Vehicle control method, electronic device, and vehicle

Abstract

This disclosure relates to the field of intelligent driving technology for vehicles, providing a vehicle control method, electronic device, and vehicle. The method involves acquiring the road surface adhesion coefficient of the road surface where the vehicle is located, determining that the road surface adhesion coefficient is less than a preset adhesion threshold, acquiring the wheel speeds of the four wheels and vehicle operating information, acquiring the target torque change rate corresponding to the vehicle's drive wheels, determining a first drive torque and a target slip level based on the wheel speeds, operating information, and target torque change rate, acquiring the initial drive torque of the vehicle's drive wheels, determining a target drive torque based on the target slip level, initial drive torque, and first drive torque, and controlling vehicle movement based on the target drive torque. This disclosure comprehensively considers wheel speeds and slip levels when determining the target drive torque, resulting in a target drive torque that better matches the vehicle's current actual situation. This achieves graded slip control under high-frequency disturbance scenarios, ensuring vehicle stability and safety, and thus improving the user's driving experience.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent driving technology for vehicles, and in particular to a vehicle control method, electronic device, and vehicle. Background Technology

[0002] With the rapid development of vehicle technology, vehicles have become an important means of transportation in people's daily lives. Currently, in complex road conditions with low traction, the tire traction state of vehicles often switches rapidly within 0.5 seconds, resulting in frequent wheel speed disturbances and strong sudden slippage.

[0003] Currently, torque control strategies largely rely on closed-loop feedback and empirical threshold judgments, resulting in a single control strategy that cannot achieve dynamic identification and differentiated response to risk levels. Summary of the Invention

[0004] In view of this, the purpose of this disclosure is to propose a vehicle control method, electronic device and vehicle to solve the problem that current torque control strategies rely heavily on closed-loop feedback and empirical threshold judgment, lack modeling of inter-wheel disturbance relationships, and are unable to achieve dynamic identification and differentiated response to risk levels.

[0005] To achieve the above objectives, a first aspect of this disclosure provides a vehicle control method, the method comprising:

[0006] Obtain the road surface adhesion coefficient of the road surface where the vehicle is located, determine that the road surface adhesion coefficient is less than a preset adhesion threshold, and obtain the wheel speed of the four wheels of the vehicle and the vehicle operation information; Obtain the target torque change rate corresponding to the vehicle's drive wheels, and determine the first drive torque and target slip level based on the four wheel speeds, the operating information, and the target torque change rate. The initial drive torque of the vehicle's drive wheels is obtained, and a target drive torque is determined based on the target slip level, the initial drive torque, and the first drive torque. The vehicle's movement is controlled based on the target drive torque.

[0007] Based on the same inventive concept, a second aspect of this disclosure provides a vehicle control device, comprising: The data acquisition module is configured to acquire the road surface adhesion coefficient of the road surface where the vehicle is located, determine that the road surface adhesion coefficient is less than a preset adhesion threshold, and acquire the wheel speed of the four wheels of the vehicle and vehicle operation information. The grade determination module is configured to acquire the target torque change rate corresponding to the vehicle's drive wheels, and determine the first drive torque and target slip grade based on the four wheel speeds, the operating information, and the target torque change rate. The vehicle control module is configured to acquire the initial drive torque of the vehicle's drive wheels, determine a target drive torque based on the target slip level, the initial drive torque, and the first drive torque, and control the vehicle's movement based on the target drive torque.

[0008] Based on the same inventive concept, a third aspect of this disclosure proposes an electronic device including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the vehicle control method as described above when executing the computer program.

[0009] Based on the same inventive concept, a fourth aspect of this disclosure provides a non-transitory computer-readable storage medium that stores computer instructions for causing a computer to perform the vehicle control method as described above.

[0010] Based on the same inventive concept, the fifth aspect of this disclosure provides a vehicle including the vehicle control device described in the second aspect, the electronic device described in the third aspect, or the storage medium described in the fourth aspect.

[0011] As can be seen from the above, this disclosure proposes a vehicle control method, electronic device, and vehicle. The method involves acquiring the road surface adhesion coefficient of the road surface where the vehicle is located, determining that the road surface adhesion coefficient is less than a preset adhesion threshold, and acquiring the wheel speeds of the four wheels and vehicle operation information. Because the tire adhesion state of the vehicle only changes frequently in a short period of time on low-adhesion roads, the wheel speeds of the four wheels and vehicle operation information are acquired only when the road surface adhesion coefficient is less than the preset adhesion threshold. This information is then used to determine the target driving torque, avoiding unnecessary data acquisition operations. The method also involves acquiring the target torque change rate corresponding to the vehicle's drive wheels, and determining the first driving torque and target slip level based on the wheel speeds, the operation information, and the target torque change rate. The target slip level represents the risk level of vehicle slippage, reflecting the impact of slippage on smooth driving. Furthermore, when determining the target driving torque, the target slip level is considered, quantifying the slippage risk into a controllable indicator. The driving torque is limited according to the slip level to prevent excessive wheel slippage, ensuring vehicle stability, while simultaneously suppressing significant wheel speed fluctuations at high slip levels and improving driving smoothness. The initial drive torque of the vehicle's drive wheels is obtained. A target drive torque is determined based on the target slip level, the initial drive torque, and the first drive torque. Vehicle movement is controlled based on the target drive torque. When determining the target drive torque, wheel speed and slip level are comprehensively considered, resulting in a target drive torque that better matches the vehicle's current actual conditions. This achieves graded slip control under high-frequency disturbance scenarios, ensuring vehicle stability and safety, and ultimately improving the user experience. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in this disclosure or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a vehicle control method according to an embodiment of the present disclosure; Figure 2 This is a structural block diagram of a vehicle control device according to an embodiment of the present disclosure; Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0015] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this disclosure should have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms "first," "second," and similar terms used in the embodiments of this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0016] The following are definitions of terms used in this disclosure: GNN: Graph Neural Network (GNN) is a deep learning model specifically designed for processing graph-structured data. It learns representations of nodes and the graph by iteratively aggregating information about the neighbors of nodes in the graph, effectively capturing complex relationships and dependencies between nodes.

[0017] LightGBM: LightGBM (Light Gradient Boosting Machine) is a high-performance gradient boosting framework developed by Microsoft, designed for handling large-scale data and improving computational efficiency. Based on the decision tree algorithm, it significantly reduces memory usage and computational complexity while maintaining high prediction accuracy through innovative techniques such as histogram-based decision tree learning, one-sided gradient sampling (GSS), and mutually exclusive feature binding (EFB).

[0018] Takagi-Sugeno: The Takagi-Sugeno (TS) fuzzy system is an efficient fuzzy inference method widely used in the modeling and control of complex nonlinear systems. Unlike traditional fuzzy systems that use fuzzy sets as outputs, the TS model designs the premise part (if part) as fuzzy language rules, while the conclusion part (then part) uses linear function expressions of the input variables.

[0019] Currently, on complex road surfaces with low adhesion, such as grasslands, slippery sand, and muddy moss, the contact state between vehicle tires and the ground often exhibits extremely unstable dynamic characteristics. The coefficient of adhesion can undergo a rapid cycle of "sticking, sliding, and sticking again" within just 0.5 seconds. This high-frequency and large-amplitude fluctuation in adhesion causes continuous disturbances in wheel speed signals, and slippage occurs with a high degree of suddenness and randomness, greatly increasing the difficulty of vehicle dynamic control.

[0020] Faced with such complex operating conditions, most torque control strategies based on wheel speed feedback in related technologies rely on post-event closed-loop adjustment and fixed empirical thresholds for slip judgment, essentially constituting a passive response control. Its core limitation lies in the lack of proactive perception of the sources of changes in road surface adhesion and the internal structure of disturbances, as well as the inability to effectively predict the generation and development trend of slip. Consequently, the control system often exhibits significant response lag and a high false judgment rate, making it difficult to achieve proactive intervention and smooth control from a feedforward perspective.

[0021] Especially in high-risk scenarios like multi-wheel asymmetric slip, where different wheels may be in different states of attachment simultaneously or sequentially, relevant control models often treat each wheel as an independent system, failing to effectively establish a dynamic description of the coupling relationship between wheel disturbances. This leads to control strategies that tend to be simplistic and static. The system cannot identify different levels of risk based on real-time road conditions, making it difficult to apply differentiated torque control to wheels in different slip states, resulting in a severe lack of overall flexibility.

[0022] Furthermore, in existing control architectures, the channels often operate independently, lacking effective linkage and decoupling mechanisms. When dealing with complex disturbances, intervention in a single wheel may trigger conflicts or oscillations in torque distribution, easily leading to vehicle drive instability and even exacerbating slippage, thus creating negative control feedback.

[0023] To address the aforementioned problems, this embodiment proposes a vehicle control method, such as... Figure 1 As shown, the method includes: Step 101: Obtain the road surface adhesion coefficient of the road surface where the vehicle is located, determine that the road surface adhesion coefficient is less than a preset adhesion threshold, and obtain the wheel speed of the four wheels of the vehicle and the vehicle operation information.

[0024] In practice, the road surface adhesion coefficient of the road surface where the vehicle is located is obtained, and the road surface adhesion coefficient is compared with a preset adhesion threshold. If it is determined that the road surface adhesion coefficient is greater than or equal to the preset adhesion threshold, it means that the road surface where the vehicle is located is not a low-adhesion road surface. At this time, the vehicle tire adhesion state will not switch rapidly, thus avoiding frequent wheel speed disturbances. Therefore, there is no need to control the drive torque.

[0025] If the road surface adhesion coefficient is less than the preset adhesion threshold, it means that the vehicle is driving on a low-adhesion road surface. The vehicle's four-wheel wheel speed and vehicle operation information are obtained, including vehicle acceleration, vehicle yaw rate, vehicle slip ratio, etc.

[0026] In this embodiment, the vehicle acceleration includes lateral acceleration and longitudinal acceleration. The lateral acceleration represents the acceleration component along the Y-axis of the vehicle's coordinate system, which is perpendicular to the direction of travel and points to the driver's left. The longitudinal acceleration represents the acceleration component along the X-axis of the vehicle's coordinate system, which is the acceleration component along the direction of travel.

[0027] Step 102: Obtain the target torque change rate corresponding to the vehicle drive wheels, and determine the first drive torque and target slip level based on the four wheel speeds, the operating information and the target torque change rate.

[0028] In specific implementation, the target torque change rate corresponding to the vehicle's drive wheels is obtained, wherein the target torque change rate represents the real-time torque change of the vehicle's drive wheels per unit time. The first drive torque and the target slip level are determined based on the four wheel speeds, the operating information, and the target torque change rate.

[0029] In this embodiment, the target slip level represents the risk level of vehicle slippage, reflecting the impact of slippage on vehicle smooth driving. Specifically, the target slip level characterizes the risk level corresponding to slippage during vehicle operation, directly reflecting the degree of impact of slippage on the vehicle's smooth driving state. Specifically, the slip level is related to the slip ratio; a higher slip ratio means poorer adhesion between the tires and the road surface, making the vehicle more prone to loss of stability, and consequently, a higher slip level, indicating a greater risk during vehicle operation. Furthermore, a higher slip level results in more pronounced wheel speed fluctuations, specifically manifested as increased amplitude fluctuations and more drastic frequency changes in the wheel speed signal, further affecting the coordinated control of the vehicle's powertrain and driving stability.

[0030] For example, the target slip level is divided into five levels, from lowest to highest: Level 1, Level 2, Level 3, Level 4, and Level 5. The slip ratio at target slip level 1 is lower than that at target slip level 5. Simultaneously, at target slip level 1, wheel speed fluctuations are relatively small. At target slip level 2, wheel speed exhibits short-term fluctuations. At target slip level 3, wheel speed fluctuations are significant but do not lead to attitude instability. At target slip level 4, wheel speed fluctuates violently and is accompanied by abnormal acceleration. At target slip level 5, wheel speed deviates significantly from vehicle speed, causing a significant decrease in directional stability.

[0031] Step 103: Obtain the initial drive torque of the vehicle's drive wheels, determine the target drive torque based on the target slip level, the initial drive torque, and the first drive torque, and control the vehicle's movement based on the target drive torque.

[0032] In specific implementation, the initial drive torque of the vehicle's drive wheels is obtained, wherein the initial drive torque is the drive torque corresponding to the vehicle's original control strategy on the road surface. The target drive torque is determined based on the target slip level, the initial drive torque, and the first drive torque; that is, one of the initial drive torque and the first drive torque is selected as the target drive torque according to the target slip level.

[0033] After determining the target driving torque, the vehicle driving is controlled based on the target driving torque. That is, the target driving torque is sent to the underlying execution unit, such as the motor controller or engine ECU, so as to achieve intelligent, linear and efficient control of the vehicle driving force, and ultimately ensure that the actual driving state of the vehicle is highly consistent with the preset driving target.

[0034] The above method obtains the road surface adhesion coefficient of the vehicle's location, determines that the road surface adhesion coefficient is less than a preset adhesion threshold, and acquires the vehicle's four-wheel wheel speeds and vehicle operation information. Because the vehicle's tire adhesion state only changes frequently in a short period on low-traction surfaces, the vehicle's four-wheel wheel speeds and vehicle operation information are only acquired when the road surface adhesion coefficient is less than the preset adhesion threshold. This information is used to determine the target driving torque, avoiding unnecessary data acquisition operations. The target torque change rate corresponding to the vehicle's drive wheels is acquired, and a first driving torque and a target slip level are determined based on the four-wheel wheel speeds, the operation information, and the target torque change rate. The target slip level represents the risk level of vehicle slippage, reflecting the impact of slippage on smooth driving. Therefore, when determining the target driving torque, the target slip level is considered, quantifying the slippage risk into a controllable indicator. The driving torque is limited according to the slip level to prevent excessive wheel slippage, ensuring vehicle stability, while suppressing significant wheel speed fluctuations at high slip levels and improving driving smoothness. The initial drive torque of the vehicle's drive wheels is obtained. A target drive torque is determined based on the target slip level, the initial drive torque, and the first drive torque. Vehicle movement is controlled based on the target drive torque. When determining the target drive torque, wheel speed and slip level are comprehensively considered, resulting in a target drive torque that better matches the vehicle's current actual conditions. This achieves graded slip control under high-frequency disturbance scenarios, ensuring vehicle stability and safety, and ultimately improving the user experience.

[0035] In this embodiment, a raw data acquisition and preprocessing mechanism is designed to train the three-model system, focusing on the high-frequency slip behavior of the vehicle under the conditions of rapid switching between sticky, slippery and sticky road surfaces. This ensures that wheel speed disturbances, vehicle body posture changes and drive response signals can be synchronously aligned at the millisecond level.

[0036] Specifically, the input signals include data from the four-wheel speed sensors (sampling rate above 200Hz), vehicle yaw rate, longitudinal acceleration, lateral acceleration, steering wheel angle, accelerator pedal opening, drive motor torque request value, and actual output value. The acquisition system uses a combination of TBox local buffering and real-time CAN bus flow to ensure that data frame loss or out-of-order issues do not occur under high dynamic conditions. To address the high-frequency noise in the wheel speed signals and abnormal disturbances caused by terrain perturbations, a combination of sliding median filtering and zero-phase IIR filtering is used to reduce noise in all channels.

[0037] Meanwhile, to accurately capture wheel speed abrupt changes caused by adhesion mutations, a perturbation point labeling method based on dynamic time windows is introduced. When a wheel speed differential peak exceeding a set threshold is detected, the start and end boundaries of the perturbation are automatically located and labeled. All sensor data are normalized using a unified timestamp to construct fixed-dimensional training sample frames. Each frame contains complete multi-channel data within 0.5 seconds before and after the frame, serving as the raw input for the downstream GNN and LightGBM training stages.

[0038] In this embodiment, during the data cleaning process, abnormal slip samples caused by non-adhesion factors such as braking system intervention and sudden vehicle stops are filtered out, and road surface condition labels (grass, wet sand, moss, etc.) are labeled to form a private vehicle manufacturer road condition dataset with fast adhesion switching characteristics, providing high-reliability sample support for subsequent models.

[0039] In some embodiments, because the rapid switching of vehicle tire adhesion states in low-adhesion complex road conditions leads to frequent wheel speed disturbances and strong sudden slippage, the influence of the disturbing wheels and the direction of disturbance propagation is considered when determining the target drive torque to improve vehicle stability. Specifically, step 102, which involves determining the first drive torque and target slip level based on the four wheel speeds, the operating information, and the target torque change rate, includes: Step 1021: Determine the target slip level and target disturbance data based on the wheel speeds of the four wheels and the operating information, wherein the target disturbance data includes the target disturbance significance value, the target disturbance wheel, and the target disturbance propagation direction, and the target disturbance wheel is the starting wheel of the vehicle slip propagation; Step 1022: Determine the first driving torque based on the target slip level and the target disturbance significance value.

[0040] In practice, the target slip level and target disturbance data are determined based on the wheel speeds and operating information of the four wheels, and the first driving torque is determined based on the target slip level and the target disturbance significance value. The target disturbance data includes the target disturbance significance value, the target disturbance wheel, and the target disturbance propagation direction.

[0041] In this embodiment, the target disturbance significance value is used to determine whether the vehicle has a slip-level disturbance. If the target disturbance significance value is greater than a preset significance threshold, it is determined that the vehicle has a slip-level disturbance. If the target disturbance significance value is less than or equal to the preset significance threshold, it is determined that the vehicle does not have a slip-level disturbance.

[0042] In this embodiment, when the vehicle is at or near its grip limit, the slippage of one or more tires, whether due to longitudinal braking lock-up / driving slippage or lateral sideslip, will alter the force and torque balance of the entire vehicle, causing a dynamic transfer of load between the wheels. This may subsequently lead to other wheels losing grip, a process known as slip propagation. The target disturbance wheel is the initiating wheel of vehicle slip propagation. The initiating wheel is the tire that bears a greater longitudinal or lateral force, i.e., the tire that exceeds its grip limit due to excessive driving, braking, or steering input.

[0043] In this embodiment, when one or more tires exceed their grip limit, the loss of grip propagates from the initial wheel to other wheels through dynamic coupling (mainly load transfer and yaw moment imbalance) and potentially superimposed improper driver correction. The target disturbance propagation direction is the propagation direction corresponding to the spread from the initial wheel to other wheels.

[0044] Specifically, vehicle operating information includes vehicle acceleration, vehicle yaw rate, and vehicle slip ratio. The target disturbance data can be determined first, and then the slip level can be determined based on the target disturbance significance value in the target disturbance data, thus achieving accurate determination of the slip level. That is, step 1021, which involves determining the target slip level and target disturbance data based on the four-wheel wheel speeds and the operating information, specifically includes: Step 10211: Input the wheel speeds of the four wheels, the vehicle acceleration, and the vehicle yaw rate into the pre-trained disturbance recognition model, and process them through the disturbance recognition model to obtain the target disturbance data; Step 10212: Determine the target acceleration change rate based on the vehicle acceleration, input the target disturbance significance value, the vehicle slip rate, and the target acceleration change rate into the pre-trained slip level determination model, and obtain the target slip level through the slip level determination model.

[0045] In practice, a pre-trained disturbance recognition model is obtained, and the wheel speeds of the four wheels, the vehicle acceleration, and the vehicle yaw rate are input into the disturbance recognition model. The disturbance recognition model processes the data to obtain the target disturbance data.

[0046] In this embodiment, the training process of the disturbance recognition model specifically includes: Step a, obtain the first training dataset and the initial disturbance recognition model, wherein the first training dataset includes historical four-wheel wheel speeds, historical vehicle accelerations, historical yaw rates and historical disturbance data.

[0047] Step b: Input the training data in the first training dataset into the initial perturbation recognition model for training, determine that the first preset training termination condition is met, and obtain the perturbation recognition model.

[0048] In specific implementation, a first training dataset and an initial disturbance recognition model are obtained, wherein the first training dataset includes historical four-wheel wheel speeds, historical vehicle accelerations, historical yaw rates, and historical disturbance data. The training data in the first training dataset is input into the initial disturbance recognition model for training, and a first preset training termination condition is determined to be met to obtain the disturbance recognition model.

[0049] The first preset training termination condition includes at least one of the following: determining that all data in the first training dataset has been input into the initial perturbation recognition model for training, determining that the loss function of the initial perturbation recognition model has converged to a first convergence threshold, or determining that the initial perturbation recognition model has been iteratively trained to a first preset number of iterations.

[0050] For example, the first preset training termination condition is to determine that all data in the first training dataset has been input into the initial perturbation recognition model for training: The first training dataset contains fifty sets of data, each set including historical four-wheel wheel speeds, historical vehicle acceleration, historical yaw rate, and historical disturbance data. The first preset training termination condition is determined when all data in the first training dataset has been input into the initial disturbance recognition model for training. That is, when all fifty sets of data have been input into the initial disturbance recognition model, there is no training data in the first training dataset that has not yet been input into the initial disturbance recognition model. At this point, the initial disturbance recognition model training is considered complete, and the disturbance recognition model is obtained.

[0051] In another example, the first preset training termination condition is to determine that the loss function of the initial perturbation recognition model converges to a first convergence threshold: The training data from the first training dataset is input into the initial perturbation recognition model for training, and the training results are output. A loss function is determined based on the training results and the actual perturbation data. The loss function may include at least one of the following: mean squared error loss function, cross-entropy loss function, logarithmic loss function, exponential loss function, squared loss function, or absolute value loss function, etc. When the loss function converges to a first convergence threshold, a first preset training termination condition is satisfied, and the perturbation recognition model is obtained.

[0052] In another example, the first preset training termination condition is to determine the initial perturbation recognition model to be iterated and trained to the first preset number of iterations.

[0053] The training data in the first training dataset is input into the initial perturbation recognition model for iterative training. The number of iterations is recorded. When the number of iterations is equal to the first preset number of iterations, the first preset training termination condition is met, and the perturbation recognition model is obtained.

[0054] In this embodiment, the disturbance recognition model is a neural network structure model. Preferably, the disturbance recognition model is a graph neural network-based model, used to model the asymmetric disturbance transmission structure between the four wheels caused by changes in adhesion, so as to achieve deep perception of the multi-wheel slip triggering characteristics.

[0055] Specifically, the input is a sample frame, each frame including the wheel speeds of the four wheels, vehicle acceleration, and lateral angular velocity. The time dimension window is fixed within 1 second, and is converted into a fixed-time sequence through sampling frequency. First, a dynamic topology graph between wheels is constructed, with the four wheels as nodes. The initial weights of the edges are calculated by the dynamic increment of the wheel speed difference between the wheels, and are updated at each time step as a disturbance correlation metric to capture the slip transmission path. The dynamic increment is calculated using first-order difference, representing the intensity of the instantaneous wheel speed change of the vehicle at the current moment. The disturbance correlation metric is expressed as the absolute value of the difference between the wheel speed change rates of the two wheels; the larger the disturbance correlation metric, the stronger the inconsistency in the slip response between the two wheels.

[0056] After graph construction, the input features undergo initial representation learning through a node encoder. Node features include the wheel's historical rate of change of speed, acceleration at adjacent time points, and driving torque input. The graph neural network model employs a multi-layer GCN structure, where each layer aggregates features from adjacent nodes to achieve the propagation and fusion of perturbation information between wheels, specifically modeling the dynamic inconsistencies between wheels exhibited by the vehicle during sudden changes in adhesion.

[0057] Specifically, the perturbation recognition model employs a three-layer GCN network structure, with the feature dimensions of nodes in each layer set to 32, 16, and 8, respectively. Each layer uses the ReLU activation function to enhance its ability to represent nonlinear wheel velocity perturbation patterns, and residual connections are introduced between adjacent GCN layers to avoid gradient vanishing. During graph convolution, the adjacency matrix is ​​processed using a symmetric normalization method based on the node degree matrix.

[0058] To enhance the sensitivity of edge weight updates, a graph attention mechanism is introduced to improve the model's ability to focus more on wheelset relationships where adhesion changes most drastically, thereby improving the modeling accuracy for asymmetric slip. Simultaneously, during training, the disturbance recognition model uses training labels jointly generated by manual annotation and self-supervised disturbance detection to ensure the model can identify sticky, slippery, and sticky disturbance structures on various road surfaces, including grass, moss, and wet sand. This model is deployed in the vehicle's local decision control unit, serving as a key structural input source for subsequent risk prediction and control mapping.

[0059] The target acceleration change rate is determined based on the vehicle acceleration, where the target acceleration change rate represents the instantaneous rate of change of vehicle acceleration. The target disturbance significance value, the vehicle slip rate, and the target acceleration change rate are input into a pre-trained slip level determination model, and processed by the slip level determination model to obtain the target slip level.

[0060] In this embodiment, the training process of the slip level determination model specifically includes: Step a, obtain the second training dataset and the initial slip level determination model, wherein the second training dataset includes historical perturbation significance values, historical vehicle slip rate, historical acceleration change rate and historical slip level.

[0061] Step b: Input the training data from the second training dataset into the initial slip level determination model for training, and determine that the second preset training termination condition is met to obtain the slip level determination model.

[0062] In specific implementation, a second training dataset and an initial slip level determination model are obtained. The second training dataset includes historical disturbance significance values, historical vehicle slip rates, historical acceleration change rates, and historical slip levels. The training data from the second training dataset is input into the initial slip level determination model for training. Once a second preset training termination condition is met, the slip level determination model is obtained.

[0063] The second preset training termination condition includes at least one of the following: determining that all data in the second training dataset has been input into the initial slip level determination model for training; determining that the loss function of the initial slip level determination model has converged to the second convergence threshold; or determining that the initial slip level determination model has been iterated for training up to the second preset number of iterations.

[0064] For example, the second preset training termination condition is to determine that all data in the second training dataset has been input into the initial slip level determination model for training: The second training dataset contains fifty sets of data, each set including historical disturbance significance values, historical vehicle slip rate, historical acceleration change rate, and historical slip level. The second preset training termination condition is that all data in the second training dataset has been input into the initial slip level determination model for training. That is, when all fifty sets of data have been input into the initial slip level determination model, there is no training data in the second training dataset that has not yet been input into the initial slip level determination model. At this point, the initial slip level determination model training is considered complete, and the slip level determination model is obtained.

[0065] In another example, the second preset training termination condition is to determine the model's loss function converges to a second convergence threshold after determining the initial slip level: The training data from the second training dataset is input into the initial slip level determination model for training, and the training results are output. A loss function is determined based on the training results and the actual slip level. The loss function may be of at least one of the following types: mean squared error loss function, cross-entropy loss function, logarithmic loss function, exponential loss function, squared loss function, or absolute value loss function, etc. When the loss function converges to a second convergence threshold, it is determined that the second preset training termination condition is met, and the slip level determination model is obtained.

[0066] In another example, the second preset training termination condition is to determine the initial slip level and then iterate the model to the second preset number of iterations.

[0067] The training data in the second training dataset is input into the initial slip level determination model for iterative training. The number of iterations is recorded. When the number of iterations is equal to the second preset number of iterations, the second preset training termination condition is met, and the slip level determination model is obtained.

[0068] In this embodiment, the slip level determination model is a slip risk level prediction model based on wheel speed perturbation characteristics. Specifically, the slip level determination model is a neural network structure model, which uses the gradient boosting decision tree (LightGBM) algorithm to quickly classify slip risks and output confidence scores during the transition between sticky, slippery, and sticky states.

[0069] Specifically, the input to the slip level determination model is a multi-dimensional time series sample, including the target disturbance significance value, the vehicle slip rate, and the target acceleration change rate. In this embodiment, the multi-dimensional time series sample may also include the differential characteristics of the four wheel speeds over a short period of time, the drive torque response delay, and the edge weight change information of the inter-wheel interaction output by the disturbance identification model.

[0070] After standardization, all features form a fixed-dimensional feature vector, which serves as the input space for the LightGBM model. The output is the target slip level, along with a probability score for each predicted level as a confidence metric. The target slip level is divided into five levels, from lowest to highest: Level 1, Level 2, Level 3, Level 4, and Level 5, corresponding to stable adhesion, mild slip, poisoned slip, severe slip, and unstable slip.

[0071] In this embodiment, during the training of the slip level determination model, the model utilizes real slip labeled samples from the automaker's self-developed dataset to learn the classification boundary through a supervised approach. The label is determined comprehensively based on the duration and degree of wheel speed deviating from the vehicle speed and whether it causes a loss of directional stability. Simultaneously, to improve the model's ability to distinguish nonlinear disturbance patterns, multiple second-order cross features are introduced, including a combination of the peak wheel speed change rate and the driving torque gradient, as well as the time delay difference between the initial side wheel and the subsequent spreading wheels.

[0072] Specifically, the training and validation sets are strictly stratified according to different terrain categories to ensure that the model's generalization ability spans typical conditions with low and rapidly changing adhesion, such as grassland, moss, and wet sand. The model evaluation metrics use weighted values ​​and confidence ranking stability, focusing on whether it can accurately judge slippage trends and respond in advance within a 0.5-second window of dramatic adhesion changes.

[0073] In some embodiments, when determining the first driving torque based on the target slip level and the target disturbance significance value, if both the target slip level and the target disturbance significance value indicate a high risk, then the first driving torque needs to be determined in conjunction with disturbance data. That is, determining the first driving torque based on the target slip level and the target disturbance significance value in step 1022 specifically includes: Step 10221: In response to the target disturbance significance value being greater than a preset significance threshold and the target slip level being less than a first preset level threshold, a first driving torque is determined based on the target disturbance significance value.

[0074] In practice, the target disturbance significance value is compared with a preset significance threshold, and the target slip level is compared with a first preset level threshold. If the target disturbance significance value is greater than the preset significance threshold and the target slip level is less than the first preset level threshold, it indicates that the disturbance identification model determines that the vehicle has a disturbance of the slip level, and the risk is inconsistent with the risk corresponding to the target slip level determined by the slip level determination model. In this case, the result of the disturbance identification model shall prevail, and the first driving torque shall be determined based on the target disturbance significance value.

[0075] Specifically, the process of determining the first driving torque based on the target disturbance significance value includes: Step A: Obtain the first initial torque corresponding to the target disturbance wheel, and determine the target adjustment coefficient based on the target disturbance significance value; Step B: Determine the first driving torque based on the first initial torque and the target adjustment coefficient.

[0076] In practice, the first initial torque corresponding to the target disturbance wheel is obtained, and the target adjustment coefficient is determined based on the target disturbance significance value. Specifically, the target adjustment coefficient corresponding to the target disturbance significance value can be determined by searching a database. The target adjustment coefficient is a coefficient greater than 0 and less than 1.

[0077] The first driving torque is determined based on the first initial torque and the target adjustment coefficient. The first initial torque and the target adjustment coefficient are then multiplied to obtain the first driving torque. Specifically, when the target disturbance significance value is greater than a preset significance threshold and the target slip level is less than a first preset level threshold, the driving torque of the target disturbance wheel determined by the disturbance identification model is reduced based on the target adjustment coefficient.

[0078] or, Step 10222: In response to the target disturbance significance value being greater than a preset significance threshold and the target slip level being greater than a first preset level threshold, a first driving torque is determined based on the target slip level, the target disturbance data, and the target torque change rate.

[0079] In specific implementation, if the target disturbance significance value is greater than the preset significance threshold and the target slip level is greater than the first preset level threshold, then the first driving torque is determined based on the target slip level, the target disturbance data and the target torque change rate.

[0080] Specifically, the method for determining the first driving torque includes: Step a: Obtain the pre-trained torque determination model; Step b: Input the target slip level, the target disturbance data, and the target torque change rate into the torque determination model, and process them through the torque determination model to obtain the first driving torque.

[0081] In specific implementation, a pre-trained torque determination model is obtained, and the target slip level, the target disturbance data, and the target torque change rate are input into the torque determination model. The first driving torque is obtained through processing by the torque determination model.

[0082] In this embodiment, the torque determination model is a neural network model. Preferably, the torque determination model is a model based on the Takagi-Sugeno fuzzy control architecture. That is, the torque determination model is a nonlinear control mapping model between slip risk and drive torque output based on the Takagi-Sugeno fuzzy control architecture, which is used to realize feedforward active torque adjustment based on the predicted risk level in complex low-adhesion road surfaces.

[0083] Specifically, the process of obtaining the first driving torque by processing the torque determination model includes: Step b01: Input the target slip level, the target disturbance data, and the target torque change rate into the torque determination model; Step b02, the torque determination model performs fuzzy processing on the target slip level, the target disturbance data and the target torque change rate to obtain a fuzzy semantic vector; Step b03: The torque determination model analyzes the fuzzy semantic vector to obtain and output the first driving torque.

[0084] In practice, the target torque change rate, vehicle longitudinal acceleration, target slip level output by the slip level determination model, and target disturbance data determined by the disturbance identification model are input into the torque determination model.

[0085] The torque determination model performs fuzzy processing on the target slip level, the target disturbance data, and the target torque change rate to obtain a fuzzy semantic vector. That is, through fuzzification, it maps to a fuzzy semantic set, where each variable corresponds to several linguistic descriptive terms, such as high risk, disturbance concentrated in the front wheels, and rapid torque growth. Each semantic term is equipped with a membership function curve, constructed using a combination of triangular and trapezoidal functions to ensure sensitive response to intermediate states and smooth transitions between boundary intervals.

[0086] In this embodiment, the torque determination model includes a fuzzy control rule base, which is jointly generated by an expert system and data-driven approach. This base contains dozens of control rules covering typical slip evolution paths and dynamic scenarios induced by sudden changes in adhesion. For example, when the slip level is high, the disturbance is concentrated on the rear wheels, and the longitudinal acceleration is high, the output torque suppression is strong. Alternatively, when the slip level is moderate and the disturbance spreads slowly, the output torque decreases slightly.

[0087] The torque determination model analyzes the fuzzy semantic vector to obtain and output the first driving torque. Specifically, it adopts a Takagi-Sugeno model structure, defining the output of each rule as a linear function. The final output is generated by a weighted linear combination of the activation degrees of each rule, exhibiting higher numerical continuity and response accuracy, making it suitable for deployment in vehicle electric drive controllers to perform fast dynamic responses. The obtained first driving torque serves as the feedforward input to the vehicle torque controller, working in conjunction with the underlying closed-loop controller to form a feedforward feedback combination structure.

[0088] In this embodiment, in order to improve real-time response efficiency and algorithm calculation stability, the torque determination model adopts a fixed-structure lookup table strategy to pre-compile key membership calculation. When deployed in an electronic control unit (ECU) or domain controller, it can achieve a 1ms-level response delay, thereby meeting the real-time control requirements of high-frequency slip scenarios.

[0089] In some embodiments, when selecting a target drive torque based on the target slip level from the initial drive torque and the first drive torque, the target slip level can be compared with a second preset level threshold, and the target drive torque can be selected based on the comparison result. That is, determining the target drive torque based on the target slip level, the initial drive torque, and the first drive torque in step 103 specifically includes: Step 1031: In response to the target slip level being less than a second preset level threshold, the initial driving torque is used as the target driving torque; or, Step 1032: In response to the slip level being greater than or equal to the second preset level threshold, the first driving torque is taken as the target driving torque.

[0090] In practice, the target slip level is compared with a second preset level threshold to obtain the comparison result. If the target slip level is less than the second preset level threshold, it indicates that the slip risk level is low. In this case, there is no need to adjust the drive torque, so the initial drive torque is directly used as the target drive torque.

[0091] If the target slip level is greater than or equal to the second preset level threshold, it indicates that the slip risk level is high. Therefore, the first drive torque determined based on the four wheel speeds, operating information and the target torque change rate is used as the target drive torque.

[0092] In this embodiment, the feedforward channel is the first driving torque determined in this embodiment, and the feedback channel is the initial driving torque determined by the original electric drive control strategy. When the system is identified as a low-risk state, the feedback channel dominates, that is, the target driving torque is determined as the initial driving torque. When the system is identified as a medium-to-high-risk state, the target driving torque is determined as the first driving torque, and the torque distribution strategy is adjusted according to the identified target disturbance wheel position, applying a higher proportion of control reduction at the position of the wheel with high slippage to avoid the spread of inter-wheel interference.

[0093] In this embodiment, to prevent control oscillations caused by high-frequency disturbances, a control signal smoother module is introduced. This module buffers the continuous control signal using a sliding window weighted average method, keeping torque output changes within the acceptable dynamic range of the physical system. Simultaneously, at the parameter decoupling level, a parameter switching mechanism driven by slip levels is constructed. Each level corresponds to an independent set of control parameters, including torque adjustment rate, control cycle, and response hysteresis threshold. The system can instantly switch parameter sets upon recognizing level changes, ensuring that the control strategy adapts to the state level in real time.

[0094] The above solution avoids the problem of insufficient adaptability of a single parameter under different attachment states, and improves the fault tolerance of the control system to varying road conditions. Meanwhile, the control structure for the first drive torque is deployed within the vehicle's electric drive domain controller, inserted as a feedforward control module into the existing control framework. It maintains a decoupled operation from the existing vehicle control logic, does not interfere with the existing feedback control link, and possesses good modular integration capabilities and engineering feasibility.

[0095] It should be noted that the method of this disclosure embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this disclosure embodiment, and the multiple devices will interact with each other to complete the method described.

[0096] It should be noted that the above description describes some embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] Based on the same inventive concept, corresponding to any of the above-described embodiments, this disclosure also provides a vehicle control device.

[0098] refer to Figure 2 , Figure 2 The vehicle control device, as described in this embodiment, includes: The data acquisition module 201 is configured to acquire the road surface adhesion coefficient of the road surface where the vehicle is located, determine that the road surface adhesion coefficient is less than a preset adhesion threshold, and acquire the wheel speed of the four wheels of the vehicle and the vehicle operation information. The grade determination module 202 is configured to acquire the target torque change rate corresponding to the vehicle drive wheels, and determine the first drive torque and the target slip grade based on the four wheel speeds, the operating information and the target torque change rate. The vehicle control module 203 is configured to acquire the initial drive torque of the vehicle's drive wheels, determine a target drive torque based on the target slip level, the initial drive torque, and the first drive torque, and control the vehicle's movement based on the target drive torque.

[0099] In some embodiments, the level determination module 202 is specifically configured as follows: The target slip level and target disturbance data are determined based on the wheel speeds of the four wheels and the operating information. The target disturbance data includes the target disturbance significance value, the target disturbance wheel, and the target disturbance propagation direction. The target disturbance wheel is the starting wheel of the vehicle slip propagation. The first driving torque is determined based on the target slip level and the target disturbance significance value.

[0100] In some embodiments, the operating information includes vehicle acceleration, vehicle yaw rate, and vehicle slip ratio, and the level determination module 202 is specifically configured as follows: The wheel speeds of the four wheels, the vehicle acceleration, and the vehicle yaw rate are input into a pre-trained disturbance recognition model, and the target disturbance data is obtained through processing by the disturbance recognition model. The target acceleration change rate is determined based on the vehicle acceleration. The target disturbance significance value, the vehicle slip rate, and the target acceleration change rate are input into a pre-trained slip level determination model. The target slip level is obtained by processing the slip level determination model.

[0101] In some embodiments, the level determination module 202 is specifically configured as follows: In response to the target disturbance significance value being greater than a preset significance threshold and the target slip level being less than a first preset level threshold, a first driving torque is determined based on the target disturbance significance value; or, In response to the target disturbance significance value being greater than a preset significance threshold and the target slip level being greater than a first preset level threshold, a first driving torque is determined based on the target slip level, the target disturbance data, and the target torque change rate.

[0102] In some embodiments, the level determination module 202 is specifically configured as follows: Obtain the first initial torque corresponding to the target disturbance wheel, and determine the target adjustment coefficient based on the target disturbance significance value; The first drive torque is determined based on the first initial torque and the target adjustment coefficient.

[0103] In some embodiments, the level determination module 202 is specifically configured as follows: Obtain the pre-trained torque determination model; The target slip level, the target disturbance data, and the target torque change rate are input into the torque determination model, and the first driving torque is obtained by processing the torque determination model.

[0104] In some embodiments, the level determination module 202 is specifically configured as follows: The target slip level, the target disturbance data, and the target torque change rate are input into the torque determination model; The torque determination model performs fuzzy processing on the target slip level, the target disturbance data, and the target torque change rate to obtain a fuzzy semantic vector; The torque determination model analyzes the fuzzy semantic vector to obtain and output the first driving torque.

[0105] In some embodiments, the vehicle control module 203 is specifically configured as follows: In response to the target slip level being less than a second preset level threshold, the initial drive torque is used as the target drive torque; or... In response to the slip level being greater than or equal to a second preset level threshold, the first driving torque is taken as the target driving torque.

[0106] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing this disclosure, the functions of each module can be implemented in one or more software and / or hardware.

[0107] The apparatus of the above embodiments is used to implement the corresponding vehicle control method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0108] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle control method described in any of the above embodiments.

[0109] Figure 3 This embodiment illustrates a more specific hardware structure of an electronic device. The device may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0110] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0111] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0112] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0113] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0114] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0115] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0116] The electronic devices described above are used to implement the corresponding vehicle control methods in any of the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0117] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to execute the vehicle control method as described in any of the above embodiments.

[0118] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0119] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the vehicle control method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0120] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a vehicle, including the vehicle control device, the electronic device, and the computer-readable storage medium in the above embodiments, wherein the vehicle device implements the vehicle control method described in any of the above embodiments.

[0121] The vehicles described in the above embodiments are used to implement the vehicle control method described in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0122] It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0123] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0124] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0125] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0126] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this disclosure (including the claims) is limited to these examples; within the framework of this disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this disclosure as described above, which are not provided in detail for the sake of brevity.

[0127] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this disclosure, the provided drawings may or may not show well-known power / ground connections to integrated circuit (IC) chips and other components. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this disclosure, and this also takes into account the fact that the details of implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this disclosure will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this disclosure, it will be apparent to those skilled in the art that the embodiments of this disclosure can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0128] Although this disclosure has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0129] This disclosure is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A vehicle control method, characterized in that, include: Obtain the road surface adhesion coefficient of the road surface where the vehicle is located, determine that the road surface adhesion coefficient is less than a preset adhesion threshold, and obtain the wheel speed of the four wheels of the vehicle and the vehicle operation information; Obtain the target torque change rate corresponding to the vehicle's drive wheels, and determine the first drive torque and target slip level based on the four wheel speeds, the operating information, and the target torque change rate. The initial drive torque of the vehicle's drive wheels is obtained, and a target drive torque is determined based on the target slip level, the initial drive torque, and the first drive torque. The vehicle's movement is controlled based on the target drive torque.

2. The method according to claim 1, characterized in that, The step of determining the first drive torque and target slip level based on the four wheel speeds, the operating information, and the target torque change rate includes: The target slip level and target disturbance data are determined based on the wheel speeds of the four wheels and the operating information. The target disturbance data includes the target disturbance significance value, the target disturbance wheel, and the target disturbance propagation direction. The target disturbance wheel is the starting wheel of the vehicle slip propagation. The first driving torque is determined based on the target slip level and the target disturbance significance value.

3. The method according to claim 2, characterized in that, The operational information includes vehicle acceleration, vehicle yaw rate, and vehicle slip ratio. The step of determining the target slip level and target disturbance data based on the four wheel speeds and the operating information includes: The wheel speeds of the four wheels, the vehicle acceleration, and the vehicle yaw rate are input into a pre-trained disturbance recognition model, and the target disturbance data is obtained through processing by the disturbance recognition model. The target acceleration change rate is determined based on the vehicle acceleration. The target disturbance significance value, the vehicle slip rate, and the target acceleration change rate are input into a pre-trained slip level determination model. The target slip level is obtained by processing the slip level determination model.

4. The method according to claim 2, characterized in that, The step of determining the first driving torque based on the target slip level and the target disturbance significance value includes: In response to the target disturbance significance value being greater than a preset significance threshold and the target slip level being less than a first preset level threshold, a first driving torque is determined based on the target disturbance significance value; or, In response to the target disturbance significance value being greater than a preset significance threshold and the target slip level being greater than a first preset level threshold, a first driving torque is determined based on the target slip level, the target disturbance data, and the target torque change rate.

5. The method according to claim 4, characterized in that, The step of determining the first driving torque based on the target disturbance significance value includes: Obtain the first initial torque corresponding to the target disturbance wheel, and determine the target adjustment coefficient based on the target disturbance significance value; The first drive torque is determined based on the first initial torque and the target adjustment coefficient.

6. The method according to claim 4, characterized in that, The step of determining the first driving torque based on the target slip level, the target disturbance data, and the target torque change rate includes: Obtain the pre-trained torque determination model; The target slip level, the target disturbance data, and the target torque change rate are input into the torque determination model, and the first driving torque is obtained by processing the torque determination model.

7. The method according to claim 6, characterized in that, The step of inputting the target slip level, the target disturbance data, and the target torque change rate into a pre-trained torque determination model, and processing them through the torque determination model to obtain the first driving torque, includes: The target slip level, the target disturbance data, and the target torque change rate are input into the torque determination model; The torque determination model performs fuzzy processing on the target slip level, the target disturbance data, and the target torque change rate to obtain a fuzzy semantic vector; The torque determination model analyzes the fuzzy semantic vector to obtain and output the first driving torque.

8. The method according to claim 1, characterized in that, The step of determining the target drive torque based on the target slip level, the initial drive torque, and the first drive torque includes: In response to the target slip level being less than a second preset level threshold, the initial drive torque is used as the target drive torque; or... In response to the slip level being greater than or equal to a second preset level threshold, the first driving torque is taken as the target driving torque.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as described in any one of claims 1 to 8.

10. A vehicle, characterized in that, The vehicle includes the electronic equipment as described in claim 9.

Images