; determine that the vehicle's steering intention is in a state of activation when the total confidence level is greater than the second threshold and less than a third threshold; determine that the vehicle's steering intention is in a state of stability when the total confidence level is greater than the third threshold and the duration is greater than the second time threshold; and determine that the vehicle's steering intention is in a state of recovery when the total confidence level is less than a fourth threshold and greater than the first threshold; wherein the fourth threshold is less than the first threshold, the first threshold is less than the second threshold, the second threshold is less than the third threshold, and the first time threshold is less than the second time threshold.

[0202] In some embodiments of this application, the second determining unit 802 is further configured to: determine a first predicted value of lateral acceleration at a second time moment based on statistical values ​​between the steering wheel angle, steering ratio, wheelbase, driving speed, and steering gradient at a first time moment; the first time moment being before the second time moment; determine a second predicted value of lateral acceleration at the second time moment based on the lateral acceleration at the first time moment, the first derivative of the lateral acceleration, and the second derivative of the lateral acceleration; perform fitting processing on one or more historical lateral accelerations to determine a third predicted value of lateral acceleration at the second time moment; and perform weighted fusion processing on the first predicted value, the second predicted value, and the third predicted value to determine a target predicted value of lateral acceleration.

[0203] In some embodiments of this application, the fourth determining unit 804 is further configured to determine the target support pressure as the first support pressure when the predicted lateral displacement is less than the first displacement threshold; determine the target support pressure as the second support pressure when the predicted lateral displacement is greater than the first displacement threshold and less than the second displacement threshold; and determine the target support pressure as the third support pressure when the predicted lateral displacement is greater than the second displacement threshold; wherein the first support pressure is less than the second support pressure, and the second support pressure is less than the third support pressure.

[0204] In some embodiments of this application, after the fourth determining unit 804, the vehicle control device 800 further includes: an acquisition unit, a fifth determining unit, and a sixth determining unit, wherein: The acquisition unit is used to acquire the driving mode of the vehicle during the current driving process; wherein the driving mode includes comfort mode, standard mode and sport mode; The fifth determining unit is used to determine the correction coefficient corresponding to the driving mode based on the vehicle's driving mode. The sixth determining unit is used to determine the fourth supporting pressure based on the correction coefficient corresponding to the driving mode and the target supporting pressure.

[0205] In some embodiments of this application, the fifth determining unit is further configured to determine the correction coefficient corresponding to the driving mode as a first correction coefficient when the driving mode is comfort mode; determine the correction coefficient corresponding to the driving mode as a second correction coefficient when the driving mode is standard mode; and determine the correction coefficient corresponding to the driving mode as a third correction coefficient when the driving mode is sport mode; wherein the first correction coefficient is less than the second correction coefficient, and the second correction coefficient is less than the third correction coefficient.

[0206] It should be noted that, in the embodiments of this application, if the above methods are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of software products. These software products are stored in a storage medium and include several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0207] This application also provides a vehicle including a memory and a processor, the memory storing a computer program that can run on the processor, and the processor executing the computer program to implement the above-described method.

[0208] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. The computer-readable storage medium can be transient or non-transient.

[0209] This application also provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement some or all of the steps in the above-described method. This computer program product can be implemented specifically through hardware, software, or a combination thereof. In one optional embodiment, the computer program product is specifically embodied as a computer storage medium; in another optional embodiment, the computer program product is specifically embodied as a software product, such as a software development kit (SDK), etc.

[0210] It should be noted that, Figure 9 This is a schematic diagram of the hardware entity of a vehicle provided in an embodiment of this application, such as... ​As shown, the hardware entity of the vehicle 900 includes: a processor 901, a communication interface 902, and a memory 903, wherein: The processor 901 typically controls the overall operation of the vehicle 900.

[0211] The communication interface 902 enables the vehicle 900 to communicate with other terminals or servers via a network.

[0212] The memory 903 is configured to store instructions and applications executable by the processor 901, and can also cache data to be processed or already processed by the processor 901 and various modules in the vehicle 900 (e.g., image data, audio data, voice communication data, and video communication data), and can be implemented using flash memory (FLASH) or RAM. Data can be transferred between the processor 901, the communication interface 902, and the memory 903 via the bus 904.

[0213] It should be noted that the descriptions of the storage medium and device embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage medium and device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0214] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above steps / processes do not imply a sequential order of execution; the execution order of each step / process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above embodiments of this application are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0215] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0216] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components may be combined, or integrated into another system, or some features may be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0217] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0218] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0219] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory, magnetic disks, or optical disks.

[0220] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence or the part that contributes to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0221] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A vehicle control method, characterized in that, The method includes: Based on the acquired vehicle operating status parameters, the vehicle's steering intention state is determined; wherein, the steering intention state includes no intention state, ready state, active state, stable state, and recovery state; When the steering intention state is in any of the ready state, the active state, and the stable state, the target predicted value of the lateral acceleration and the user type parameter are determined; wherein, the user type parameter includes the acceleration threshold corresponding to the user type, the mass parameter corresponding to the user type, and the response factor corresponding to the user type; The predicted lateral displacement of the user is determined based on the statistical values ​​between the target predicted value of the lateral acceleration, the acceleration threshold, the mass parameter, the standard mass parameter, and the response factor. Based on the user's predicted lateral displacement, the corresponding target support pressure is determined to control the vehicle's seat airbags to inflate to the target support pressure.

2. The method according to claim 1, characterized in that, The operating status parameters include steering angle, steering angular velocity, lateral acceleration, yaw rate, and driving speed; determining the vehicle's steering intention state based on the acquired operating status parameters includes: The steering angle, steering angular velocity, lateral acceleration, yaw rate, and driving speed are input into the steering intention state machine to determine the confidence levels of the steering angle, steering angular velocity, lateral acceleration, yaw rate, and signal consistency. The confidence scores of the steering angle, steering angular velocity, lateral acceleration, yaw rate, and signal consistency are weighted and fused to determine the total confidence score. The vehicle's steering intention state is determined based on the total confidence level.

3. The method according to claim 2, characterized in that, Determining the vehicle's steering intention state based on the total confidence level includes: If the total confidence level is less than a first threshold, the vehicle's steering intention is determined to be in an inattentive state. If the total confidence level is greater than a first threshold and less than a second threshold, and the duration is greater than a first time threshold, the vehicle's steering intention is determined to be in a ready state. If the total confidence level is greater than the second threshold and less than the third threshold, the vehicle's steering intention is determined to be active. If the total confidence level is greater than the third threshold and the duration is greater than the second time threshold, the vehicle's steering intention is determined to be in a stable state. If the total confidence level is less than a fourth threshold and greater than the first threshold, the vehicle's steering intention is determined to be in a recovery state; wherein the fourth threshold is less than the first threshold, the first threshold is less than the second threshold, the second threshold is less than the third threshold, and the first time threshold is less than the second time threshold.

4. The method according to claim 1, characterized in that, Determine the target predicted value of lateral acceleration, including: Based on the statistical values ​​between the steering wheel angle, steering ratio, wheelbase, driving speed, and steering gradient at the first moment, a first predicted value of the lateral acceleration at the second moment is determined; the first moment is before the second moment. Based on the lateral acceleration at the first moment, the first derivative of the lateral acceleration, and the second derivative of the lateral acceleration, the second predicted value of the lateral acceleration at the second moment is determined. By fitting one or more historical lateral accelerations, a third predicted value of the lateral acceleration at the second time moment is determined; The first predicted value, the second predicted value, and the third predicted value are weighted and fused to determine the target predicted value of the lateral acceleration.

5. The method according to claim 1, characterized in that, The step of determining the corresponding target support pressure based on the user's predicted lateral displacement includes: If the predicted lateral displacement is less than the first displacement threshold, the target support pressure is determined as the first support pressure. If the predicted lateral displacement is greater than a first displacement threshold and less than a second displacement threshold, the target support pressure is determined to be the second support pressure. If the predicted lateral displacement is greater than the second displacement threshold, the target support pressure is determined as the third support pressure; wherein the first support pressure is less than the second support pressure, and the second support pressure is less than the third support pressure.

6. The method according to claim 5, characterized in that, After determining the target support pressure corresponding to the predicted lateral displacement, the method further includes: The driving mode of the vehicle during the current driving process is obtained; wherein the driving mode includes comfort mode, standard mode and sport mode; Based on the vehicle's driving mode, determine the correction coefficient corresponding to the driving mode; The fourth support pressure is determined based on the correction coefficient corresponding to the driving mode and the target support pressure.

7. The method according to claim 6, characterized in that, The step of determining the correction coefficient corresponding to the driving mode based on the vehicle's driving mode includes: When the driving mode is the comfort mode, the correction factor corresponding to the driving mode is determined to be the first correction factor; When the driving mode is the standard mode, the correction factor corresponding to the driving mode is determined to be the second correction factor; When the driving mode is the sport mode, the correction coefficient corresponding to the driving mode is determined to be the third correction coefficient; wherein the first correction coefficient is less than the second correction coefficient, and the second correction coefficient is less than the third correction coefficient.

8. A vehicle control device, characterized in that, The device includes: The first determining unit is used to determine the steering intention state of the vehicle based on the acquired operating state parameters of the vehicle and the steering intention state machine; wherein the steering intention state includes no intention state, ready state, active state, stable state and recovery state. The second determining unit is configured to determine the user type parameter and the target predicted value of the lateral acceleration when the steering intention state is in any one of the ready state, the active state, and the stable state; wherein the user type parameter includes an acceleration threshold corresponding to the user type, a mass parameter corresponding to the user type, and a response factor corresponding to the user type. The third determining unit is used to determine the user's predicted lateral displacement based on the target predicted value of the lateral acceleration, the acceleration threshold, the mass parameter, the standard mass parameter, and the statistical value between the response factor; The fourth determining unit is used to determine the corresponding target support pressure based on the user's predicted lateral displacement, so as to control the vehicle's seat airbag to inflate to the target support pressure.

9. A vehicle comprising a processor and a memory, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1 to 7.

Images