Method and system for adjusting a vehicle seat

Abstract

The application discloses a kind of adjusting method and system of vehicle seat, method includes: in the case where determining that there is user in vehicle seat, at least one first pressure parameter and each first pressure parameter corresponding first weight parameter are determined, wherein, first pressure parameter is the pressure parameter of user to vehicle seat;At least one first inflation parameter is determined according to first pressure parameter, first weight parameter and neural network model;According to each first inflation parameter, adjustment operation is carried out to the air bag corresponding to first inflation parameter in vehicle seat, to complete the adjustment of vehicle seat, adjustment operation includes inflation operation, deflation operation and one of not operating operation.It can be made so that vehicle seat can be based on the pressure of user to vehicle seat corresponding adjustment, to make adjusted seat more fit human body, more can satisfy the use demand of user to vehicle seat, and, without user manual adjustment, improve the experience of user.

Description

Technical Field

[0001] This application relates to the field of vehicle seat adjustment technology, and more particularly to a vehicle seat adjustment method and system. Background Technology

[0002] With the development of automotive technology, more and more manufacturers are defining more scenario functions in the seat cabin. For example, enabling car seats to be adjusted according to user needs to make the seat more ergonomic and improve the user experience.

[0003] In existing technology, vehicle seats offer a comfort mode. When a user wants to adjust the seat to better suit their needs, they trigger a one-button start, and the vehicle controls the seat to recline the backrest and raise the leg rest, allowing the user to lie down more comfortably. However, this method is unsafe while driving because the seatbelt's restraint function is lost when the backrest is flat. Furthermore, the limited range of seat adjustments is too simplistic and cannot meet the diverse needs of users. Additionally, existing vehicle seats also offer some fixed back and lumbar massage modes, activated by a control button. This method, however, limits the types of seat adjustments and cannot meet the needs of users with different body types and sitting postures.

[0004] Therefore, the existing seat adjustment methods have the problem of limited seat adjustment types, which cannot meet the seat adjustment needs of different users. Summary of the Invention

[0005] The purpose of this application is to solve the problem that the existing technology has limited seat adjustment types, which cannot meet the seat adjustment needs of different users.

[0006] In a first aspect, the present application provides a method for adjusting a vehicle seat, the method comprising: when it is determined that a user is present in the vehicle seat, determining at least one first pressure parameter and a first weight parameter corresponding to each first pressure parameter, wherein the first pressure parameter is the pressure parameter exerted by the user on the vehicle seat; determining at least one first inflation parameter based on the first pressure parameter and the first weight parameter; and adjusting an airbag disposed in the vehicle seat corresponding to the first inflation parameter according to each first inflation parameter, thereby completing the adjustment of the vehicle seat, the adjustment operation including one of an inflation operation, a deflation operation, and no operation.

[0007] By employing the above technical solution, when it is confirmed that a user is present in the vehicle seat, at least one pressure parameter and a corresponding first weight parameter of the user's pressure on the vehicle seat are determined. At least one first inflation parameter is obtained based on the first pressure parameter and the first weight parameter. Adjustments are then made to the airbags in the vehicle seat corresponding to each first inflation parameter, such as inflating, deflating, or not operating, to complete the adjustment of the vehicle seat. In this way, the vehicle seat can be adjusted according to the pressure applied by the user, making the adjusted seat more ergonomic and better meeting the user's needs. Furthermore, manual adjustment is not required from the user, and automatic adjustment based on the user's pressure on the vehicle seat can still be achieved during driving, improving the user experience.

[0008] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application further includes: determining a first mean square error based on at least one first pressure parameter; determining at least one second pressure parameter when it is determined that the vehicle seat adjustment is complete; determining a second mean square error based on at least one second pressure parameter; performing corresponding processing based on the first mean square error and the second mean square error when it is determined that the second mean square error has not converged to zero; and ending the adjustment of the vehicle seat when it is determined that the second mean square error has converged to zero.

[0009] In the implementation of this application, after determining that the vehicle seat has been adjusted, a second pressure parameter and a second mean square deviation are determined. If the second mean square deviation has not converged to zero, corresponding processing is performed based on the first mean square deviation of the first pressure parameter and the second mean square deviation of the second pressure parameter. If the second mean square deviation has converged to zero, the adjustment of the vehicle seat is terminated. In this way, multiple adjustments of the vehicle seat can be achieved, making the vehicle seat more ergonomic and improving the user's seat experience.

[0010] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application performs corresponding processing based on a first mean square deviation and a second mean square deviation, including: when it is determined that the first mean square deviation is less than the second mean square deviation, adjusting each first weight parameter to obtain a second weight parameter, determining at least one second inflation parameter based on a second pressure parameter, the second weight parameter, and a neural network model, and adjusting the airbags installed in the vehicle seat corresponding to the second inflation parameters according to each second inflation parameter to complete the adjustment of the vehicle seat; when it is determined that the first mean square deviation is greater than or equal to the second mean square deviation, determining at least one third inflation parameter based on the second pressure parameter, the first weight parameter, and the neural network model, and adjusting the airbags installed in the vehicle seat corresponding to the third inflation parameter according to the third inflation parameter to complete the adjustment of the vehicle seat.

[0011] By adopting the above technical solution, the first weighting parameter is adjusted according to the relationship between the first and second mean square deviations, so that the inflation parameters are calculated based on the latest weighting parameters, making the obtained inflation parameters more accurate. Furthermore, the second adjustment of the car seat based on the inflation parameters is more precise and better meets the user's needs.

[0012] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application adjusts each first weight parameter to obtain a second weight parameter, including: determining a corresponding first average pressure parameter based on at least one first pressure parameter; determining a first square value of the difference between the first average pressure parameter and the first pressure parameter based on the first pressure parameter and the first average pressure parameter; if the first square value is greater than a preset first threshold, adjusting the weight parameter related to the first pressure parameter in the first weight parameters to obtain the second weight parameter; if the first square value is less than or equal to the preset first threshold, not adjusting the weight parameter related to the first pressure parameter in the first weight parameters, and using the first weight parameter as the second weight parameter.

[0013] By adopting the above technical solution, the adjustment of the weight parameter is determined based on the square of the difference between the pressure parameter and the average pressure parameter and the size of the preset threshold. This makes the inflation parameter calculated by inputting the final weight parameter and the latest pressure parameter into the neural network model more in line with the user's needs, so that the finally adjusted vehicle seat can better meet the user's needs.

[0014] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application adjusts the weight parameter related to the first pressure parameter in the first weight parameter, including: when the first average pressure parameter is greater than the first pressure parameter, determining a first adjustment parameter based on a first square value, so as to increase the weight parameter related to the first pressure parameter in the first weight parameter according to the first adjustment parameter to obtain a second weight parameter; when the first average pressure parameter is less than the first pressure parameter, determining a second adjustment parameter based on a first square value, so as to decrease the weight parameter related to the first pressure parameter in the first weight parameter according to the second adjustment parameter to obtain a second weight parameter.

[0015] By adopting the above technical solution, the weight parameter is increased or decreased based on the magnitude of the pressure parameter and the average pressure parameter. This makes the inflation parameter calculated by inputting the final adjusted weight parameter and the latest pressure parameter into the neural network model more accurate, so that the finally adjusted vehicle seat can better meet the user's needs.

[0016] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application determines at least one first inflation parameter based on a first pressure parameter, a first weight parameter, and a neural network model, including obtaining the first inflation parameter in the following manner:

[0017]

[0018] in, b ij Let b be the first inflation parameter, 1≤i≤n, 1≤j≤n, and b ij ∈[0,1], n*n corresponds to the number of airbags, and σ is the preset activation function. a km Let i be the first pressure parameter, 1≤k≤i, 1≤m≤i, and i*i correspond to the number of the first pressure parameters. x km Let be the first weight parameter, 1≤k≤i, 1≤m≤i, i*i corresponds to the number of first weight parameters, |W|=1.

[0019] By employing the above technical solution, the first inflation parameter is obtained using a preset neural network model calculation formula, resulting in more accurate inflation parameters. Furthermore, vehicle seat adjustment based on the neural network model allows for real-time updates of the model's parameter data, further enhancing its accuracy.

[0020] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application adjusts the airbags in the vehicle seat corresponding to the first inflation parameters according to each first inflation parameter, including: if the first inflation parameter is greater than or equal to a first inflation threshold and less than or equal to a second inflation threshold, then inflating the airbags in the vehicle seat corresponding to the first inflation parameter according to the first inflation parameter; if the first inflation parameter is greater than the second inflation threshold and less than or equal to a third inflation threshold, then not operating the airbags in the vehicle seat corresponding to the first inflation parameter; if the first inflation parameter is greater than the third inflation threshold and less than or equal to a fourth inflation threshold, then deflating the airbags in the vehicle seat corresponding to the first inflation parameter according to the first inflation parameter.

[0021] By adopting the above technical solution, the system determines whether to inflate, deflate, or not inflate the corresponding airbag based on the relationship between the inflation parameters and the inflation threshold. This allows for different adjustment methods for different airbags, resulting in a vehicle seat that is more ergonomically fitted and more satisfying to the user.

[0022] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application uses a first pressure parameter obtained by correspondingly processing the voltage or current value collected by at least one pressure sensor installed in the vehicle seat.

[0023] By employing the above technical solution, a pressure sensor is used to collect pressure or current values, and these values ​​are then processed accordingly to obtain pressure parameters. In this way, using a pressure sensor reduces the bandwidth requirements for determining pressure parameters.

[0024] According to another specific embodiment of this application, the vehicle seat adjustment method disclosed in this application obtains the first pressure parameter in the following manner: if the pressure sensors are connected in series, the pressure value of each pressure sensor and the total pressure value are determined, and the first pressure parameter corresponding to each pressure sensor is obtained according to the ratio of the voltage value of each pressure sensor to the total voltage value; if the pressure sensors are connected in parallel, the current value of each pressure sensor is determined, the pressure value of each pressure sensor and the total pressure value are determined according to the current value of each pressure sensor, and the first pressure parameter corresponding to each pressure sensor is obtained according to the ratio of the pressure value of each pressure sensor to the total pressure value.

[0025] By adopting the above technical solution, pressure sensors are arranged in series or parallel, and the ratio of the obtained voltage value to the total voltage value is used as the first pressure parameter, or the voltage value is calculated based on the obtained current value and the ratio of the voltage value to the total voltage value is used as the first pressure parameter. In this way, the gravity distribution ratio of the human body to each pressure sensor can be obtained simply and quickly, thereby enabling the adjustment of the inflation volume of airbags in each part, making the seat adjustment more precise and accurate.

[0026] Secondly, embodiments of this application provide a vehicle seat adjustment system, including a sensor and a controller. The sensor is used to collect at least one pressure value or at least one current value and send the pressure value or current value to the controller. The controller is used to determine a corresponding first pressure parameter based on the pressure value or current value when it is determined that a user is present in the vehicle seat, and to determine a first weight parameter corresponding to each first pressure parameter. The controller determines at least one first inflation parameter based on the first pressure parameter, the first weight parameter, and a neural network model. The controller adjusts the airbags in the vehicle seat corresponding to each first inflation parameter according to each first inflation parameter to complete the adjustment of the vehicle seat. The adjustment operation includes one of the following: inflation operation, deflation operation, and no operation. The first pressure parameter is the pressure parameter of the user on the vehicle seat.

[0027] The vehicle seat adjustment system provided in this application includes a module for performing the vehicle seat adjustment method provided in the first aspect, and thus can also achieve the beneficial effects (or advantages) of the vehicle seat adjustment method provided in the first aspect.

[0028] Thirdly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by an electronic device, are used to implement the vehicle seat adjustment method provided by the implementation of the first aspect described above.

[0029] It is understood that the beneficial effects of the second and third aspects mentioned above can also be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0030] The vehicle seat adjustment method provided in this application determines at least one pressure parameter and a corresponding first weight parameter applied by the user to the vehicle seat. Based on the first pressure parameter, the first weight parameter, and a neural network model, at least one first inflation parameter is obtained. Adjustments are then made to the airbags in the vehicle seat corresponding to each first inflation parameter, such as inflating, deflating, or not operating, to complete the adjustment of the vehicle seat. This allows the vehicle seat to adjust according to the pressure applied by the user, resulting in a more ergonomic fit and better meeting the user's needs. Furthermore, manual adjustment is unnecessary, and the seat can automatically adjust based on the user's pressure while driving, enhancing the user experience.

[0031] Furthermore, after the first adjustment is completed, the second pressure parameter is collected again, and the first mean square deviation of the first pressure parameter and the second mean square deviation of the second pressure parameter are calculated. Then, based on the relationship between the magnitudes of the first and second mean square deviations, a second adjustment is made until the mean square deviation of the latest collected pressure parameter converges to zero. Only then is the adjustment considered appropriate. In this way, multiple adjustments can be made for the user, and the inflation parameters obtained from the seat adjustment are different for different users, making the seat adjustment more user-friendly, more comfortable for the user, and improving the user experience. Attached Figure Description

[0032] Figure 1 A flowchart illustrating the method for adjusting a vehicle seat as provided in this application;

[0033] Figure 2 A schematic diagram showing the distribution of gravity sensors provided for the implementation of this application;

[0034] Figure 3 A schematic diagram of the gravity sensor provided for the implementation of this application;

[0035] Figure 4 A flowchart illustrating another method for adjusting a vehicle seat as provided in this application;

[0036] Figure 5 A flowchart illustrating another method for adjusting a vehicle seat as provided in this application;

[0037] Figure 6 A schematic diagram of a vehicle seat adjustment system provided for the implementation of this application;

[0038] Figure 7 A schematic diagram of another vehicle seat adjustment system provided for the implementation of this application. Detailed Implementation

[0039] The following specific embodiments illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Although the description of this application will be presented in conjunction with preferred embodiments, this does not mean that the features of this invention are limited to this embodiment. On the contrary, the purpose of describing the invention in conjunction with embodiments is to cover other options or modifications that may be derived based on the claims of this application. To provide a thorough understanding of this application, many specific details will be included in the following description. This application may also be implemented without using these details. Furthermore, to avoid confusion or obscuring the focus of this application, some specific details will be omitted in the description. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0040] The terms “first”, “second”, etc., are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0041] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] As mentioned above, the existing seat adjustment methods have the problem of limited seat adjustment types, which cannot meet the seat adjustment needs of different users.

[0043] With the increasing popularity of family cars and product iterations, more and more manufacturers are defining new scenario functions for seats, including the function of car seats conforming to the human body to meet the comfort requirements of users during use, especially during long-distance driving.

[0044] For example, many vehicle brands have added comfort features such as zero-gravity seats and seat back massage functions. However, these features all have limitations. Take zero-gravity seats as an example. Zero-gravity seats are a comfort cabin scenario. Their function is achieved by raising the seat cushion, reclining the backrest, and raising the leg rest after a one-button start. The problem is that they cannot be applied to the driver's seat; they can only be used in non-driver areas such as the front passenger seat and rear seats. Furthermore, reclining the backrest during driving can cause the seatbelt's restraint and protection function to fail, resulting in a lack of safety and making them less safe to use while driving. Additionally, their comfort functions cannot dynamically adjust to the body's posture. Therefore, a new seat adjustment method that can adjust the seat according to the body's posture to provide sufficient support is needed.

[0045] As for functions such as seat back / lumbar massage, only some fixed modes of back / lumbar massage are provided. Since these massage comfort functions only have a fixed inflation volume, the massage stroke is fixed and cannot provide personalized human comfort support for passengers of different body types and sitting postures.

[0046] Based on this, a seat adjustment method that can automatically adjust the seat shape according to the driver's body posture, thereby improving the user's physical comfort while ensuring driving safety, is emerging.

[0047] This application provides a method for adjusting a vehicle seat, executed by the vehicle or a cloud server. When a user is present in the vehicle seat, a first pressure parameter of the user on the seat is determined. A first inflation parameter is obtained based on the first pressure parameter. The airbags installed in the vehicle seat are adjusted accordingly based on the first inflation parameter, including inflating, deflating, or not initiating any operation on the airbags. This completes the adjustment of the vehicle seat, making the current seat posture more responsive to the user's needs and allowing the user to experience the seat more comfortably.

[0048] The vehicle seat adjustment method provided in this application, taking a vehicle as an example, is described in [reference needed]. Figure 1 Specifically, it includes the following steps.

[0049] S110, if it is determined that there is a user in the vehicle seat, at least one first pressure parameter and a first weight parameter corresponding to each first pressure parameter are determined, wherein the first pressure parameter is the pressure parameter of the user on the vehicle seat.

[0050] In one implementation of this application, the first pressure parameter is obtained by processing the voltage or current values ​​collected by at least one gravity sensor installed in the vehicle seat. Furthermore, the number of first pressure parameters is the same as the number of gravity sensors. The number of first weighting parameters is the same as the number of first pressure parameters.

[0051] For example, such as Figure 2 As shown, at least one microgravity sensor (as an example of a gravity sensor) for acquiring human posture is arranged in the seat cushion and / or backrest. For example, a total of 16 microgravity sensors are arranged in different areas of the seat cushion and / or backrest. When the microgravity sensor detects a pressure change, it is assumed that a user is present in the vehicle seat, and a first pressure parameter is then acquired.

[0052] Of course, in another implementation of this application, the in-vehicle camera can detect that a passenger (i.e., a user) is sitting in the seat, which triggers the microgravity sensor to collect the first pressure parameter. Of course, other methods can be used to detect whether a user is in the vehicle seat. The implementation method in this application is only for illustrative purposes.

[0053] Furthermore, while existing pressure distribution testing technologies can be applied to this function, the excessive amount of pressure distribution cloud map data exceeds the bandwidth requirements of the vehicle's CANFD bus. This results in time delays in the microcontroller unit's (MCU) calculations and communication, causing the adaptive function to react slowly and fail to conform to the human body in real time. Therefore, this application proposes a redesigned data analysis and processing algorithm to achieve human posture recognition in the adaptive seat.

[0054] In one implementation of this application, the 16 microgravity sensors can be connected in series or in parallel.

[0055] If the gravity sensors are connected in series, the voltage value of each gravity sensor is determined, and the first pressure parameter corresponding to each gravity sensor is obtained based on the ratio of the voltage value of each gravity sensor to the total voltage value.

[0056] For example, if a series connection is used, the resistance value of the microgravity sensor will change when it is compressed. The pressure of each microgravity sensor can be calculated by detecting the voltage value of each microgravity sensor. Furthermore, if the resistance value of the electronic control unit (ECU) at the end of the sensor is much smaller than the resistance value of each microgravity sensor, the ratio of the voltage value of each microgravity sensor to the total pressure value can directly represent the proportion of the human body weight in the area of ​​each microgravity sensor being compressed.

[0057] That is, the ratio of the pressure value collected by the microgravity sensor to the total voltage value of all microgravity sensors is used as the ratio of the human body's weight in the microgravity sensor area (i.e., the first pressure parameter).

[0058] Furthermore, if the pressure sensors are connected in parallel, the current value of each pressure sensor is determined, and the pressure value of each pressure sensor and the total pressure value are determined based on the current value of each pressure sensor. The first pressure parameter corresponding to each pressure sensor is obtained based on the ratio of the pressure value of each pressure sensor to the total pressure value, and then the overall sitting posture of the human body is calculated based on the first pressure parameter.

[0059] For example, if data is collected in parallel, the voltage value of each microgravity sensor is calculated based on the shunt value (i.e., current value) of the current of each microgravity sensor and the resistance value of each microgravity sensor. Then, the proportion ratio is calculated based on the voltage value of each microgravity sensor and the total voltage value, which is used as the proportion ratio of human body weight in the microgravity sensor area (i.e., the first pressure parameter). Then, the overall sitting posture of the human body is calculated based on the first pressure parameter.

[0060] In one implementation of this application, such as Figure 3 As shown, the microgravity sensor is proposed to use Qrovo's FT-4x00 sensor, which consists of a surface structure connected to a rubber interlayer, a varistor (FT-4x00), and a microprocessor substrate (PCB). The rubber interlayer is connected to the surface structure under the seat cover surface by hot melt adhesive to absorb manufacturing tolerances. When the varistor is compressed, its resistance changes. The microprocessor collects the voltage or current drop and sends the data to the seat controller. The seat controller calculates the first inflation parameter based on the obtained data and sends inflation / deflation commands to the airbag.

[0061] The seat controller can be a massage controller used to inflate the airbags.

[0062] S120, determine at least one first inflation parameter based on the first pressure parameter, the first weight parameter and the neural network model.

[0063] In this implementation, the matrix formed by the first pressure parameters and the matrix formed by the first weight parameters are input into a convolutional neural network model (as an example of a neural network model) to obtain the matrix corresponding to the first inflation parameters. The number of first inflation parameters is the same as the number of airbags installed in the vehicle seat.

[0064] That is, in this application, the number of first pressure parameters and the number of first weight parameters in the neural network model are specified, as are the number of first inflation parameters in the output.

[0065] For example, at least one first inflation parameter can be determined based on a first pressure parameter, a first weight parameter, and a neural network model, and the first inflation parameter can be obtained in the following manner:

[0066]

[0067] in, b ij Let b be the first inflation parameter, 1≤i≤n, 1≤j≤n, and b ij ∈[0,1], n*n corresponds to the number of airbags.

[0068] σ is the preset activation function.

[0069] a km Let i be the first pressure parameter, 1≤k≤i, 1≤m≤i, and i*i corresponds to the number of the first pressure parameters (i.e., the number of gravity sensors).

[0070] x km Let be the first weight parameter, 1≤k≤i, 1≤m≤i, i*i corresponds to the number of first weight parameters, |W|=1.

[0071] For example, taking the 16 first pressure parameters obtained from 16 microgravity sensors as an example, the matrix obtained from the 16 first pressure parameters is as follows:

[0072]

[0073] Among them, a 11 a 12 a 13 a 14 a 21 a 22 a 23 a 24 a 31 a 32 a 33 a 34 a 41 a 42 a 42 a 43 a 44 This is the first pressure parameter.

[0074] Let the weight function be Where |W|=1, the normalization of the final first inflation parameter can be achieved based on the weighting function. Where x 11 x 12 x 13 x 14 x 21 x 22 x 23 x 24 x 31 x 32x 33 x 34 x 41 x 42 x 43 x 44 This is the first weight parameter.

[0075] In the implementation of this application, the weight function is recursively obtained based on the actual vehicle calibration. It is a function that is pre-obtained based on the actual vehicle calibration and preset in the vehicle controller.

[0076] Furthermore, the first layer of recursive propagation is performed using a single-layer convolutional neural network model, and the result is divided by the matrix magnitude of the first pressure parameter to obtain the first inflation parameter.

[0077] For example, according to the formula get A 3x3 matrix. Where σ is the activation function, σ=[c1 c2 c3 c4 c5 c6], which is a 1x6 matrix, |σ|=1, and b1, b2, b3, b4, b5, b6, b7, b8, b9 are the first inflation parameters.

[0078] Thus, the computation result H1 passed by the single-layer convolutional neural network model is obtained, which is a 3*3 matrix. Since it has been normalized, b1 to b9 are all values ​​in the range of [0, 1].

[0079] S130, according to each first inflation parameter, the airbag installed in the vehicle seat corresponding to the first inflation parameter is adjusted to complete the adjustment of the vehicle seat. The adjustment operation includes one of the following operations: inflation operation, deflation operation, and no operation.

[0080] In one implementation of this application, the corresponding adjustment level is first determined based on the calculated first inflation parameter, and then the airbag installed in the vehicle seat is inflated or deflated according to the adjustment level.

[0081] If the first inflation parameter is in the range of [0-0.2], then the adjustment position of the airbag in the vehicle seat corresponding to the first inflation parameter is determined to be the first adjustment position.

[0082] If the first inflation parameter is in the range of [0.2-0.4], then the adjustment position of the airbag in the vehicle seat corresponding to the first inflation parameter is determined to be the second adjustment position.

[0083] In this application, if the adjustment setting is the first adjustment setting or the second adjustment setting, the corresponding airbag is inflated. For example, if the calculated first inflation parameter result is any value in [0-0.2] or [0.2-0.4], it indicates that the support force is weak at this time, so an appropriate inflation operation is required.

[0084] That is, if the first inflation parameter is greater than or equal to 0 (as an example of the first inflation threshold) and less than or equal to 0.4 (as an example of the second inflation threshold), then the airbag in the vehicle seat corresponding to the first inflation parameter is inflated according to the first inflation parameter.

[0085] Furthermore, if the first inflation parameter is located in the range of [0.4-0.6], then the adjustment position of the airbag in the vehicle seat corresponding to the first inflation parameter is determined to be the third adjustment position.

[0086] In this application, if the adjustment gear is the third adjustment gear, the corresponding airbag will not be operated.

[0087] That is, if the first inflation parameter is greater than 0.4 and less than or equal to 0.6 (as an example of the third inflation threshold), the airbags in the vehicle seats corresponding to the first inflation parameter will not be operated.

[0088] For example, if the calculated first inflation parameter is any value in the range of [0.4-0.6], it indicates that the support at this location is adequate, and therefore no inflation or deflation operation is required.

[0089] If the first inflation parameter is in the range of [0.6-0.8], then the adjustment position of the airbag in the vehicle seat corresponding to the first inflation parameter is determined to be the fourth adjustment position.

[0090] If the first inflation parameter is in the range of [0.8-1], then the adjustment position of the airbag in the vehicle seat corresponding to the first inflation parameter is determined to be the fifth adjustment position.

[0091] In this application, if the adjustment gear is the fourth adjustment gear or the fifth adjustment gear, the corresponding airbag is deflated.

[0092] For example, if the calculated first inflation parameter is any value in [0.6-0.8] or [0.8-1.0], it indicates that the support force is too strong and therefore an appropriate deflation operation is required.

[0093] That is, if the first inflation parameter is greater than 0.6 and less than or equal to 1 (as an example of the fourth inflation threshold), then the airbag in the vehicle seat corresponding to the first inflation parameter is deflated according to the first inflation parameter.

[0094] It should be noted that the smaller the value, the greater the amount of air needed for inflation. Conversely, the larger the value, the greater the amount of air needed for deflation. After calculating the inflation parameters for multiple airbags, each airbag is either inflated, deflated, or left unattended based on these parameters. In other words, when calculating using a single-layer convolutional neural network, the position of the first inflation parameter in the matrix corresponds to the airbag.

[0095] In this application, the first inflation parameter is also the inflation amount corresponding to the airbag. Taking nine first inflation parameters as an example, the values ​​of the nine first inflation parameters correspond to the inflation amounts of the airbags at nine different positions. These nine values ​​are the corresponding inflation amounts of the airbags. Only a 5-bit hexadecimal signal needs to be output to describe the inflation amount of this airbag. If the current inflation amount of the airbag is greater than the calculated first inflation parameter, a deflation operation is required. If the current inflation amount is less than the calculated first inflation parameter, an inflation operation is required. If the current inflation amount is equal to the calculated first inflation parameter, no operation is required, and the current inflation amount can be maintained.

[0096] Furthermore, such as Figure 4 As shown, the vehicle seat adjustment method provided in this application further includes the following steps.

[0097] S140, determine the first mean square error based on at least one first pressure parameter.

[0098] For example, the first mean square error is obtained according to the following formula:

[0099]

[0100] in, This is the average value obtained based on each of the first pressure parameters.

[0101] Of course, it can also be based on Calculate the first mean squared error.

[0102] It should be noted that step S140 can be executed after step S110 and before step S120, or it can be executed after step S130.

[0103] S150, after determining that the vehicle seat has been adjusted, determines at least one second pressure parameter.

[0104] For example, after the first inflation adjustment is completed, a microgravity sensor is used to collect body pressure data (i.e., the second pressure parameter) again. The collection method is the same as that for the first pressure parameter, and will not be described in detail here.

[0105] Sixteen secondary pressure parameters were collected and formed into the following matrix:

[0106]

[0107] Among them, a' 11 and a' 12 and a' 13 and a' 14 and a' 21 and a' 22 and a' 23 and a' 24 and a' 31 and a' 32 and a' 33 and a' 34 and a' 41 and a' 42 and a' 43 and a' 44 are the second pressure parameters collected.

[0108] S160. Determine the second mean square error according to at least one second pressure parameter.

[0109] Exemplarily, the second mean square error of the following formula.

[0110]

[0111] Among them, is the average value obtained based on each second pressure parameter.

[0112] Of course, the first mean square error can also be calculated according to Calculate the first mean square error.

[0113] S170. In the case where it is determined that the second mean square error does not converge to zero, perform corresponding processing according to the first mean square error and the second mean square error.

[0114] Exemplarily, use the Cauchy convergence criterion (that is, the Cauchy limit existence criterion) and other methods to determine whether the second mean square error converges to zero. If it does not converge to zero, perform corresponding processing according to the first mean square error and the second mean square error.

[0115] In the implementation manner of this application, performing corresponding processing according to the first mean square error and the second mean square error includes, in the case where it is determined that the first mean square error is less than the second mean square error, adjusting each first weight parameter to obtain a second weight parameter, determining at least one second inflation parameter according to the second pressure parameter, the second weight parameter, and the neural network model, and performing an adjustment operation on the airbag corresponding to the second inflation parameter provided in the vehicle seat according to each second inflation parameter to complete the adjustment of the vehicle seat.

[0116] Exemplarily, compare the magnitudes of S1 (that is, the first mean square error) and S2 (that is, the second mean square error). If S1 < S2, then for the first weight parameter x in the weight function w kmAdjustments are made to obtain the second weight parameter, and then the second weight function w1 is obtained. The new second inflation parameter is determined based on the second weight parameter and the second pressure parameter, and the corresponding airbag is adjusted according to the second inflation parameter.

[0117] In the implementation of this application, adjusting each first weight parameter to obtain a second weight parameter includes determining a corresponding first average pressure parameter based on at least one first pressure parameter, determining a first squared value of the difference between the first average pressure parameter and the first pressure parameter based on the first pressure parameter and the first average pressure parameter, and adjusting the weight parameters related to the first pressure parameter in the first weight parameters when the first squared value is greater than a preset first threshold to obtain the second weight parameter.

[0118] For example, the first mean pressure parameter is determined. Then, the first squared value is determined based on the square of the difference between the first average pressure parameter and the first pressure parameter. Take the first threshold μ, if When this happens, then select a′. ij Position, selected from the first weighting function and a′ ij The corresponding weight function x ij Adjust x ij and with x ij The adjacent weight parameters are used to obtain the second weight parameter x′. ij .

[0119] For example, when If the value is greater than the threshold, then adjust the weight parameters of the adjacent weight functions to x′. 22 、x′ 23 、x′ 24 、x′ 32 、x′ 33 、x′ 34 、x′ 42 、x′ 43 、x′ 44 (Several examples as the second weight parameter) yield the weight function as follows:

[0120]

[0121] In the implementation of this application, during the matrix calculation process, the weight parameters that have a calculation process with the first pressure parameter are considered to be weight parameters related to the first pressure parameter.

[0122] In the implementation of this application, adjusting the weight parameters related to the first pressure parameter in the first weight parameters includes, when the first average pressure parameter is greater than the first pressure parameter, determining a first adjustment parameter based on a first squared value, and increasing the weight parameters related to the first pressure parameter in the first weight parameters according to the first adjustment parameter to obtain a second weight parameter.

[0123] For example, according to The weight is adjusted based on whether it is positive or negative. This indicates that the force applied here is too small, so increase x. ij And the values ​​of its adjacent weighting factors (i.e., weighting parameters). Furthermore, according to... The value determines how much it needs to be increased; for example, it can be determined based on... The value determines the corresponding first adjustment parameter, the existing x ij Add the first adjustment parameter as the second adjustment parameter x′ ij .

[0124] It should be noted that, The correspondence between the value and the first adjustment parameter can be obtained through physical calibration. The implementation method of this application will... The correspondence between the values ​​and the first adjustment parameter is formed in a table and stored in the vehicle controller.

[0125] If the first average pressure parameter is less than the first pressure parameter, a second adjustment parameter is determined based on the first squared value. The weight parameter related to the first pressure parameter in the first weight parameter is reduced based on the second adjustment parameter to obtain the second weight parameter.

[0126] For example, if This indicates that the force applied here is too great, so adjust x to be smaller. ij And the values ​​of its adjacent weighting factors (i.e., weighting parameters). Furthermore, according to... The value determines how much it needs to be reduced; for example, it can be determined based on... The value determines the corresponding second adjustment parameter, the existing x ij Subtract the second adjustment parameter to obtain the second adjustment parameter x′ ij .

[0127] It should be noted that, The correspondence between the value and the first adjustment parameter can be obtained through physical calibration. The implementation method of this application will... The correspondence between the values ​​and the first adjustment parameter is formed in a table and stored in the vehicle controller.

[0128] In this application, the number of pressure parameters corresponds to the number of weight parameter adjustments required. Therefore, a single weight parameter may involve multiple adjustments, which can be made in the order of adjustment, with the last adjustment taking precedence.

[0129] Furthermore, if the first squared value is less than or equal to a preset first threshold, the weight parameters related to the first pressure parameter in the first weight parameters are not adjusted, and the first weight parameters are used as the second weight parameters.

[0130] For example, if If the value of the first weight parameter is appropriate, then there is no need to adjust the weight parameters related to the first pressure parameter. The first weight parameter can be directly used as the second weight parameter, and a new weight function matrix can be formed based on the second weight parameter.

[0131]

[0132] The size of the first threshold can be set according to the actual situation.

[0133] Furthermore, after obtaining the second weight parameter, the matrix formed by the second pressure parameter and the matrix formed by the second weight parameter are input into a single-layer convolutional neural network model to output the second inflation parameter. Then, based on the second inflation parameter, the airbags in the vehicle seat corresponding to the second inflation parameter are adjusted to complete the secondary adjustment of the vehicle seat. The pressure parameter is re-acquired, and the mean square error is recalculated until the calculated mean square error converges to 0, thus completing the adjustment. If it does not converge to 0, the weight parameter is adjusted according to the aforementioned method, and a new inflation parameter is calculated based on the newly acquired pressure parameter and the new weight parameter, or a new inflation parameter is calculated based on the newly acquired pressure parameter and the previously calculated weight parameter.

[0134] The process of adjusting the airbag using the second inflation parameter is the same as the process of adjusting the airbag using the first inflation parameter, and will not be described again here.

[0135] Furthermore, if the first mean square deviation is greater than or equal to the second mean square deviation, at least one third inflation parameter is determined based on the second pressure parameter, the first weight parameter, and the neural network model. The airbag corresponding to the third inflation parameter in the vehicle seat is adjusted according to the third inflation parameter to complete the adjustment of the vehicle seat.

[0136] For example, if it is determined that the first mean square error is greater than or equal to the second mean square error, it indicates that the mean square error is gradually converging, and that the preset first weight parameter is appropriate. Therefore, the matrix generated by the collected second pressure parameter and the matrix generated by the first weight parameter are input into the single-layer convolutional neural network model to obtain the third inflation parameter. Based on the third inflation parameter, the airbag set in the vehicle seat corresponding to the third inflation parameter is adjusted to complete the adjustment of the vehicle seat.

[0137] The process of adjusting the airbag using the third inflation parameter is the same as the process of adjusting the airbag using the first inflation parameter, and will not be described again here.

[0138] S180, once the second mean square error has converged to zero, the adjustment of the vehicle seat is terminated.

[0139] For example, if it is determined that the second mean square error has converged to 0, this indicates that the airbag inflation level is appropriate, and the current inflation level can be maintained without further adjustment of the vehicle seat. Therefore, the adjustment of the vehicle seat is stopped until the microgravity sensor detects the pressure change again.

[0140] When a user is confirmed to be present in the vehicle seat, at least one pressure parameter and a corresponding first weight parameter are determined based on the user's pressure on the vehicle seat. At least one first inflation parameter is obtained based on the first pressure parameter, the first weight parameter, and a neural network model. Adjustments are then made to the airbags in the vehicle seat corresponding to each first inflation parameter, such as inflating, deflating, or not operating, to adjust the vehicle seat. This allows the vehicle seat to adjust based on the pressure applied by the user, resulting in a more ergonomic fit and better meeting the user's needs, without requiring manual adjustment by the user, thus improving the user experience.

[0141] Furthermore, in one implementation of this application, after the first adjustment is completed, a second pressure parameter is collected again, and the root mean square error of the first and second pressure parameters is calculated. Then, based on the relationship between the first and second root mean square errors, subsequent secondary and tertiary adjustments are made until the root mean square error of the latest collected pressure parameter converges to zero. Only then is the adjustment considered appropriate. In this way, multiple adjustments can be made for the user, and the inflation parameters obtained from seat adjustments will be different for different users, making the seat adjustment more suitable for the user, making the user more comfortable to use the seat, and improving the user experience.

[0142] In another implementation of this application, such as Figure 5 As shown, the vehicle seat adjustment method provided in this application further includes the following steps.

[0143] S210, adaptive mode startup.

[0144] For example, after determining that the seat adaptive mode has been activated and that there is a user in the current vehicle seat, the collection of the first pressure parameter begins.

[0145] S220, a microgravity sensor that detects body pressure.

[0146] For example, microgravity sensors are used to obtain the pressure of the human body on each microgravity sensor (i.e., the first pressure parameter).

[0147] S230, single-layer convolutional neural network model transmission.

[0148] For example, the first pressure parameter and the preset first weight parameter are input into a single-layer convolutional neural network model to obtain the first inflation parameter.

[0149] S240, the controller drives the air valve and air pump to perform inflation and deflation operations according to the first inflation parameter.

[0150] S250, the microgravity sensor detects body pressure again to obtain a second pressure parameter.

[0151] S260, determine whether the root mean square error (or second root mean square error) corresponding to the second pressure parameter collected by the pressure sensor (i.e., microgravity sensor) converges to 0. If yes, proceed to step S270; otherwise, proceed to step S280.

[0152] S270, confirm that inflation / deflation is complete, the airbag maintains the current inflation level until the microgravity sensor detects a change in pressure value (i.e., the detected value) again.

[0153] S280, determine whether the root mean square error of the second pressure parameter collected by the pressure sensor after the single-layer convolutional neural network transmits and adjusts the airbag inflation volume has decreased. If yes, proceed to step S230; otherwise, proceed to step S290.

[0154] S290, adjust the weight function w, and execute step S230.

[0155] In this implementation, after the adaptive mode is activated, the microgravity sensor detects body pressure to obtain a first pressure parameter, and a single-layer convolutional neural network model is used to obtain a first inflation parameter. The controller drives the air valve and air pump to perform inflation and deflation operations according to the first inflation parameter. The microgravity sensor detects body pressure again to obtain a second pressure parameter. Based on the second mean square error of the second pressure parameter, it is determined whether the second mean square error has converged to 0. If so, inflation and deflation are stopped, and the airbag inflation volume remains at the current inflation volume until the microgravity sensor detects a change in pressure value again and the seat is readjusted. If not, it is determined whether the second mean square error of the second pressure parameter collected by the microgravity sensor after the single-layer convolutional neural network model is transferred and the controller adjusts the airbag inflation volume is less than the first mean square error of the first inflation parameter (i.e., whether it has decreased). If so, the single-layer convolutional neural network model is transferred again based on the second pressure parameter and the first weight parameter. If not, the weight parameter in the weight function is adjusted to obtain the second weight parameter, and the single-layer convolutional neural network model is transferred based on the second pressure parameter and the second weight parameter.

[0156] The vehicle seat adjustment method provided in this application is a method for adjusting a seat based on an adaptive human posture algorithm derived from a single-layer convolutional neural network. It utilizes pressure sensors at the seat cushion's acquisition end and software algorithms to calculate the human posture, obtaining calculation results (i.e., inflation parameters). Based on these results, air valves under the seat cushion inflate the airbags, causing the cushion to conform to the human body and optimize comfort. Thus, by arranging airbags in the seat cushion, the air valves inflate and deflate the airbags when the seat firmness needs adjustment. Specifically, by calculating the pressure of the human body in different areas of the seat cushion to extract human posture information, and then, based on the human posture measured by the pressure sensors, appropriately inflating the airbags in different areas of the seat cushion, the inflation volume of the airbags in each area changes with the human posture to change the contact surface and wrap around the human body, making the seat cushion's firmness and shape conform to the human body, providing better support and improving comfort.

[0157] Furthermore, by using a neural network model to predict inflation volume, the movement state of the human body during driving can be detected, the most comfortable posture and sitting habits of the human body can be learned, and the calculated inflation parameters can be used for comfort optimization.

[0158] Furthermore, due to the limited communication bandwidth of the vehicle system, the pressure distribution cloud map is not applicable to this scenario. Also, because the controller applied to the seat is a slave node of the CAN bus, the total data size transmitted is limited, and only processed and compressed signals can be used as the logical criteria for driving the airbag operation. Using the algorithm model provided in this application (i.e., a single-layer convolutional neural network model), body pressure data (i.e., the first pressure parameter) can be compressed after collecting human body pressure data and output a 5-bit hexadecimal signal describing human body pressure information (i.e., the first inflation parameter). The airbag inflation volume is adjusted according to the corresponding sitting posture. The data volume meets the communication bandwidth limitations of CANFD, ensuring that the function can be implemented under the existing hardware conditions of the seat controller and vehicle communication bandwidth.

[0159] Furthermore, the controller iterates according to changes in body pressure, adjusting the weight function transmitted by the convolutional neural network to best fit the human body as the posture changes.

[0160] In this implementation, when a user sits on the seat, pressure sensors detect pressure values ​​at various positions. The controller then transmits these values ​​through a convolutional neural network and outputs the airbag inflation value, inflating or deflating the seat. Based on the inflation / deflating results, the software calculates and iterates the weight function of the neural network until the root mean square (RMS) value (mean square error) of the pressure values ​​detected by each sensor converges to 0. At this point, the pressure support on the human body surface is evenly distributed, optimizing comfort. When the user's posture changes, the software continues to adjust the airbag inflation amount according to the change in body pressure, allowing the seat cushion to conform to the human body surface again.

[0161] Furthermore, the vehicle seat adjustment method provided in this application can also be applied to a cloud server. The cloud server receives voltage or current values ​​sent by microgravity sensors installed in the vehicle seat, calculates a first pressure parameter, determines a first inflation parameter based on the first pressure parameter, a first weight parameter, and a neural network model, and sends the first inflation parameter to the vehicle seat controller. The vehicle seat controller adjusts the airbags installed in the vehicle seat corresponding to the first inflation parameter according to each first inflation parameter to complete the adjustment of the vehicle seat. The adjustment operation includes one of the following: inflation operation, deflation operation, or no operation.

[0162] The implementation method of this application is as follows: Figure 6 As shown, a vehicle seat adjustment system, including sensors and controllers, is also disclosed.

[0163] The sensor is used to acquire at least one pressure value or at least one current value and send the pressure value or current value to the controller.

[0164] The controller is used to determine at least one first pressure parameter based on a pressure value or a current value when it is determined that a user is present in the vehicle seat, and to determine a first weight parameter corresponding to each first pressure parameter. Based on the first pressure parameter, the first weight parameter and the neural network, it determines at least one first inflation parameter, determines the adjustment level of each airbag in the vehicle seat corresponding to each first inflation parameter based on each first inflation parameter, and performs an adjustment operation on the corresponding airbag according to the adjustment level to complete the adjustment of the vehicle seat. The adjustment operation includes one of the following operations: inflation operation, deflation operation, and no operation. The first pressure parameter is the pressure parameter of the user on the vehicle seat.

[0165] In this implementation, the vehicle seat adjustment system is the hardware device that executes the vehicle seat adjustment method. The sensor can be the aforementioned microgravity sensor, but it can also be other sensors used to collect pressure information. The controller can specifically be a microcontroller unit (MCU) for controlling the vehicle seat; however, it can also be other in-vehicle terminals, the vehicle's main controller, etc.

[0166] Furthermore, such as Figure 7As shown in the figure, the overall circuit of the control system is as follows. When the microcontroller unit (MCU) receives the vehicle's signal, the MCU calculates the pressure parameters based on the voltage or current values ​​sent by the sensors. Based on the pressure parameters, it calculates nine inflation parameters (i.e., values) representing the human body posture. The MCU controls the opening and closing of the air valve and drives the air pump to work. Each airbag is inflated in five levels of inflation based on the calculation results of b1 to b9, so that the seat / backrest conforms to the human body position and shape to optimize the support effect.

[0167] In this implementation, the microcontroller unit is connected to a DC-DC converter (DCDC), a low-dropout linear stabilizer (LDO), a controller area network (CAN), and a pump drive. The DC-DC converter (DCDC) and the LDO are connected to the power supply (VCC) and ground (GND), respectively. The controller area network (CAN) and the pump drive (PUMPDrive) are connected to ground (GND), respectively. The microcontroller unit communicates with the vehicle via the controller area network, obtains the calculated inflation parameters based on the DC-DC converter, and controls the pump drive to inflate and deflate the airbag.

[0168] This application also provides a chip for executing instructions, which is used to execute the technical solution of the vehicle seat adjustment method in the above embodiments.

[0169] This application also provides a computer-readable storage medium storing computer instructions that, when executed on a computer, cause the computer to perform the technical solution of the vehicle seat adjustment method described above.

[0170] In some possible implementations, various aspects of the methods provided in this application may also be implemented as a program product comprising program code that, when run on a computer device, causes the computer device to perform the steps of the methods described above according to various exemplary embodiments of this application. For example, the computer device may perform a vehicle seat adjustment method described in an embodiment of this application.

[0171] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0172] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when the at least one processor executes the computer program, it can implement the technical solution of the vehicle seat adjustment method in the above embodiments.

[0173] Although this application has been illustrated and described with reference to certain preferred embodiments, those skilled in the art should understand that the above description is a further detailed explanation of the application in conjunction with specific embodiments, and should not be construed as limiting the specific implementation of the application to these descriptions. Those skilled in the art can make various changes in form and detail, including some simple deductions or substitutions, without departing from the spirit and scope of this application.

Claims

1. A method for adjusting a vehicle seat, characterized in that, The method includes: If it is determined that there is a user in the vehicle seat, at least one first pressure parameter and a first weight parameter corresponding to each first pressure parameter are determined, wherein the first pressure parameter is the pressure parameter of the user on the vehicle seat; At least one first inflation parameter is determined based on the first pressure parameter, the first weight parameter, and the neural network model; The airbags in the vehicle seat corresponding to the first inflation parameters are adjusted according to the first inflation parameters to complete the adjustment of the vehicle seat. The adjustment operation includes one of the following operations: inflation operation, deflation operation, and no operation. If it is determined that the vehicle seat has been adjusted, at least one second pressure parameter is determined; The second mean square error is determined based on the at least one second pressure parameter. The second mean square error is obtained by calculating the square value of the difference between the average value of each second pressure parameter and the difference between each second pressure parameter, and summing the square values, or by summing the square values ​​and then calculating the square root of the summed square value. Once the second mean square error converges to zero, the adjustment of the vehicle seat is terminated; If the second mean square error does not converge to zero, the airbag installed in the vehicle seat is adjusted according to the first mean square error and the second mean square error to complete the adjustment of the vehicle seat; the first mean square error is obtained by calculating the square value of the difference between the average value of each first pressure parameter and the difference between each first pressure parameter, and summing the square values, or by summing the square values ​​and then calculating the square root of the summed square value.

2. The vehicle seat adjustment method as described in claim 1, characterized in that, Based on the first mean square error and the second mean square error, corresponding processing is performed, including: If the first mean square error is determined to be less than the second mean square error, the first weight parameter is adjusted to obtain the second weight parameter. At least one second inflation parameter is determined based on the second pressure parameter, the second weight parameter and the neural network model. The airbags in the vehicle seat corresponding to the second inflation parameter are adjusted according to each second inflation parameter to complete the adjustment of the vehicle seat. If the first mean square error is greater than or equal to the second mean square error, at least one third inflation parameter is determined based on the second pressure parameter, the first weight parameter, and the neural network model. The airbag in the vehicle seat corresponding to the third inflation parameter is adjusted according to the third inflation parameter to complete the adjustment of the vehicle seat.

3. The vehicle seat adjustment method as described in claim 2, characterized in that, The second weight parameter is obtained by adjusting the first weight parameter, including: A corresponding first average pressure parameter is determined based on the at least one first pressure parameter; The first square value of the difference between the first average pressure parameter and the first pressure parameter is determined based on the first pressure parameter and the first average pressure parameter. If the first squared value is greater than a preset first threshold, the weight parameter related to the first pressure parameter in the first weight parameter is adjusted to obtain the second weight parameter; If the first squared value is less than or equal to a preset first threshold, the weight parameter related to the first pressure parameter in the first weight parameter is not adjusted, and the first weight parameter is used as the second weight parameter.

4. The vehicle seat adjustment method as described in claim 3, characterized in that, Adjusting the weight parameters related to the first pressure parameter in the first weight parameters includes: When the first average pressure parameter is greater than the first pressure parameter, a first adjustment parameter is determined based on the first square value, and the weight parameter related to the first pressure parameter in the first weight parameter is increased according to the first adjustment parameter to obtain the second weight parameter; If the first average pressure parameter is less than the first pressure parameter, a second adjustment parameter is determined based on the first squared value. The weight parameter related to the first pressure parameter in the first weight parameter is reduced according to the second adjustment parameter to obtain the second weight parameter.

5. The method for adjusting a vehicle seat as described in any one of claims 1-4, characterized in that, Determining at least one first inflation parameter based on the first pressure parameter, the first weight parameter, and the neural network model, including obtaining the first inflation parameter in the following manner: in, , The first inflation parameter is... , , , Corresponding to the number of airbags, The preset activation function, = , The first pressure parameter is... , , Corresponding to the number of the first pressure parameters, , The first weight parameter, , , Corresponding to the number of the first weight parameters, .

6. The method for adjusting a vehicle seat as described in any one of claims 1-4, characterized in that, Adjusting the airbags in the vehicle seat corresponding to the first inflation parameters according to the first inflation parameters includes: If the first inflation parameter is greater than or equal to the first inflation threshold and less than or equal to the second inflation threshold, then the airbag corresponding to the first inflation parameter in the vehicle seat is inflated according to the first inflation parameter. If the first inflation parameter is greater than the second inflation threshold and less than or equal to the third inflation threshold, then the airbag corresponding to the first inflation parameter installed in the vehicle seat will not be operated. If the first inflation parameter is greater than the third inflation threshold and less than or equal to the fourth inflation threshold, then the airbag corresponding to the first inflation parameter in the vehicle seat is deflated according to the first inflation parameter.

7. The method for adjusting a vehicle seat as described in any one of claims 1-4, characterized in that, The first pressure parameter is obtained by processing the voltage or current value collected by at least one pressure sensor installed in the vehicle seat.

8. The method for adjusting a vehicle seat as described in claim 7, characterized in that, The first pressure parameter is obtained in the following manner: If the pressure sensors are connected in series, the voltage value of each pressure sensor and the total voltage value are determined, and the first pressure parameter corresponding to each pressure sensor is obtained according to the ratio of the voltage value of each pressure sensor to the total voltage value. If the pressure sensors are connected in parallel, the current value of each pressure sensor is determined, the pressure value of each pressure sensor and the total pressure value are determined based on the current value of each pressure sensor, and the first pressure parameter corresponding to each pressure sensor is obtained based on the ratio of the pressure value of each pressure sensor to the total pressure value.

9. A vehicle seat adjustment system, characterized in that, Includes sensors and controllers, among which The sensor is used to acquire at least one pressure value or at least one current value, and send the pressure value or the current value to the controller; The controller, upon determining that a user is present in the vehicle seat, determines a corresponding first pressure parameter based on the pressure value or current value, and determines a first weight parameter corresponding to each first pressure parameter. It then determines at least one first inflation parameter based on the first pressure parameter, the first weight parameter, and a neural network model. Based on each first inflation parameter, it adjusts the airbags in the vehicle seat corresponding to that first inflation parameter to complete the adjustment of the vehicle seat. The adjustment operation includes one of inflation, deflation, or no operation. Upon determining that the vehicle seat adjustment is complete, it determines at least one second pressure parameter, and determines a second mean square error based on the at least one second pressure parameter. Finally, it determines that the second mean square error converges to zero. After the adjustment of the vehicle seat is completed, if it is determined that the second mean square error has not converged to zero, the airbag installed in the vehicle seat is adjusted according to the first mean square error and the second mean square error to complete the adjustment of the vehicle seat. The first pressure parameter is the pressure parameter of the user on the vehicle seat. The first mean square error is obtained by calculating the square value of the average value of each first pressure parameter and the difference of each first pressure parameter, and summing the square values, or by summing the square values ​​and then calculating the square root of the summed square value. The second mean square error is obtained by calculating the square value of the average value of each second pressure parameter and the difference of each second pressure parameter, and summing the square values, or by summing the square values ​​and then calculating the square root of the summed square value.

Images