Self-adaptive adjusting method, device and equipment for vehicle seat and medium

Abstract

The invention provides a vehicle seat self-adaptive adjusting method and device, equipment and a medium, and the method comprises the steps: obtaining current driving scene information, inputting the current driving scene information into a seat adjusting self-adaptive model, so as to obtain a seat adjusting parameter corresponding to the current driving scene information, the seat adjustment adaptive model obtaining method comprises the steps of constructing an initial neural network model, training the initial neural network model based on historical driving scene information to obtain the seat adjustment adaptive model, and finally adjusting a vehicle seat based on seat parameters corresponding to current driving scene information. The self-adaptive adjustment of the vehicle seat is realized; according to the method, the association relationship between the driving scene information and the seat adjustment parameters is established, so that the seat adjustment parameters strongly related to the current scene information are obtained based on the current scene information, seat adjustment is more suitable for the driving environment, and the driving safety and comfort are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and in particular to a vehicle seat adaptive adjustment method, device, equipment and medium. Background Art

[0002] With the development of science and technology, cars have become an indispensable part of people's daily life. The safety and comfort of cars are also attracting more and more attention. Among them, the driver's field of vision is one of the important factors affecting car safety. However, due to the design of the car structure, the driver may encounter some visual blind spots during driving, which may affect driving safety. In order to solve this problem, seat adjustment technology came into being. By adjusting the height and angle of the seat, the driver's field of vision can be changed, thereby reducing the visual blind spots and improving driving safety.

[0003] Existing seat adjustment technologies mainly adjust the height and angle of the seat manually or electrically. The manual adjustment method requires the driver to adjust it according to their needs, which is inconvenient to operate, while the electric adjustment method can be adjusted through buttons or knobs, which is relatively convenient to operate. However, whether it is manual adjustment or automatic adjustment, the seat mode is relatively fixed, and the adaptability of the seat adjustment mode to the real-time environment is low. In certain environments, it is easy to cause the driver to have a large blind spot, thus causing safety hazards. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a vehicle seat adaptive adjustment method, device, equipment and medium to solve the technical problem of low adaptability between the above-mentioned seat adjustment mode and the real-time environment.

[0005] The present application provides a method for adaptive adjustment of vehicle seats, the method comprising: obtaining current driving scene information; inputting the current driving scene information into a seat adjustment adaptive model to obtain seat adjustment parameters corresponding to the current driving scene information, the method for obtaining the seat adjustment adaptive model comprising: constructing an initial neural network model, and training the initial neural network model based on historical driving scene information to obtain the seat adjustment adaptive model; adjusting the vehicle seat based on the seat adjustment parameters corresponding to the current driving scene information to achieve adaptive adjustment of the vehicle seat.

[0006] In one embodiment of the present application, the current driving scene information includes driving environment information, driver information, and vehicle status information, and the vehicle seat is adjusted based on the seat parameters corresponding to the current driving scene information. Specifically, when any of the driving environment information, driver information, or vehicle status information changes, new driving scene information is generated based on the changed information, and new seat adjustment parameters are generated based on the new driving scene information to achieve automatic adjustment of the seat.

[0007] In one embodiment of the present application, an initial neural network model is constructed and a seat adjustment adaptive model is obtained, including: constructing an initial neural network model including at least three input nodes, at least two output nodes, and at least one hidden layer, wherein the input nodes are used to receive driving environment information, driver information, and vehicle status information in driving scene information, and the output nodes are used to output seat adjustment parameters; initializing weights and biases in the initial neural network, and setting a first transfer function between the input layer and the hidden layer, and a second transfer function between the hidden layer and the output layer, wherein the first transfer function is a nonlinear activation function, and the second transfer function is a linear activation function or a nonlinear activation function; dividing the historical driving scene information to obtain at least a training data set and a verification data set; training the initial neural network model based on the training data set to learn an initial mapping relationship between the input driving scene information and the seat adjustment parameters; verifying the mapping relationship based on the verification data set, and optimizing the initial mapping relationship based on the verification result to obtain an optimized neural network model, and determining the optimized neural network model as the seat adjustment adaptive model.

[0008] In one embodiment of the present application, an initial neural network model is trained based on the historical driving scene information to obtain a seat adjustment adaptive model, including: dividing the historical driving scene information to obtain at least a training data set and a verification data set; training the initial neural network model based on the training data set to learn the initial mapping relationship between the input driving scene information and the seat adjustment parameters; verifying the mapping relationship based on the verification data set, and optimizing the initial mapping relationship based on the verification result to obtain an optimized neural network model, and determining the optimized neural network model as the seat adjustment adaptive model.

[0009] In one embodiment of the present application, the vehicle seat adaptive adjustment method also includes: extracting numerical data from the current driving scene information, the current driving scene information including numerical data and quantitative data; performing a linear transformation on the numerical data, and using the converted value after the linear transformation as the input value of the seat adjustment adaptive model; predicting the input value based on the seat adjustment adaptive model to obtain a corresponding prediction result; performing a reverse linear transformation on the prediction result to convert the prediction result back to the original data scale.

[0010] In one embodiment of the present application, the numerical data is linearly transformed, including: selecting any type of numerical data as target data, and determining the maximum and minimum values ​​of the target data; calculating a conversion factor based on the maximum and minimum values ​​of the target data, and calculating the conversion value of each data point in the target data based on the conversion factor; traversing all numerical data to obtain the conversion value corresponding to each data point of each type of numerical data.

[0011] In one embodiment of the present application, the vehicle seat adaptive adjustment method also includes: classifying historical driving scene information based on driver identity information to obtain target historical information corresponding to any target driver; obtaining all seat adjustment parameters in the target historical information, and classifying the seat adjustment parameters to obtain multiple seat adjustment models; calculating the frequency of occurrence of each seat adjustment mode, and determining a seat adjustment mode as a historical seat model based on the frequency of occurrence; when the existence of a historical seat mode of the driver is identified, preferentially using the historical seat model as the initial adjustment basis of the vehicle seat, so as to adjust the vehicle seat for the first time before obtaining the current seat adjustment parameters.

[0012] In one embodiment of the present application, the vehicle seat adaptive adjustment method also includes: determining the driver identity based on the driver information in the driving scene information; identifying whether there is a historical seat pattern corresponding to the driver based on the determined driver identity; if so, adjusting the vehicle seat for the first time based on the historical seat pattern before obtaining the current seat adjustment parameters; if not, generating corresponding seat adjustment parameters through the current driving scene information, and recording the seat adjustment parameters as the driver's historical seat pattern.

[0013] The present application provides a vehicle seat adaptive adjustment device, the device comprising: a data acquisition module, used to obtain current driving scene information; a seat adjustment parameter determination module, used to input the current driving scene information into a seat adjustment adaptive model to obtain seat adjustment parameters corresponding to the current driving scene information, the method for obtaining the seat adjustment adaptive model comprising: constructing an initial neural network model, and training the initial neural network model based on historical driving scene information to obtain the seat adjustment adaptive model; a seat adjustment module, used to adjust the vehicle seat based on the seat adjustment parameters corresponding to the current driving scene information to achieve adaptive adjustment of the vehicle seat.

[0014] The present application provides an electronic device, comprising a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the vehicle seat adaptive adjustment method as described above.

[0015] The present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is used to enable a computer to execute the vehicle seat adaptive adjustment method as described above.

[0016] Beneficial effects of the present application: The vehicle seat adaptive adjustment method in the present application includes collecting driving scene information and dividing it into historical driving scene information and current driving scene information; constructing an initial neural network model and training it with historical driving scene information to obtain a seat adjustment adaptive model; inputting current driving scene information into the model to obtain corresponding current seat adjustment parameters; and adjusting the vehicle seat based on these parameters. By establishing a correlation between driving scene information and seat adjustment parameters, the method can generate highly correlated seat adjustment parameters based on current driving scene information, so that seat adjustment is more in line with the driving environment, thereby significantly improving driving safety and comfort.

[0017] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0019] Figure 1is a schematic diagram of an implementation environment of a vehicle seat adaptive adjustment method shown in an exemplary embodiment of the present application;

[0020] Figure 2 is a schematic diagram of the front and rear sight range of a seat adjustment shown in an exemplary embodiment of the present application;

[0021] Figure 3 is a schematic diagram of the steps of a vehicle seat adaptive adjustment method shown in an exemplary embodiment of the present application;

[0022] Figure 4 is a schematic diagram of a neural network structure of a seat adjustment adaptive model shown in an exemplary embodiment of the present application;

[0023] Figure 5 is a block diagram of a vehicle seat adaptive adjustment device shown in an exemplary embodiment of the present application;

[0024] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0025] The following will describe the implementation methods of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, not for limiting the scope of protection of the present application.

[0026] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0027] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0028] Figure 1 It is a schematic diagram of an implementation environment of a vehicle seat adaptive adjustment method shown in an exemplary embodiment of the present application.

[0029] like Figure 1 As shown, the implementation environment of the vehicle seat adaptive adjustment method includes a data acquisition module 101 and a computer module 102. Among them, the data acquisition module 101 is responsible for collecting raw data during the operation of the vehicle, which includes but is not limited to driving environment information, driver information and vehicle status information. Driving environment information is further subdivided into environment type (such as urban roads, highways, rural roads, etc.) and road surface information (such as flatness, wetness, etc.). Its data acquisition equipment can be various sensors installed on the vehicle (such as cameras, pressure sensors, environmental perception sensors, etc.), or it can be data collected by external devices after being connected to the vehicle wirelessly or wired. This application does not impose any restrictions on the specific implementation form of the data acquisition module.

[0030] The computer module 102 is used to process the raw data collected by the data acquisition module 101, specifically including: constructing an initial neural network model, and training it with historical data in the raw data to obtain a seat adjustment adaptive model, and then inputting real-time data in the raw data into the model to obtain the corresponding current seat adjustment parameters, thereby adjusting the vehicle operation based on the seat adjustment parameters to ensure that the seat can quickly adapt to different driving environments and personal preferences. The computer module 102 can be a high-performance processor, a dedicated vehicle-mounted computing platform, or an embedded system integrated in the vehicle. This application also does not impose any restrictions on the specific type of computer equipment.

[0031] Figure 2 It is a schematic diagram of the front and rear viewing range of a seat adjustment shown in an exemplary embodiment of the present application.

[0032] like Figure 2As shown, when the vehicle is in an uphill state, given the combined influence of the driver's height and the vehicle seat height information, the range of its blind spot may be about 10 meters; based on the vehicle seat adaptive adjustment proposed in this application, the vehicle driving scene information in the current state can be collected to determine that its 10-meter blind spot range will greatly increase the risk of vehicle driving, and even cause the vehicle to be in a dangerous driving state. Because the driver's height is non-adjustable information, if the driver straightens his back and stands on tiptoe to check the blind spot in front of him, it will also lead to other unsafe factors. Therefore, this application proposes a method to adaptively adjust the vehicle seat based on the current driving scene information. Its purpose is: first, to enable the main driver's seat adjustment to respond to environmental changes in real time, especially to automatically optimize the viewing angle when the vehicle enters the blind spot, improve the driver's visibility of potential threats, reduce the risk of accidents, and have a revolutionary improvement on driving safety; second, to personalize the viewing angle adjustment strategy, reduce manual operation, reduce driving burden, and improve driving comfort and convenience; third, to ensure that the viewing angle adjustment strategy is executed efficiently without causing unnecessary burdens on other vehicle systems, reflecting the green energy-saving concept of technology.

[0033] Figure 3 It is a schematic diagram of the steps of a vehicle seat adaptive adjustment method shown in an exemplary embodiment of the present application.

[0034] like Figure 3 As shown, in an exemplary embodiment, the vehicle seat adaptive adjustment method includes at least steps S310 to S330, which are described in detail as follows:

[0035] Step S310, obtaining current driving scene information.

[0036] In one embodiment of the present application, the current driving scene information includes driving environment information, driver information, and vehicle status information, and the vehicle seat is adjusted based on the seat parameters corresponding to the current driving scene information. Specifically, when any of the driving environment information, driver information, or vehicle status information changes, a new driving scene information is generated based on the changed information, and new seat adjustment parameters are generated based on the new driving scene information to achieve automatic adjustment of the seat.

[0037] In one embodiment of the present application, the environmental information around the vehicle is obtained in real time through the intelligent driving system (e.g., millimeter wave radar, laser radar, camera and other sensors) carried by the vehicle, including but not limited to traffic conditions, weather conditions, light intensity, etc. on the road. With the help of the driver monitoring system (DMS), the driver's eye height information and visual angle information are collected. The DMS system can detect the driver's eye position and line of sight direction through a camera or other type of sensor installed above the steering wheel column, so as to determine whether the driver is focusing on the road ahead and whether the driver's sitting posture needs to be adjusted to improve the field of vision. The seat's built-in sensor is used to obtain the seat's height, angle and other state information. In addition, the inertial measurement unit (IMU) or other attitude detection device inside the vehicle can also be used to obtain the vehicle's attitude information, including the vehicle's pitch angle, roll angle, heading angle and acceleration in the XYZ direction. All driving scene information collected in the past period of time (such as the last few weeks or months) is archived as historical driving scene information, and the data received from the above sensors and systems at the current moment is regarded as the current driving scene information.

[0038] Step S320, inputting the current driving scene information into the seat adjustment adaptive model to obtain seat adjustment parameters corresponding to the current driving scene information. The method for obtaining the seat adjustment adaptive model includes: constructing an initial neural network model, and training the initial neural network model based on historical driving scene information to obtain the seat adjustment adaptive model.

[0039] In one embodiment of the present application, seat adjustment parameters include adjustment height and adjustment angle; constructing an initial neural network model, including: constructing an initial neural network model including at least three input nodes, at least two output nodes, and at least one hidden layer, the target data input node is used to receive driving environment information, driver information, and vehicle status information in the driving scene information, and the target data output node is used to output the adjustment height and adjustment angle in the seat adjustment parameters; initialize the weights and biases in the target data initial neural network, and set a first transfer function between the input layer and the hidden layer, and a second transfer function between the hidden layer and the output layer, the first transfer function of the target data is a nonlinear activation function, and the second transfer function of the target data is a linear activation function or a nonlinear activation function.

[0040] Figure 4 It is a schematic diagram of the neural network structure of a seat adjustment adaptive model shown in an exemplary embodiment of the present application.

[0041] In one embodiment of the present application, the above driving environment information is further refined into environment type and road surface information, and the driver information is refined into the driver's field of view angle, and an initial neural network model is constructed based on the environment type, road surface information, driver's field of view angle, and vehicle state information. The details are as follows:

[0042] Define the environment type as C, C∈{0,1,2,3,4}, where 0 represents a flat road, 1 represents a flat road to an uphill slope, 2 represents a downhill slope, 3 represents a turn, 4 represents 13, and 5 represents 23;

[0043] The driver's field of view is defined as Among them, ρ z represents the top-down angle, ρ y represents the side view angle, and h represents the driver's field of vision height;

[0044] The road surface information is defined as φ, φ = {p0, p1, p2, p3, p4}, where p i Waypoints representing the path of the vehicle;

[0045] The vehicle state information is defined as X, X = {yaw, pitch, roll}, where yaw represents the heading angle of the vehicle relative to the path, pitch represents the pitch angle of the vehicle relative to the path, and roll represents the roll angle of the vehicle relative to the path;

[0046] In addition, the seat adjustment parameter is defined as D, Where h is the seat height, Indicates the front-to-back angle of the seat.

[0047] Then, 3,000 data samples were collected, which included the environment type (such as urban roads, highways, rural roads, etc.), the driver's field of view (eye height and viewing angle information obtained through the DMS system), road surface information (including road surface flatness, wetness, etc.), and vehicle status information (including the vehicle's heading angle, pitch angle, roll angle, etc. relative to the path). The 3,000 data samples collected were divided into training set, validation set and test set in proportion, where the training set contained 2,400 samples (accounting for 80% of the total number of samples) for training the neural network model; the validation set contained 300 samples (accounting for 10% of the total number of samples) for adjusting the model hyperparameters and preventing overfitting; the test set also contained 300 samples (accounting for 10% of the total number of samples) for the final evaluation of model performance.

[0048] Then, considering the performance and complexity of the neural network model, a 5×3×2 neural network structure is proposed in this embodiment, such as Figure 4As shown, the input layer contains 4 nodes, representing the environment type C, the driver's field of view angle The hidden layer contains 3 nodes, which are used to process the input data and extract features; the output layer contains 2 nodes, which represent the ideal state information of the seat, i.e., the seat adjustment height h and adjustment angle In addition, considering that the ideal state information of the seat is usually a continuous value and the linear function can provide an unconstrained output, tansigmoid is selected as the transfer function of the first and second layers in the embodiment, and the transfer function of the output layer is selected as linear. The weights and biases of the model are initialized using a random initialization method to ensure the stability of the model in the initial stage of training. Subsequently, the model is trained using the training set data, and the model parameters are adjusted using the back propagation algorithm to minimize the loss function until the model converges; the hyperparameters of the model are adjusted using the validation set data and the overfitting phenomenon is monitored; finally, the final performance of the model is evaluated on the test set to ensure that the model can accurately predict the ideal state information of the seat on unseen data, and the final neural network model is defined as the seat adjustment adaptive model.

[0049] It should be noted that adjusting the height and adjusting the angle are only exemplary descriptions of the seat adjustment parameters. In actual applications, the adjustable parameters of the seat may also include the fore-and-aft position, backrest inclination, seat cushion inclination, and lumbar support, etc., to meet the physical conditions and personal preferences of different drivers and provide a more comfortable and safe driving experience. Therefore, this application does not impose any restrictions on the specific categories of its seat adjustment parameters.

[0050] In one embodiment of the present application, an initial neural network model is trained based on historical driving scene information of target data to obtain a seat adjustment adaptive model, including: dividing the historical driving scene information of the target data to obtain at least a training data set and a verification data set; training the target data initial neural network model based on the target data training data set to learn the initial mapping relationship between the input driving scene information and the seat adjustment parameters; verifying the target data mapping relationship based on the target data verification data set, and optimizing the target data initial mapping relationship based on the verification result to obtain an optimized neural network model, and determining the optimized neural network model as the seat adjustment adaptive model.

[0051] In a specific embodiment of the present application, during the training process, the weights and biases of the model are continuously adjusted through the back propagation algorithm to minimize the loss function, thereby learning the initial mapping relationship between the input driving scene information and the seat adjustment parameters. After the training is completed, the validation data set is used to verify the validity of the learned mapping relationship, and the model parameters are adjusted according to the verification results to further optimize the mapping relationship. Through the process of repeated iterations of training and verification, an optimized neural network model is finally obtained. This model can more accurately reflect the relationship between the driving scene information and the seat adjustment parameters, thereby realizing adaptive adjustment of the seat. Finally, this optimized neural network model is determined as a seat adjustment adaptive model and used in subsequent practical applications to ensure that the seat can automatically adjust to the most suitable position according to different driving environments and the needs of the driver, thereby improving driving comfort and safety.

[0052] In addition, in one embodiment of the present application, the vehicle seat adaptive adjustment method proposed in the present application also includes: extracting numerical data from the current driving scene information of the target data, the current driving scene information of the target data includes numerical data and quantitative data; performing a linear transformation on the numerical data of the target data, and using the converted value after the linear transformation as the input value of the target data seat adjustment adaptive model; predicting the target data input value based on the target data seat adjustment adaptive model to obtain a corresponding prediction result; performing a reverse linear transformation on the target data prediction result to convert the prediction result back to the original data scale.

[0053] In a specific embodiment of the present application, first, numerical data is extracted from the current driving scene information, and the current driving scene information includes multiple types of data, including both numerical data and quantitative data. Numerical data include the driver's field of view angle (eye height and viewing angle information obtained through the DMS system), road surface information (such as road surface flatness, wetness, etc.) and vehicle state information (such as vehicle speed, acceleration, attitude angle, etc.), and quantitative data include environment type (such as urban roads, highways, etc.). Then, the extracted numerical data is linearly transformed. The purpose of linear transformation is to standardize the data so that it is suitable for the input of the neural network model. For example, methods such as Min-MaxNormalization or Z-score normalization can be used to convert numerical data into values ​​within a fixed range, such as values ​​in the interval [0,1]. The transformation formula can be expressed as:

[0054]

[0055] Where x′ represents the converted value after conversion, x represents the original data value, min(x) represents the minimum value in the data set, and max(x) represents the maximum value in the data set.

[0056] Then, the converted value after linear transformation is used as the input value of the seat adjustment adaptive model. The seat adjustment adaptive model predicts the input value according to the previously trained mapping relationship to obtain the prediction result corresponding to the current driving scene information, namely the seat adjustment parameter, and the seat adjustment parameter includes the adjustment height and the adjustment angle.

[0057] Finally, the obtained prediction results are subjected to an inverse linear transformation to convert the prediction results back to the original data scale. The inverse transformation formula can be expressed as:

[0058] x=x′(max(x)-min(x))+min(x) Formula (2)

[0059] Where x′ represents the converted value after conversion, x represents the original data value, min(x) represents the minimum value in the data set, and max(x) represents the maximum value in the data set.

[0060] In one embodiment of the present application, a linear transformation is performed on target numerical data, including: selecting any type of numerical data as the target data, and determining the maximum and minimum values ​​of the target data; calculating a conversion factor based on the maximum and minimum values ​​of the target data, and calculating a conversion value of each data point in the target data based on the target data conversion factor; traversing all numerical data to obtain a conversion value corresponding to each data point of each type of numerical data.

[0061] In a specific embodiment of the present application, the driver's field of view angle is selected as the target data, and the maximum value of the driver's field of view angle MaxAngle and the minimum value of the driver's field of view angle MinAngle are obtained through historical data statistics, and the corresponding conversion factors are obtained according to the maximum value of the driver's field of view angle MaxAngle and the minimum value of the driver's field of view angle MinAngle as shown below:

[0062]

[0063] Among them, ScaleFactor represents the driver's field of view angle conversion factor, MaxAngle represents the maximum value of the driver's field of view angle, and MinAngle represents the minimum value of the driver's field of view angle.

[0064] Then, use the conversion factor to calculate the conversion value of each data point in the target data as follows:

[0065] NormalizedAngle=ScaleFactor×(CurrentAngle-MinAngle) Formula (4)

[0066] Among them, NormalizedAngle represents the standard value of the driver's field of view angle after conversion, CurrentAngle represents the current driver's field of view angle value, ScaleFactor represents the driver's field of view angle conversion factor, and MinAngle represents the minimum value of the driver's field of view angle.

[0067] Then, the same method is used to perform linear transformation on other numerical data in the current driving scene information, such as road surface information and vehicle status information. The specific steps are as follows: for each type of numerical data, determine the respective maximum and minimum values, calculate the respective conversion factors based on the maximum and minimum values, and use the conversion factors to calculate the conversion values ​​of each data point in each type of numerical data.

[0068] Finally, all the converted values ​​after linear transformation are used as the input values ​​of the seat adjustment adaptive model. It should be noted that the standardized data obtained based on the above data transformation can better meet the requirements of the model, thereby improving the accuracy of the model prediction.

[0069] It should be noted that standardizing the driver's field of view angle and other numerical data to make them suitable as inputs to the seat adjustment adaptive model can ensure that the model is more efficient and accurate when processing the current driving scene information, thereby achieving precise adaptive adjustment of the seat.

[0070] In addition, it should be noted that in the process of training the initial neural network model or identifying the current driving scene information, specifically, before training the initial neural network model based on the historical driving scene information, and before identifying the current driving scene information based on the seat adjustment adaptive model, it also includes preprocessing of the historical driving scene information and the historical driving scene information, and the preprocessing includes but is not limited to data cleaning, time synchronization, data standardization processing, etc.

[0071] In the process of training the initial neural network model or identifying the current driving scene information, in order to ensure the quality and consistency of the data, a series of preprocessing of the historical driving scene information and the current driving scene information is required. Specifically, before training the initial neural network model based on the historical driving scene information, and before identifying the current driving scene information based on the seat adjustment adaptive model, the preprocessing steps include but are not limited to data cleaning, time synchronization, and data standardization. To ensure the quality of the historical driving scene information and the current driving scene information, thereby improving the training effect of the initial neural network model and the recognition accuracy of the seat adjustment adaptive model for the current driving scene information. These preprocessing measures help eliminate noise, unify data formats, and make data more suitable for model training, thereby achieving more accurate seat adjustment adaptive adjustment.

[0072] Among them, data cleaning includes: eliminating data points of key fields that are obviously wrong or missing to ensure the integrity and validity of the data; detecting and processing outliers through statistical methods or machine learning algorithms to prevent outliers from affecting the effect of model training; using interpolation, mean filling or other appropriate methods to fill in some missing data to maintain data continuity.

[0073] Time synchronization includes: ensuring that data from different sensors are synchronized in time to eliminate time offsets caused by different sensor sampling frequencies; and aligning time series data that need to be used in combination, such as driver's field of view angle and vehicle speed data, on the time axis to ensure data consistency.

[0074] Data standardization processing includes: linear transformation of numerical data, scaling it to a fixed range, such as the interval [0,1], to improve the training efficiency of the model; for non-numerical data (such as environment type), one-hot encoding (One-Hot Encoding) or other encoding methods are used to convert it into numerical data so that the model can process it correctly; and methods such as Z-score standardization or minimum-maximum normalization are used to make different features have similar scales to avoid some features dominating the model training due to excessively large values.

[0075] In another embodiment of the present application, the driver's field of view angle is selected as the target data. In the linear transformation stage, the maximum and minimum values ​​of the driver's field of view angle have been determined, assuming that the maximum value is MaxAngle and the minimum value is MinAngle. The previous maximum and minimum values ​​are used to calculate the reverse conversion factor, so that the standardized data is restored to the original data based on the reverse conversion factor. The formula for calculating the reverse conversion factor is as follows:

[0076] InverseScaleFactor=MaxAngle-MinAngle Formula (5)

[0077] Among them, InverseScaleFactor represents the inverse conversion factor of the driver's field of view angle, MaxAngle represents the maximum value of the driver's field of view angle, and MinAngle represents the minimum value of the driver's field of view angle.

[0078] The formula for calculating the inverse conversion value of each data point in the driver's field of view angle data using the inverse conversion factor is as follows:

[0079] OriginalAngle=NormalizedAngle×InverseScaleFactor+MinAngle Formula (6)

[0080] Among them, OriginalAngle represents the field of view angle value after conversion back to the original scale, NormalizedAngle is the field of view angle value after normalization, InverseScaleFactor represents the inverse conversion factor of the driver's field of view angle, and MinAngle represents the minimum value of the driver's field of view angle.

[0081] In addition, the same method is used to perform inverse linear transformation on other numerical data in the current driving scene information, such as temperature data in the environment type. The specific steps include: for each type of numerical data, using the previously calculated maximum and minimum values ​​to calculate the inverse conversion factor; using the inverse conversion factor to calculate the inverse conversion value of each data point in each type of numerical data.

[0082] Finally, all the data after the inverse linear transformation are converted back to the original scale. It should be noted that the data after the conversion back to the original scale is used as the output of the seat adjustment adaptive model to ensure that the model prediction results can be directly applied to the actual seat adjustment operation.

[0083] It should be noted that the standardized driver's field of view angle and other numerical data are restored to the original data scale so that the output of the model can be directly used in practical applications, thereby achieving precise adaptive adjustment of the seat.

[0084] Step S330, adjusting the vehicle seat based on the seat adjustment parameter corresponding to the current driving scene information to achieve adaptive adjustment of the vehicle seat.

[0085] In one embodiment of the present application, it is assumed that the adjustment height output by the model is H target , adjust the angle to A target In other words, the seat should be adjusted to a height of H target and angle A target Therefore, firstly, the actual height H of the current seat is read through the built-in sensor of the seat. current and angle A current , and calculate the difference between the current state of the seat and the target state, as follows:

[0086] ΔH=H target -H current

[0087] ΔA=A target -A current Formula (7)

[0088] Among them, ΔH represents the adjustment height difference, H target Indicates the target adjustment height, H current Indicates the current seat height; ΔA indicates the adjustment angle difference, A targetIndicates the target adjustment angle, A current Indicates the current seat angle.

[0089] Then, according to the calculated differences ΔH and ΔA, the corresponding adjustment instructions are generated. Specifically, if ΔH>0, an instruction to increase the seat height is sent; if ΔH<0, an instruction to lower the seat height is sent. Similarly, if ΔA>0, an instruction to increase the seat tilt angle is sent; if ΔA<0, an instruction to reduce the seat tilt angle is sent.

[0090] Finally, after receiving the adjustment command, the seat control system drives the seat motor or hydraulic system to adjust the seat. The seat height is adjusted by the electric lifting mechanism until it reaches H target The seat angle is adjusted by the electric tilt mechanism until it reaches A target .

[0091] In addition, after the adjustment is completed, the actual height and angle of the seat can be read again through the built-in sensor of the seat, and this information can be fed back to the control system to determine whether the seat has accurately reached the target position. If there is still a deviation, the adjustment process is repeated until the seat reaches the specified position. When the seat is successfully adjusted to H target and A target After that, the control system confirms that the adjustment is completed and stops sending adjustment instructions.

[0092] It should be noted that, based on the solution proposed in this embodiment, the seat can automatically adjust to the most suitable position according to the current driving scene information, thereby realizing adaptive adjustment of the vehicle seat, which not only improves driving comfort but also enhances driving safety.

[0093] In one embodiment of the present application, the target data method also includes: determining the driver identity based on the driver information in the target data driving scene information; identifying whether there is a historical seat pattern of the driver based on the determined driver identity; if so, adjusting the vehicle seat for the first time before obtaining the current seat adjustment parameters based on the target data historical seat pattern; if not, generating corresponding seat adjustment parameters through the current driving scene information, and recording the target data seat adjustment parameters as the driver's historical seat pattern.

[0094] In a specific embodiment of the present application, the current driving scene information is collected through sensors and systems on the vehicle. After such information includes but is not limited to the environment type (such as urban roads, highways, etc.), driver information (such as driver identity information), road surface information (such as road surface flatness, wetness, etc.) and vehicle status information (such as vehicle speed, acceleration, attitude angle, etc.), the driver monitoring system (DMS) of the vehicle is further used to obtain the driver's identity information. This may include facial recognition, fingerprint recognition or other biometric recognition technologies to determine who the current driver is. Take the system identifying the current driver as "Zhang San" as an example. According to the determined driver identity (Zhang San), the system is queried whether there is a historical seat mode of the driver. The historical seat mode stores the seat adjustment parameters used by Zhang San in the past, including the height and angle of the seat. If there is a historical seat mode of Zhang San, the system will perform initial adjustment on the vehicle seat based on these historical seat modes.

[0095] If the system reads Zhang San's historical seat mode and finds that his preferred seat height is H history and angle A history , then send a command to the seat control system to adjust the seat to H history and A history location.

[0096] If the system query result shows that there is no historical seat mode for Zhang San, the system will enter the learning mode. The corresponding seat adjustment parameters are generated based on the driver's field of view angle, vehicle posture information, etc. in the current driving scene information. Assume that the seat adjustment parameters generated by the system based on the current driving scene information are H current and A current . Adjust the seat to H current and A current position, and adjust the parameter (H current and A current ) is recorded as Zhang San’s historical seat pattern.

[0097] It should be noted that, whether in the case of an existing historical seat mode or in the case of the first generation and recording, the system will ensure that the seat is adjusted to a position suitable for the current driver. In other words, the above historical seat mode is only used as the first adjustment of the vehicle seat after the driver enters the cab. It will enter the learning mode as usual later, and adaptively adjust the vehicle seat when the driving scene information changes. In addition, based on the scheme proposed in the above embodiment, it is possible to automatically identify whether the historical seat mode needs to be used for initial adjustment based on the driver's identity information. If there is no historical mode, new seat adjustment parameters are generated based on the current driving scene information and recorded for future use, which not only improves the efficiency of seat adjustment, but also ensures that each driver can get a personalized seat adjustment experience.

[0098] In one embodiment of the present application, after determining the driver identity based on the driver information in the target data driving scene information, it also includes: classifying the historical driving scene information based on the driver identity information to obtain the target historical information corresponding to any target driver; obtaining all seat adjustment parameters in the target historical information of the target data, and classifying the target data seat adjustment parameters to obtain multiple seat adjustment models; calculating the frequency of occurrence of each seat adjustment mode, and determining a seat adjustment mode as a historical seat model based on the target data occurrence frequency; when it is identified that there is a historical seat mode of the target data driver, the target data historical seat model is preferentially used as the initial adjustment basis for the target data vehicle seat, so as to adjust the vehicle seat for the first time before obtaining the current seat adjustment parameters.

[0099] In a specific embodiment of the present application, the current driving scene information is collected through sensors and systems on the vehicle. After this information includes but is not limited to the environment type (such as urban roads, highways, etc.), driver information (such as driver identity information), road surface information (such as road surface flatness, wetness, etc.) and vehicle status information (such as vehicle speed, acceleration, attitude angle, etc.), the driver's identity information is obtained using the vehicle's driver monitoring system (DMS). The identity authentication can be achieved through technologies such as facial recognition, fingerprint recognition or other biometric recognition, and the present application does not impose any specific restrictions on the method of identity authentication. Take the system identifying the current driver as "Li Si" as an example.

[0100] First, according to the determined driver identity (Li Si), all historical records related to Li Si are screened from the historical driving scene information database to obtain the target historical information corresponding to Li Si, which includes various driving scene information and corresponding seat adjustment parameters when Li Si drove in the past. All seat adjustment parameters are extracted from Li Si's target historical information, including the height and angle of the seat.

[0101] Secondly, the extracted seat adjustment parameters are classified to form multiple seat adjustment models. Each model represents a typical seat adjustment mode. For example, according to different driving environments (such as urban roads, highways) and driver habits, the seat adjustment parameters can be divided into several main modes.

[0102] Then, the number of times each seat adjustment model appears in Li Si's historical driving records is counted, and the seat adjustment model with the highest frequency of occurrence is selected as the historical seat model. For example, if a seat adjustment model A appears 100 times and another model B appears 50 times, then model A has the highest frequency of occurrence, so model A is selected as the historical seat model.

[0103] Finally, after identifying that Li Si's historical seat mode is mode A, the system will give priority to this mode as the basis for the initial adjustment of the vehicle seat, and send instructions to the seat control system according to the seat adjustment parameters (such as height and angle) in mode A to adjust the seat to the corresponding position.

[0104] It should be noted that after confirming that the seat has been adjusted to the position specified by the historical seat model, the adjustment process is recorded and the self-learning mode is automatically entered. If the system detects new driving scene information during the subsequent driving process, the seat adjustment parameters are further optimized and the historical seat model is updated. In addition, based on the solution proposed in this embodiment, the seat adjustment mode that best suits the current driver can be automatically identified and adopted according to the driver's identity information and historical driving records, thereby achieving adaptive adjustment of the seat and improving driving comfort and safety.

[0105] Finally, it should be noted that the vehicle seat adaptive adjustment method proposed in the present application has the following effects: First, the method proposed in the present application can realize automatic adjustment of the seat through a neural network model. When the driver encounters a blind spot, the system will automatically calculate the seat adjustment reference amount, such as the seat height and angle, based on the collected driver's eye height information and viewing angle information, seat status information, and vehicle posture information, thereby realizing automatic adjustment of the seat and greatly improving the convenience of operation; secondly, the neural network model is used to calculate the seat adjustment reference amount. Compared with the existing manual adjustment method and electric adjustment method, its adjustment accuracy and flexibility have been greatly improved, which can better meet the driver's personalized needs; finally, by dynamically adjusting the seat, the driver can have a better driving field of vision, thereby reducing visual blind spots and improving driving safety.

[0106] Figure 5 is a block diagram of a vehicle seat adaptive adjustment device shown in an exemplary embodiment of the present application. The device can be applied to Figure 1 The device may also be applicable to other exemplary implementation environments and specifically configured in other devices, and this embodiment does not limit the implementation environment to which the device is applicable.

[0107] like Figure 5 As shown, the exemplary vehicle seat adaptive adjustment device includes: a data acquisition module 510 , a seat adjustment parameter determination module 520 , and a seat adjustment module 530 .

[0108] Among them, the data acquisition module 510 is used to obtain the current driving scene information; the seat adjustment parameter determination module 520 is used to input the current driving scene information into the seat adjustment adaptive model to obtain the seat adjustment parameters corresponding to the current driving scene information. The method for obtaining the seat adjustment adaptive model includes: constructing an initial neural network model, and training the initial neural network model based on historical driving scene information to obtain the seat adjustment adaptive model; the seat adjustment module 530 is used to adjust the vehicle seat based on the seat adjustment parameters corresponding to the current driving scene information to achieve adaptive adjustment of the vehicle seat.

[0109] It should be noted that the vehicle seat adaptive adjustment device provided in the above embodiment and the vehicle seat adaptive adjustment method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, and will not be repeated here. In actual application, the vehicle seat adaptive adjustment device provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0110] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the electronic device implements the vehicle seat adaptive adjustment method provided in the above-mentioned embodiments.

[0111] Figure 6 The structure diagram of the computer system suitable for implementing the electronic device of the embodiment of the present application is shown. It should be noted that: Figure 6 The computer system 600 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0112] like Figure 6As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage part 608 to the random access memory (RAM) 603, such as executing the method described in the above embodiment. In the RAM 603, various programs and data required for system operation are also stored. The CPU 601, ROM 602 and RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0113] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read therefrom is installed into the storage section 608 as needed.

[0114] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 609, and / or installed from a removable medium 611. When the computer program is executed by a central processing unit (CPU) 601, various functions defined in the system of the present application are executed.

[0115] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0116] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0117] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0118] Another aspect of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the vehicle seat adaptive adjustment method as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiment, or may exist independently without being assembled into the electronic device.

[0119] Another aspect of the present application also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the vehicle seat adaptive adjustment method provided in each of the above embodiments.

[0120] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A vehicle seat adaptive adjustment method, characterized in that: The method comprises: Get current driving scene information; Inputting the current driving scene information into a seat adjustment adaptive model to obtain a seat adjustment parameter corresponding to the current driving scene information, wherein the method for obtaining the seat adjustment adaptive model comprises: constructing an initial neural network model, and training the initial neural network model based on historical driving scene information to obtain the seat adjustment adaptive model; The vehicle seat is adjusted based on the seat adjustment parameter corresponding to the current driving scene information to achieve adaptive adjustment of the vehicle seat.

2. The vehicle seat adaptive adjustment method according to claim 1, characterized in that: The current driving scene information includes driving environment information, driver information, and vehicle status information. The vehicle seat is adjusted based on the seat parameter corresponding to the current driving scene information, specifically, When any of the driving environment information, driver information, or vehicle status information changes, new driving scene information is generated based on the changed information, and new seat adjustment parameters are generated based on the new driving scene information to achieve automatic adjustment of the seat.

3. The vehicle seat adaptive adjustment method according to claim 1, characterized in that: Construct an initial neural network model and obtain a seat adjustment adaptive model, including: Constructing an initial neural network model including at least three input nodes, at least two output nodes, and at least one hidden layer, wherein the input nodes are used to receive driving environment information, driver information, and vehicle status information in the driving scene information, and the output nodes are used to output seat adjustment parameters; Initializing weights and biases in the initial neural network, and setting a first transfer function from the input layer to the hidden layer, and a second transfer function from the hidden layer to the output layer, wherein the first transfer function is a nonlinear activation function, and the second transfer function is a linear activation function or a nonlinear activation function; Dividing the historical driving scene information to obtain at least a training data set and a verification data set; Training the initial neural network model based on the training data set to learn an initial mapping relationship between input driving scene information and seat adjustment parameters; The mapping relationship is verified based on the verification data set, and the initial mapping relationship is optimized based on the verification result to obtain an optimized neural network model, and the optimized neural network model is determined as the seat adjustment adaptive model.

4. The vehicle seat adaptive adjustment method according to claim 1, characterized in that: The method further comprises: extracting numerical data from the current driving scene information, wherein the current driving scene information includes numerical data and quantitative data; Performing a linear transformation on the numerical data, and using the converted value after the linear transformation as an input value of the seat adjustment adaptive model; Predicting the input value based on the seat adjustment adaptive model to obtain a corresponding prediction result; An inverse linear transformation is performed on the prediction results to convert the prediction results back to the original data scale.

5. The vehicle seat adaptive adjustment method according to claim 4, characterized in that: Performing a linear transformation on the numerical data includes: Select any type of numerical data as target data, and determine the maximum and minimum values ​​of the target data; Calculating a conversion factor based on the maximum value and the minimum value of the target data, and calculating a conversion value of each data point in the target data based on the conversion factor; Traverse all numerical data to obtain the conversion value corresponding to each data point of each type of numerical data.

6. The vehicle seat adaptive adjustment method according to any one of claims 1 to 5, characterized in that: The method further comprises: Classify the historical driving scene information based on the driver identity information to obtain the target historical information corresponding to any target driver; Acquiring all seat adjustment parameters in the target historical information, and classifying the seat adjustment parameters to obtain a plurality of seat adjustment models; Calculating the occurrence frequency of each seat adjustment mode, and determining a seat adjustment mode as a historical seat model based on the occurrence frequency; When the driver's historical seat model is identified, the historical seat model is preferentially used as the basis for initial adjustment of the vehicle seat, so as to adjust the vehicle seat for the first time before obtaining the current seat adjustment parameters.

7. The vehicle seat adaptive adjustment method according to claim 6, characterized in that: The method further comprises: determining a driver identity based on the driver information in the driving scene information; Whether there is a historical seat mode corresponding to the driver according to the determined driver identity identification; If so, adjusting the vehicle seat for the first time based on the historical seat mode before obtaining the current seat adjustment parameter; If not, the corresponding seat adjustment parameters are generated according to the current driving scene information, and the seat adjustment parameters are recorded as the historical seat mode of the driver.

8. A vehicle seat adaptive adjustment device, characterized in that: The device comprises: Data acquisition module, used to obtain current driving scene information; A seat adjustment parameter determination module, used for inputting the current driving scene information into a seat adjustment adaptive model to obtain a seat adjustment parameter corresponding to the current driving scene information, wherein the method for obtaining the seat adjustment adaptive model comprises: constructing an initial neural network model, and training the initial neural network model based on historical driving scene information to obtain the seat adjustment adaptive model; The seat adjustment module is used to adjust the vehicle seat based on the seat adjustment parameter corresponding to the current driving scene information to achieve adaptive adjustment of the vehicle seat.

9. An electronic device, characterized in that: It comprises a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute the computer program stored in the memory to implement the vehicle seat adaptive adjustment method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is used to enable a computer to execute the vehicle seat adaptive adjustment method as described in any one of claims 1 to 7.

Images