Vehicle empennage control method, device and equipment based on road adhesion coefficient estimation

Abstract

The invention discloses a vehicle empennage control method, device and equipment based on road adhesion coefficient estimation, and the method comprises the steps: obtaining a driving state parameter of a target vehicle in a current time period, and determining adhesion coefficient time sequence data and slip rate time sequence data of the target vehicle; acquiring image information of a road in front of the target vehicle, and extracting a road texture feature matrix and a road color feature matrix; inputting the attachment coefficient time series data, the slip rate time series data, the road texture feature matrix and the road color feature matrix into a pre-trained road attachment coefficient estimation model to obtain a road attachment coefficient estimation value at the next moment; and according to the road adhesion coefficient estimation value, determining a corresponding optimal downforce, and according to the optimal downforce, adjusting and controlling the empennage of the target vehicle. The method improves the accuracy of road adhesion coefficient estimation and the accuracy of vehicle empennage control, thereby improving the safety and stability of vehicle driving, and can be applied to the technical field of vehicle control.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to a method, device and equipment for controlling a vehicle rear wing based on road adhesion coefficient estimation. Background Technology

[0002] With the rapid increase in car ownership, road traffic safety has become an increasingly important issue. To improve vehicle safety, many cars have been equipped with active safety control systems, which help drivers better control the vehicle, especially in complex and changing road conditions.

[0003] The coefficient of friction (COP) is a crucial indicator of road surface friction and directly impacts vehicle handling stability and safety. Adding an adjustable rear wing to a vehicle allows for adjustments to the wing angle under varying road conditions, altering aerodynamic performance and generating different downforce levels, thereby enhancing handling stability and safety.

[0004] Existing technologies still have many shortcomings in real-time estimation of road adhesion coefficient. First, methods relying on a single sensor or dynamic model are insufficient to cope with complex and ever-changing real-world road conditions, especially the uncertainties brought about by abrupt road surface changes. Second, existing methods lack an effective closed-loop control mechanism, making it impossible to dynamically adjust vehicle control parameters based on real-time estimation results to achieve optimal driving performance.

[0005] The above problems urgently need to be addressed. Summary of the Invention

[0006] The purpose of this invention is to at least partially solve one of the technical problems existing in the prior art.

[0007] Therefore, one objective of this invention is to provide a vehicle rear wing control method based on road adhesion coefficient estimation, which improves the accuracy of road adhesion coefficient estimation and vehicle rear wing control, thereby improving vehicle driving safety and stability.

[0008] Another objective of this invention is to provide a vehicle rear wing control device based on road adhesion coefficient estimation.

[0009] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of the present invention include: On one hand, embodiments of the present invention provide a vehicle rear wing control method based on road adhesion coefficient estimation, comprising the following steps: Obtain the driving status parameters of the target vehicle in the current time period, and determine the adhesion coefficient time series data and slip ratio time series data of the target vehicle based on the driving status parameters; Acquire road image information in front of the target vehicle, and extract road texture feature matrix and road color feature matrix based on the road image information; The adhesion coefficient time series data, the slip ratio time series data, the road texture feature matrix, and the road color feature matrix are input into a pre-trained road adhesion coefficient estimation model to obtain the estimated road adhesion coefficient value at the next time step. The optimal downforce is determined based on the road adhesion coefficient estimate, and the tail wing of the target vehicle is adjusted and controlled based on the optimal downforce.

[0010] Furthermore, in one embodiment of the present invention, the driving state parameters include wheel longitudinal force, wheel normal force, wheel center velocity, wheel angular velocity, and wheel rolling radius. The step of determining the adhesion coefficient time-series data and slip ratio time-series data of the target vehicle based on the driving state parameters specifically includes: The real-time adhesion coefficient of the target vehicle at each moment in the current time period is calculated based on the longitudinal force and normal force of the wheel. The real-time adhesion coefficient is then processed into a time series to obtain the adhesion coefficient time series data. The real-time slip rate of the target vehicle at various moments in the current time period is calculated based on the wheel center velocity, the wheel angular velocity, and the wheel rolling radius. The real-time slip rate is then processed into a time series to obtain the slip rate time series data.

[0011] Furthermore, in one embodiment of the present invention, the road texture feature matrix is ​​extracted through the following steps: The road image information is processed to obtain a grayscale road image; The road grayscale image is segmented according to a preset pixel window to obtain multiple target image regions; Determine the gray-level co-occurrence matrix of each target image region, and calculate the energy value, entropy value, moment of inertia value, and local stationary value of each target image region based on the gray-level co-occurrence matrix to obtain the texture feature parameters of each target image region; The texture feature parameters of each target image region are matrixed to obtain the road texture feature matrix.

[0012] Furthermore, in one embodiment of the present invention, the road color feature matrix is ​​extracted through the following steps: Determine the RGB values ​​of each pixel in the road image information; The hue, saturation, and brightness of each pixel are calculated based on the RGB values ​​to obtain the color feature parameters of each pixel. The color feature parameters of each pixel are matrixed to obtain the road color feature matrix.

[0013] Furthermore, in one embodiment of the present invention, the road adhesion coefficient estimation model is trained through the following steps: Acquire time-series sample data of adhesion coefficient, slip ratio, road texture sample feature matrix, and road color sample feature matrix of the test vehicle in the test scenario; Training samples are constructed based on the adhesion coefficient time-series sample data, the slip ratio time-series sample data, the road texture sample feature matrix, and the road color sample feature matrix, and the adhesion coefficient label corresponding to the next moment of the training sample is determined by manual annotation. The training samples are input into a pre-built CNN-LSTM hybrid neural network for training to obtain the trained road adhesion coefficient estimation model.

[0014] Furthermore, in one embodiment of the present invention, the CNN-LSTM hybrid neural network includes a CNN convolutional layer, an LSTM layer, a feature fusion layer, and an output layer. The step of inputting the training samples into the pre-constructed CNN-LSTM hybrid neural network for training to obtain the trained road adhesion coefficient estimation model specifically includes: The road texture sample feature matrix and the road color sample feature matrix are convolved by two CNN convolutional layers respectively to obtain texture space feature vector and color space feature vector; Hidden state calculations are performed on the adhesion coefficient time-series sample data and the slip ratio time-series sample data using two LSTM layers respectively, to obtain the adhesion coefficient time-series feature vector and the slip ratio time-series feature vector. The feature fusion layer performs spatiotemporal fusion of the texture space feature vector, the color space feature vector, the adhesion coefficient temporal feature vector, and the slip ratio temporal feature vector based on an attention mechanism to obtain spatiotemporal fused features. The spatiotemporal fusion features are mapped to the predicted adhesion coefficient value for the next time step through the output layer; The loss value is determined based on the predicted adhesion coefficient value and the adhesion coefficient label; The parameters of the CNN-LSTM hybrid neural network are updated using the backpropagation algorithm based on the loss value to obtain the trained road adhesion coefficient estimation model.

[0015] Furthermore, in one embodiment of the present invention, the step of determining the corresponding optimal downforce based on the road adhesion coefficient estimate, and adjusting and controlling the tail wing of the target vehicle based on the optimal downforce, specifically includes: The optimal downforce is obtained by querying a pre-built adhesion coefficient-downforce mapping table based on the estimated road adhesion coefficient value. Obtain the current sprung load of the target vehicle, and determine the downforce deviation value based on the optimal downforce and the current sprung load; The angle of the tail wing of the target vehicle is adjusted according to the downforce deviation value so that the sprung load of the target vehicle after adjustment is equal to the optimal downforce.

[0016] On the other hand, embodiments of the present invention provide a vehicle rear wing control device based on road adhesion coefficient estimation, comprising: The parameter acquisition module is used to acquire the driving status parameters of the target vehicle in the current time period, and determine the adhesion coefficient time series data and slip ratio time series data of the target vehicle based on the driving status parameters. The image feature extraction module is used to acquire road image information in front of the target vehicle, and extract road texture feature matrix and road color feature matrix based on the road image information; The model estimation module is used to input the adhesion coefficient time series data, the slip ratio time series data, the road texture feature matrix and the road color feature matrix into a pre-trained road adhesion coefficient estimation model to obtain the estimated road adhesion coefficient value at the next time step. The rear wing adjustment control module is used to determine the corresponding optimal downforce based on the road adhesion coefficient estimate, and to adjust the rear wing of the target vehicle based on the optimal downforce.

[0017] On the other hand, embodiments of the present invention provide an electronic device, including: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above-described vehicle rear wing control method based on road adhesion coefficient estimation.

[0018] On the other hand, embodiments of the present invention also provide a computer-readable storage medium storing a processor-executable computer program that, when executed by a processor, implements the above-described vehicle rear wing control method based on road adhesion coefficient estimation.

[0019] On the other hand, embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the above-described vehicle rear wing control method based on road adhesion coefficient estimation.

[0020] The advantages and beneficial effects of the present invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention: This invention acquires the driving state parameters of a target vehicle in the current time period, determines the time-series data of the vehicle's adhesion coefficient and slip ratio based on these parameters, acquires road image information in front of the target vehicle, extracts a road texture feature matrix and a road color feature matrix from the road image information, and inputs these data into a pre-trained road adhesion coefficient estimation model to obtain the estimated road adhesion coefficient for the next time moment. Based on this estimated value, the corresponding optimal downforce is determined, and the tail wing of the target vehicle is adjusted and controlled accordingly. This invention determines the time-series data of the adhesion coefficient and slip ratio based on the vehicle's driving state parameters, extracts the road texture feature matrix and road color feature matrix from the road image information in front of the vehicle, and inputs these into a pre-trained road adhesion coefficient estimation model to obtain the estimated road adhesion coefficient for the next time moment. Furthermore, the optimal downforce is determined based on this estimated value, and the tail wing is dynamically adjusted, improving the accuracy of road adhesion coefficient estimation and tail wing control, thereby enhancing vehicle safety and stability. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the embodiments of the present invention are described below. It should be understood that the drawings described below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the steps of a vehicle rear wing control method based on road adhesion coefficient estimation, provided in an embodiment of the present invention; Figure 2 A structural block diagram of a vehicle rear wing control device based on road adhesion coefficient estimation is provided in an embodiment of the present invention; Figure 3 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of this invention; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this invention as detailed in the appended claims.

[0024] 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 invention pertains. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0025] The vehicle rear wing control method based on road adhesion coefficient estimation provided in this invention can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the vehicle rear wing control method based on road adhesion coefficient estimation, but is not limited to the above forms.

[0026] This invention can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0027] It should be noted that in various specific embodiments of the present invention, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of the present invention require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to a confirmation page. Only after obtaining the user's separate permission or consent is the necessary user-related data for the normal operation of the embodiments of the present invention acquired.

[0028] The existing technology has the following problems: 1) Inaccurate real-time estimation of road adhesion coefficient: Existing technologies often rely on a single sensor or dynamic model when estimating road adhesion coefficient, which is difficult to cope with complex and ever-changing actual road conditions, resulting in large errors in the estimation results, especially when there are sudden changes in the road surface, and cannot provide reliable predictions.

[0029] 2) Lack of dynamic adjustment mechanism for rear wing downforce control: Existing vehicle rear wing control methods are mostly statically set, lacking the ability to dynamically adjust according to real-time road conditions. This results in the vehicle being unable to maintain optimal downforce when driving on surfaces with different coefficients of adhesion, affecting driving stability and safety.

[0030] 3) Insufficient fusion of multi-source sensor data: Existing technologies fail to fully utilize the advantages of various sensors when processing multi-source sensor data, resulting in insufficient data fusion, which affects the accuracy and robustness of road adhesion coefficient estimation.

[0031] This invention primarily aims to estimate the road adhesion coefficient in real time based on multi-source sensor information, and optimize the downforce control of the vehicle's rear wing accordingly to improve vehicle stability and safety under different road conditions. For example, in rainy or snowy weather or on slippery roads, this solution can adjust the rear wing angle in a timely manner to provide optimal downforce, ensuring the vehicle maintains best grip and handling performance. Furthermore, this solution can continuously monitor and adjust the rear wing angle through a closed-loop control mechanism to adapt to constantly changing road conditions, thereby significantly improving the driving experience and road safety.

[0032] Reference Figure 1 This invention provides a vehicle rear wing control method based on road adhesion coefficient estimation, specifically including the following steps: S101. Obtain the driving status parameters of the target vehicle in the current time period, and determine the adhesion coefficient time series data and slip ratio time series data of the target vehicle based on the driving status parameters. S102. Obtain road image information in front of the target vehicle, and extract the road texture feature matrix and road color feature matrix based on the road image information; S103. Input the time series data of adhesion coefficient, the time series data of slip ratio, the road texture feature matrix and the road color feature matrix into the pre-trained road adhesion coefficient estimation model to obtain the estimated value of road adhesion coefficient at the next time step. S104. Determine the corresponding optimal downforce based on the estimated road adhesion coefficient, and adjust and control the tail wing of the target vehicle based on the optimal downforce.

[0033] This invention utilizes multi-source sensor data to determine the time-series data of adhesion coefficient, slip ratio, road texture feature matrix, and road color feature matrix. A neural network algorithm is used to estimate the road adhesion coefficient in real time. The estimated road adhesion coefficient is periodically input into the control unit of the adjustable rear wing, thereby realizing the dynamic adjustment of the rear wing downforce. This ensures that the vehicle maintains the best driving state on road surfaces with different adhesion conditions, significantly improving the driving experience and driving safety.

[0034] This invention determines the time-series data of adhesion coefficient and slip ratio based on the vehicle's driving state parameters, extracts the road texture feature matrix and road color feature matrix based on the road image information in front of the vehicle, and inputs them into a pre-trained road adhesion coefficient estimation model to obtain the estimated value of the road adhesion coefficient at the next moment. Then, based on the estimated value of the road adhesion coefficient, the corresponding optimal downforce is determined and the tail wing is dynamically adjusted, which improves the accuracy of road adhesion coefficient estimation and the accuracy of vehicle tail wing control, thereby improving the safety and stability of vehicle driving.

[0035] Specifically, embodiments of the present invention utilize multiple sensors such as visual sensors, accelerometers, and gyroscopes to collect current vehicle driving status parameters and road image information of the road surface ahead, then determine the adhesion coefficient time series data and slip ratio time series data, and extract the road texture feature matrix and road color feature matrix.

[0036] As a further optional implementation, the driving state parameters include wheel longitudinal force, wheel normal force, wheel center velocity, wheel angular velocity, and wheel rolling radius. Based on these driving state parameters, the adhesion coefficient time-series data and slip ratio time-series data of the target vehicle are determined, specifically including: S1011. Calculate the real-time adhesion coefficient of the target vehicle at each moment in the current time period based on the longitudinal force and normal force of the wheel, and perform time-series processing on the real-time adhesion coefficient to obtain the time-series data of the adhesion coefficient. S1012. Calculate the real-time slip ratio of the target vehicle at each moment in the current time period based on the wheel center speed, wheel angular velocity and wheel rolling radius, and perform time-series processing on the real-time slip ratio to obtain slip ratio time-series data.

[0037] Specifically, the driving state parameters include the longitudinal force F of the wheels. X Normal force F Z The wheel center velocity v, wheel angular velocity w, and wheel rolling radius r are used to calculate the real-time adhesion coefficient based on the wheel longitudinal force and wheel normal force. The real-time slip ratio is calculated based on the wheel center velocity, wheel angular velocity, and wheel rolling radius. This ultimately generates time-series data on adhesion coefficient and slip ratio.

[0038] As an optional further implementation, the road texture feature matrix is ​​extracted through the following steps: S1021. Perform grayscale processing on the road image information to obtain a road grayscale image; S1022. The road grayscale image is segmented according to a preset pixel window to obtain multiple target image regions; S1023. Determine the gray-level co-occurrence matrix of each target image region, and calculate the energy value, entropy value, moment of inertia value and local stationary value of each target image region based on the gray-level co-occurrence matrix to obtain the texture feature parameters of each target image region. S1024. Perform matrix processing on the texture feature parameters of each target image region to obtain the road texture feature matrix.

[0039] Specifically, image features of road image information are extracted and a feature matrix is ​​formed. The image features include texture features (energy E, entropy H, moment of inertia J, local stationarity L) and color features (hue T, saturation S, brightness B).

[0040] In texture feature extraction, energy (E), entropy (H), moment of inertia (J), and local stationarity (L) are commonly used statistical features extracted based on the gray-level co-occurrence matrix (GLCM), which respectively reflect different properties of image texture.

[0041] First, the road image information is converted into a grayscale image. Then, it is segmented according to a preset pixel window size (e.g., a 10×10 pixel window) to obtain multiple target image regions. For each target image region, a grayscale co-occurrence matrix in four directions is calculated, and the average value is taken to reduce directional sensitivity. Based on the grayscale co-occurrence matrix of each target image region, the energy E, entropy H, moment of inertia J, and local stationarity L are calculated to form the texture feature parameters {E, H, J, L} of each target image region. Based on the spatial position relationship of the image, the texture feature parameters of each target image region are matrixed to finally obtain the road texture feature matrix.

[0042] Specifically, energy, also known as the angular second moment, is the sum of the squares of all elements in the gray-level co-occurrence matrix. Its calculation formula is as follows:

[0043] in, The gray values ​​in the gray-level co-occurrence matrix are and The probability of pixel pairs appearing. It refers to the grayscale level. Energy reflects the uniformity of image texture. The higher the energy value, the more uniform the grayscale distribution in the image and the more regular the texture; conversely, the lower the energy value, the more complex and irregular the texture.

[0044] Entropy is a concept in information theory used to measure the complexity or randomness of image texture. Its calculation formula is:

[0045] The higher the entropy value, the greater the complexity of the image texture, the more uneven the gray-level distribution, and the more information it contains; conversely, the lower the entropy value, the simpler and more regular the texture.

[0046] The moment of inertia, also known as the contrast component, reflects the magnitude of the difference in grayscale values ​​between adjacent pixels in an image texture. Its calculation formula is:

[0047] A larger moment of inertia value indicates a greater difference in grayscale between adjacent pixels in the image, higher texture contrast, and richer details; conversely, a smaller moment of inertia value indicates a smoother texture and smaller grayscale variations.

[0048] Local stationarity (sometimes also called homogeneity) reflects the concentration of pixel pairs with similar gray values ​​in an image texture, and its calculation formula is as follows:

[0049] A higher local stationarity value indicates that there are more pairs of pixels with similar gray values ​​in the image, and the texture is smoother and more uniform; conversely, a lower local stationarity value indicates that the gray values ​​in the texture vary more and the non-uniformity is higher.

[0050] As an optional implementation, the road color feature matrix is ​​extracted through the following steps: S1025. Determine the RGB values ​​of each pixel in the road image information; S1026. Calculate the hue, saturation and brightness of each pixel based on the RGB values ​​to obtain the color feature parameters of each pixel. S1027. Perform matrix processing on the color feature parameters of each pixel to obtain the road color feature matrix.

[0051] Specifically, the hue (T), saturation (S), and brightness (B) of a pixel need to be converted and calculated from RGB values. The core is to first normalize the RGB components and then derive them through the maximum / minimum values ​​and the difference.

[0052] The RGB values ​​of the pixels (range 0-255) are first normalized to the 0-1 interval, and then H, S, and B are calculated. The key parameter is: C. max =max(R, G, B); C min =min(R, G, B); Δ= C max -C min .

[0053] The hue calculation process is as follows: If C max =R, then T=60°×[(GB) / Δmod 6]; if C max =G, then T=60°×[(BR) / Δ+ 2]; if C max =B, then T=60°×[(RG) / Δ+ 4].

[0054] The saturation calculation process is as follows: If C max If Δ = 0, then S = 0; otherwise, S = Δ / C max .

[0055] The brightness calculation process is: B=C max .

[0056] After calculating the hue (T), saturation (S), and brightness (B) of each pixel, the color feature parameters {T, S, B} of each pixel are formed. Based on the spatial position relationship of the image, the dark feature parameters of each pixel are matrixed to finally obtain the road color feature matrix.

[0057] As an optional implementation, the road adhesion coefficient estimation model is trained through the following steps: S201. Obtain the time-series sample data of the adhesion coefficient, slip ratio, road texture sample feature matrix, and road color sample feature matrix of the test vehicle in the test scenario. S202. Construct training samples based on the time series sample data of adhesion coefficient, the time series sample data of slip ratio, the feature matrix of road texture sample, and the feature matrix of road color sample, and determine the adhesion coefficient label of the next time step corresponding to the training sample through manual annotation. S203. Input the training samples into the pre-built CNN-LSTM hybrid neural network for training to obtain a trained road adhesion coefficient estimation model.

[0058] As a further optional implementation, the CNN-LSTM hybrid neural network includes CNN convolutional layers, LSTM layers, feature fusion layers, and an output layer. Training samples are input into the pre-constructed CNN-LSTM hybrid neural network for training to obtain a trained road adhesion coefficient estimation model, which specifically includes: S2031. The feature matrix of road texture samples and the feature matrix of road color samples are convolved by two CNN convolutional layers respectively to obtain the texture space feature vector and the color space feature vector. S2032. Hidden state calculations are performed on the time series sample data of adhesion coefficient and slip ratio using two LSTM layers respectively, to obtain the time series feature vectors of adhesion coefficient and slip ratio. S2033. Spatiotemporal fusion of texture space feature vector, color space feature vector, adhesion coefficient temporal feature vector and slip rate temporal feature vector is performed by the feature fusion layer based on the attention mechanism to obtain spatiotemporal fused features. S2034. The spatiotemporal fusion features are mapped to the predicted adhesion coefficient value at the next time step through the output layer; S2035. Determine the loss value based on the predicted value of the adhesion coefficient and the adhesion coefficient label; S2036. Update the parameters of the CNN-LSTM hybrid neural network based on the loss value using the backpropagation algorithm to obtain the trained road adhesion coefficient estimation model.

[0059] Specifically, multi-source sensor data of the test vehicle in the test scenario is acquired. Based on similar steps as before, time-series sample data of adhesion coefficient, slip ratio, road texture sample feature matrix, and road color sample feature matrix are determined. Training samples are constructed based on these data, while the actual adhesion coefficient at the next moment can be measured during the test. Adhesion coefficient labels are formed through manual annotation. The training samples are input into a pre-constructed CNN-LSTM hybrid neural network. Two CNN convolutional layers are used to convolve the road texture sample feature matrix and the road color sample feature matrix to obtain texture space feature vectors and color space feature vectors, respectively. Two LSTM layers are then used to process the adhesion coefficient... Hidden state calculations are performed on temporal sample data and slip ratio temporal sample data to obtain temporal feature vectors for adhesion coefficient and slip ratio. A feature fusion layer, based on an attention mechanism, performs spatiotemporal fusion of texture space feature vectors, color space feature vectors, adhesion coefficient temporal feature vectors, and slip ratio temporal feature vectors to obtain spatiotemporal fused features. Finally, the output layer maps these spatiotemporal fused features to the predicted adhesion coefficient value for the next time step. Based on the predicted adhesion coefficient value and the adhesion coefficient label, a loss value is determined using a preset loss function. The parameters of the CNN-LSTM hybrid neural network are updated using the backpropagation algorithm based on this loss value, completing one round of iterative training. Training stops when the number of iterations reaches a preset threshold or the loss value reaches a preset loss value threshold, resulting in a well-trained road adhesion coefficient estimation model.

[0060] By inputting the time series data of the target vehicle's adhesion coefficient, slip ratio, road texture feature matrix, and road color feature matrix into the road adhesion coefficient estimation model, the estimated road adhesion coefficient for the next time step can be obtained from the model output.

[0061] As a further optional implementation, the optimal downforce is determined based on the estimated road adhesion coefficient, and the rear wing of the target vehicle is adjusted and controlled according to the optimal downforce, specifically including: S1041. Based on the estimated road adhesion coefficient, query the pre-built adhesion coefficient-downforce mapping table to obtain the optimal downforce; S1042. Obtain the current sprung load of the target vehicle, and determine the downforce deviation value based on the optimal downforce and the current sprung load; S1043. Adjust the angle of the target vehicle's tail wing according to the downforce deviation value so that the sprung load of the target vehicle after adjustment is equal to the optimal downforce.

[0062] Specifically, the estimated road adhesion coefficient is periodically input into the control unit of the adjustable tail wing. A two-dimensional lookup table is performed based on the road adhesion coefficient-optimal downforce MAP (i.e., adhesion coefficient-downforce mapping table) to obtain the optimal downforce corresponding to the estimated road adhesion coefficient. The control unit adjusts the tail wing in real time based on the optimal downforce to ensure that the tail wing angle is always in the ideal position.

[0063] Specifically, the sprung load is collected in real time, and the difference between it and the optimal downforce is calculated to obtain the downforce deviation value. Based on this downforce deviation value, the angle of the target vehicle's rear wing is adjusted so that the sprung load of the target vehicle after adjustment equals the optimal downforce. It is evident that the closed-loop control mechanism ensures that the tail wing angle remains in the ideal position under constantly changing road conditions, thereby achieving dynamic adjustment of optimal downforce.

[0064] It is understood that the embodiments of the present invention estimate the road adhesion coefficient in real time based on multi-source sensor data, and combine the dynamic adjustment of the tail wing downforce and the closed-loop control mechanism to ensure that the vehicle maintains the best driving state on road surfaces with different adhesion conditions, which significantly improves the driving experience and driving safety.

[0065] The method steps of the embodiments of the present invention have been described above. It can be understood that the embodiments of the present invention determine the time-series data of the adhesion coefficient and the slip ratio based on the vehicle's driving state parameters, extract the road texture feature matrix and the road color feature matrix based on the road image information in front of the vehicle, and input them into a pre-trained road adhesion coefficient estimation model to obtain the estimated value of the road adhesion coefficient at the next moment. Then, based on the estimated value of the road adhesion coefficient, the corresponding optimal downforce is determined and the rear wing is dynamically adjusted, improving the accuracy of the road adhesion coefficient estimation and the accuracy of the vehicle's rear wing control, thereby improving the safety and stability of vehicle driving.

[0066] Reference Figure 2 This invention provides a vehicle rear wing control device based on road adhesion coefficient estimation, comprising: The parameter acquisition module is used to acquire the driving status parameters of the target vehicle in the current time period, and determine the time series data of the adhesion coefficient and slip ratio of the target vehicle based on the driving status parameters. The image feature extraction module is used to acquire road image information in front of the target vehicle, and extract the road texture feature matrix and road color feature matrix based on the road image information; The model estimation module is used to input the time series data of adhesion coefficient, the time series data of slip ratio, the road texture feature matrix, and the road color feature matrix into the pre-trained road adhesion coefficient estimation model to obtain the estimated value of the road adhesion coefficient at the next time step. The rear wing adjustment control module is used to determine the corresponding optimal downforce based on the road adhesion coefficient estimate, and to adjust the rear wing of the target vehicle according to the optimal downforce.

[0067] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0068] Reference Figure 3 This invention provides an electronic device, comprising: At least one processor; At least one memory for storing at least one program; When the above-mentioned at least one program is executed by the above-mentioned at least one processor, the above-mentioned at least one processor implements the above-mentioned vehicle rear wing control method based on road adhesion coefficient estimation.

[0069] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0070] This invention also provides a computer-readable storage medium storing a processor-executable computer program that, when executed by a processor, implements the above-described vehicle rear wing control method based on road adhesion coefficient estimation.

[0071] This invention provides a computer-readable storage medium that can execute a vehicle rear wing control method based on road adhesion coefficient estimation provided in the method embodiments of this invention. It can execute any combination of implementation steps of the method embodiments and has the corresponding functions and beneficial effects of the method.

[0072] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described vehicle rear wing control method based on road adhesion coefficient estimation.

[0073] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0074] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0075] The embodiments described in this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.

[0076] The terms "first," "second," "third," "fourth," etc. (if present) in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0077] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the aforementioned blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0078] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the aforementioned functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0079] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, 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 steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0080] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0081] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the aforementioned program can be printed, because the aforementioned program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0082] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0083] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0084] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0085] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A vehicle rear wing control method based on road adhesion coefficient estimation, characterized in that, Includes the following steps: Obtain the driving status parameters of the target vehicle in the current time period, and determine the adhesion coefficient time series data and slip ratio time series data of the target vehicle based on the driving status parameters; Acquire road image information in front of the target vehicle, and extract road texture feature matrix and road color feature matrix based on the road image information; The adhesion coefficient time series data, the slip ratio time series data, the road texture feature matrix, and the road color feature matrix are input into a pre-trained road adhesion coefficient estimation model to obtain the estimated road adhesion coefficient value at the next time step. The optimal downforce is determined based on the road adhesion coefficient estimate, and the tail wing of the target vehicle is adjusted and controlled based on the optimal downforce.

2. The vehicle rear wing control method based on road adhesion coefficient estimation according to claim 1, characterized in that, The driving state parameters include wheel longitudinal force, wheel normal force, wheel center velocity, wheel angular velocity, and wheel rolling radius. The determination of the adhesion coefficient time-series data and slip ratio time-series data of the target vehicle based on the driving state parameters specifically includes: The real-time adhesion coefficient of the target vehicle at each moment in the current time period is calculated based on the longitudinal force and normal force of the wheel. The real-time adhesion coefficient is then processed into a time series to obtain the adhesion coefficient time series data. The real-time slip rate of the target vehicle at various moments in the current time period is calculated based on the wheel center velocity, the wheel angular velocity, and the wheel rolling radius. The real-time slip rate is then processed into a time series to obtain the slip rate time series data.

3. The vehicle rear wing control method based on road adhesion coefficient estimation according to claim 1, characterized in that, The road texture feature matrix is ​​extracted through the following steps: The road image information is processed into grayscale to obtain a road grayscale image; The road grayscale image is segmented according to a preset pixel window to obtain multiple target image regions; Determine the gray-level co-occurrence matrix of each target image region, and calculate the energy value, entropy value, moment of inertia value, and local stationary value of each target image region based on the gray-level co-occurrence matrix to obtain the texture feature parameters of each target image region; The texture feature parameters of each target image region are matrixed to obtain the road texture feature matrix.

4. The vehicle rear wing control method based on road adhesion coefficient estimation according to claim 1, characterized in that, The road color feature matrix is ​​extracted through the following steps: Determine the RGB values ​​of each pixel in the road image information; The hue, saturation, and brightness of each pixel are calculated based on the RGB values ​​to obtain the color feature parameters of each pixel. The color feature parameters of each pixel are matrixed to obtain the road color feature matrix.

5. The vehicle rear wing control method based on road adhesion coefficient estimation according to claim 1, characterized in that, The road adhesion coefficient estimation model is trained through the following steps: Acquire time-series sample data of adhesion coefficient, slip ratio, road texture sample feature matrix, and road color sample feature matrix of the test vehicle in the test scenario; Training samples are constructed based on the adhesion coefficient time-series sample data, the slip ratio time-series sample data, the road texture sample feature matrix, and the road color sample feature matrix, and the adhesion coefficient label corresponding to the next moment of the training sample is determined by manual annotation. The training samples are input into a pre-built CNN-LSTM hybrid neural network for training to obtain the trained road adhesion coefficient estimation model.

6. The vehicle rear wing control method based on road adhesion coefficient estimation according to claim 5, characterized in that, The CNN-LSTM hybrid neural network includes CNN convolutional layers, LSTM layers, feature fusion layers, and an output layer. The step of inputting the training samples into the pre-constructed CNN-LSTM hybrid neural network for training to obtain the trained road adhesion coefficient estimation model specifically includes: The road texture sample feature matrix and the road color sample feature matrix are convolved by two CNN convolutional layers respectively to obtain texture space feature vector and color space feature vector; Hidden state calculations are performed on the adhesion coefficient time-series sample data and the slip ratio time-series sample data using two LSTM layers respectively, to obtain the adhesion coefficient time-series feature vector and the slip ratio time-series feature vector. The feature fusion layer performs spatiotemporal fusion of the texture space feature vector, the color space feature vector, the adhesion coefficient temporal feature vector, and the slip ratio temporal feature vector based on an attention mechanism to obtain spatiotemporal fused features. The spatiotemporal fusion features are mapped to the predicted adhesion coefficient value for the next time step through the output layer; The loss value is determined based on the predicted adhesion coefficient value and the adhesion coefficient label; The parameters of the CNN-LSTM hybrid neural network are updated using the backpropagation algorithm based on the loss value to obtain the trained road adhesion coefficient estimation model.

7. A vehicle rear wing control method based on road adhesion coefficient estimation according to any one of claims 1 to 6, characterized in that, The step of determining the optimal downforce based on the estimated road adhesion coefficient and adjusting the tail wing of the target vehicle based on the optimal downforce specifically includes: The optimal downforce is obtained by querying a pre-built adhesion coefficient-downforce mapping table based on the estimated road adhesion coefficient value. Obtain the current sprung load of the target vehicle, and determine the downforce deviation value based on the optimal downforce and the current sprung load; The angle of the tail wing of the target vehicle is adjusted according to the downforce deviation value so that the sprung load of the target vehicle after adjustment is equal to the optimal downforce.

8. A vehicle rear wing control device based on road adhesion coefficient estimation, characterized in that, include: The parameter acquisition module is used to acquire the driving status parameters of the target vehicle in the current time period, and determine the adhesion coefficient time series data and slip ratio time series data of the target vehicle based on the driving status parameters. The image feature extraction module is used to acquire road image information in front of the target vehicle, and extract road texture feature matrix and road color feature matrix based on the road image information; The model estimation module is used to input the adhesion coefficient time series data, the slip ratio time series data, the road texture feature matrix and the road color feature matrix into a pre-trained road adhesion coefficient estimation model to obtain the estimated road adhesion coefficient value at the next time step. The rear wing adjustment control module is used to determine the corresponding optimal downforce based on the road adhesion coefficient estimate, and to adjust the rear wing of the target vehicle based on the optimal downforce.

9. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a vehicle rear wing control method based on road adhesion coefficient estimation as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements a vehicle rear wing control method based on road adhesion coefficient estimation as described in any one of claims 1 to 7.

Images