Adjusting method, system and equipment for adjusting driving seat

Abstract

The invention provides a method, a system and equipment for adjusting a driving seat. The method comprises the following steps: acquiring state data of a driver and current seat parameters of the driving seat in a current driving scene; the state data is used for representing the action response of the driver in the current driving scene; determining a target state index according to the state data of the driver; the target state index is used for representing the adaptation degree of the driver to the current seat parameters; under the condition that the target state index exceeds a first threshold value, historical seat parameters in the same driving scene are obtained in a historical vehicle information set; and adjusting the driving seat to adapt to the driver based on the comparison result of the historical seat parameters and the current seat parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobiles, in particular to a driving seat adjustment method, system and device. BACKGROUND

[0002] With the continuous progress of automobile technology, driving comfort and safety have become important factors in vehicle design. As the most direct interface between the driver and the vehicle, the adjustment performance of the seat directly affects the physical state and operation stability of the driver. Especially in complex or dynamic driving scenarios, the seat needs to have higher adaptive ability to meet the individual needs of different drivers and different road conditions. In the related art, it is difficult to respond and adjust in time to relieve the driver's fatigue for complex road conditions or sudden driving behavior. SUMMARY

[0003] The present application provides a driving seat adjustment method, system and device.

[0004] To achieve the above purpose, the technical scheme adopted by the present application is as follows: In a first aspect, the present application provides a driving seat adjustment method, comprising: obtaining state data of a driver and current seat parameters of a driving seat under a current driving scenario; the state data is used to represent the action response of the driver under the current driving scenario; determining a target state index according to the state data of the driver; the target state index is used to represent the degree of adaptation of the driver to the current seat parameters; in the case that the target state index exceeds a first threshold, obtaining historical seat parameters under the same driving scenario in a historical vehicle information set; based on the comparison result of the historical seat parameters and the current seat parameters, adjusting the driving seat to adapt to the driver.

[0005] It can be understood that the scheme provided by the embodiments of the present application first acquires the state data of the driver and the current seat parameter in the current driving scene, calculates a target state index according to the state data, and the target state index represents the adaptability of the driver to the current seat parameter. When the target state index exceeds a first threshold value, it indicates that the current seat parameter does not match the driver, which will cause the driver to drive unsuitably and easily cause fatigue driving. At this time, the historical seat parameter in the same driving scene is extracted from the historical vehicle information, and the driving seat is adjusted based on the comparison result of the historical seat parameter in the same driving scene and the current seat parameter. On the one hand, by comparing the target state index with the first threshold value, the driving seat is adjusted in the case of being greater than the first threshold value, which can not only respond to the adjustment demand of the driver in time, but also avoid the resource waste caused by frequent and small adjustments, and ensure that the adjustment is only started when necessary, so as to save resources. On the other hand, based on the comparison result of the historical seat parameter in the same driving scene and the current seat parameter, the driving seat is adjusted, which can ensure that the driving seat is adjusted according to the optimal seat parameter, so that the seat parameter after adjustment is the most suitable seat parameter in the current driving scene, and the adaptability of the driving seat configuration to the current scene is improved, so as to improve the comfort of the driving seat and relieve the discomfort and / or fatigue of the driver.

[0006] In some embodiments, the adjusting the driving seat based on the comparison result of the historical seat parameter and the current seat parameter comprises: acquiring a first duration corresponding to each type of parameter in the historical seat parameter; acquiring a second duration corresponding to each type of parameter in the current seat parameter; for each first same type of parameter in the historical seat parameter and the current seat parameter: taking the parameter with the longest duration in each first same type of parameter as a target parameter to be adjusted; and adjusting the driving seat according to at least one target parameter to be adjusted.

[0007] It can be understood that the scheme provided by the embodiments of the present application: by respectively acquiring the duration of each type of parameter in the historical and current seat parameters, selecting the parameter with the longest duration as the target parameter to be adjusted, and adjusting the driving seat according to at least one target parameter to be adjusted, the false judgment or interference caused by temporary adjustment can be avoided, and the seat parameter after adjustment can have higher stability and adaptability, and the adjustment effect is further improved.

[0008] In some embodiments, the obtaining the historical seat parameters under the same driving scene in the historical vehicle information set, and adjusting the driving seat based on a comparison result of the historical seat parameters and the current seat parameters, comprises: obtaining historical vehicle information sets of various types of vehicles under the same driving scene; selecting a target information set from the historical vehicle information sets of the various types of vehicles based on the current driving scene; the target information set is the historical vehicle information set that best matches the current driving scene; and adjusting the driving seat based on a comparison result of the seat parameters in the target information set and the current seat parameters.

[0009] It can be understood that the scheme provided by the embodiments of the present application can expand the data source of seat parameter optimization and improve the generalization ability of the recommended parameters by introducing historical vehicle information sets of various types of vehicles and selecting the target information set that best matches the current driving scene. Compared with the traditional method that only relies on local data of a single vehicle, the present application can realize cross-model and cross-user optimization of seat parameters, making the adjustment strategy more adaptive and further improving the adjustment effect.

[0010] In some embodiments, the selecting the target information set from the historical vehicle information sets of the various types of vehicles based on the current driving scene comprises: determining a driving feature vector in the current driving state based on driving data of the vehicle in the current driving scene; determining an average distance between the driving feature vector in the current driving state and a plurality of driving feature vectors in a type vehicle feature vector set in the historical vehicle information set of each type of vehicle for each type of vehicle in the historical vehicle information set of each type of vehicle; determining the type vehicle feature vector set corresponding to the average distance less than a first set distance threshold as a matching feature vector set; and determining the historical vehicle information set corresponding to the matching feature vector set as the target information set.

[0011] It can be understood that the scheme provided by the embodiments of the present application can quantify the similarity between the current driving scene and the historical driving scene by constructing and comparing the average distance between the driving feature vector and the plurality of driving feature vectors in each type vehicle feature vector set, so as to accurately select the matching feature vector set, that is, to accurately select the vehicle that is closest to the driving scene of the vehicle from the different types of vehicles under the same driving scene, and then to adjust the driving seat based on a comparison result of the seat parameters of the vehicle that is closest to the driving scene of the vehicle and the seat parameters of the vehicle. This method is more flexible than the traditional fixed threshold classification method, and can effectively cope with changes in complex or sudden driving scenes, and improve the accuracy and robustness of seat parameter matching.

[0012] In some embodiments, the adjusting the driver seat based on the comparison result of the target information set and the current seat parameter comprises: for each second same type parameter in the target information set and the current seat parameter, taking a type parameter with the longest duration time in each second same type parameter as a target parameter to be adjusted; and adjusting the driver seat according to at least one target parameter to be adjusted.

[0013] It can be understood that the scheme provided by the embodiments of the present application further refines the logic of selecting the target parameter in the target information set, that is, preferentially selecting the parameter with the longest duration time in the same type parameter as the adjustment basis, so that the adjustment action has stronger rationality and stability. Compared with the adjustment mode based on only single point data or experience rules, the present application improves the scientificity of parameter selection through duration time analysis and reduces the occurrence of invalid adjustment.

[0014] In some embodiments, before the adjusting the driver seat based on the comparison result of the historical seat parameter and the current seat parameter, the method further comprises: for each parameter in the current seat parameter, obtaining a duration time of each parameter; determining a parameter with a duration time exceeding a set duration time as a valid parameter; each parameter corresponds to a set duration time; and adjusting the driver seat based on the comparison result of the valid parameters in the historical seat parameter and the current seat parameter.

[0015] It can be understood that the scheme provided by the embodiments of the present application: by setting the set duration time corresponding to different parameters and screening out the valid parameters accordingly, it can filter out those temporary adjustments that have not been fully verified, so as to ensure that the subsequent adjustment is based on a more reliable parameter basis. Compared with the problem of lacking parameter validity verification mechanism in the related art, the present application enhances the credibility and stability of the adjustment decision by introducing the screening mechanism of the duration time.

[0016] In some embodiments, the method further comprises: performing feature processing on each driving data in the plurality of driving data of the vehicle to obtain a plurality of driving feature vectors; performing classification on the plurality of driving feature vectors to obtain a plurality of driving feature vector sets of driving scenes; and for each driving feature vector set of a driving scene in the plurality of driving feature vector sets of driving scenes, determining the historical seat parameter of the same driving scene in the historical vehicle information set based on the seat parameter corresponding to each driving feature vector in each driving feature vector set of the driving scene.

[0017] It can be understood that the scheme provided in the embodiments of the present application: by feature processing on multiple driving data and classifying to form a driving feature vector set of multiple driving scenes, it is helpful to establish a more rich scene model, thereby improving the accuracy of seat parameter matching. Compared with the deficiency of relying on artificial definition of scene or relying on limited data in the traditional method, the present application realizes automatic clustering and updating of driving scenes, which can make the adjustment method of the driver's seat no longer limited to one driving scene, but can adapt to different driving scenes, thereby improving the generalization ability of the adjustment method of the driver's seat.

[0018] In some embodiments, in the case that the state data of the driver includes: the pressure of the driver's hand holding the steering wheel, the frequency of the driver's sitting posture change and the frequency of the driver's blinking, the target state index is determined according to the state data of the driver, including: determining the first weight of the pressure of the driver's hand holding the steering wheel, the second weight of the frequency of the driver's sitting posture change and the third weight of the frequency of the driver's blinking; determining the target state index based on the pressure of the driver's hand holding the steering wheel, the frequency of the driver's sitting posture change and the frequency of the driver's blinking, and the first weight, the second weight and the third weight; wherein the second weight is greater than the third weight, and the third weight is greater than the first weight.

[0019] It can be understood that the scheme provided in the embodiments of the present application: by setting the weight coefficients of different physiological indicators and sorting according to the actual influence degree, the overall state of the driver can be more accurately evaluated, thereby improving the rationality and sensitivity of the target state index. Compared with the calculation method of single index or fixed weight in the related art, the present application dynamically weights and sums, so that the state evaluation is more comprehensive and reliable, which is helpful to find the signs of fatigue earlier and intervene in time.

[0020] In a second aspect, the embodiments of the present application provide a driver's seat adjustment system, comprising: an acquisition module, a determination module and an adjustment module; the acquisition module is configured to acquire the state data of the driver and the current seat parameter of the driver's seat in the current driving scene; the state data is used to represent the action response of the driver in the current driving scene; the determination module is configured to determine a target state index according to the state data of the driver; the target state index is used to represent the adaptation degree of the driver to the current seat parameter; the adjustment module is configured to acquire the historical seat parameter of the same driving scene in the historical vehicle information set in the case that the target state index exceeds a first threshold value; based on the comparison result of the historical seat parameter and the current seat parameter, the driver's seat is adjusted to adapt to the driver.

[0021] In a third aspect, an embodiment of the present application provides a vehicle device, comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, and wherein the processor implements the method described above when executing the program.

[0022] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, wherein the memory stores a computer program capable of running on the processor, and wherein the processor implements the method of the first aspect when executing the program.

[0023] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising a computer program or instructions, wherein the computer program or instructions implement the method of the first aspect when executed. BRIEF DESCRIPTION OF DRAWINGS

[0024] The drawings in the accompanying drawings are incorporated into the specification and constitute a part of the specification, which illustrate embodiments consistent with the present application, and together with the specification serve to explain the technical solutions of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0025] The flowchart shown in the drawings is only an exemplary description, and is not necessarily to include all contents and operations / steps, nor is it necessarily executed in the order described. For example, some operations / steps can be further divided, and some operations / steps can be combined or partially combined, so that the actual execution order can be changed according to the actual situation.

[0026] Figure 1 Flowchart of the adjusting method of the driving seat provided by an embodiment of the present application Figure 1 ; Figure 2 Flowchart of the adjusting method of the driving seat provided by an embodiment of the present application Figure 2 ; Figure 3 Flowchart of the adjusting method of the driving seat provided by an embodiment of the present application Figure 3 ; Figure 4 Flowchart of the adjusting method of the driving seat provided by an embodiment of the present application Figure 4 ; Figure 5 Structure diagram of the adjusting device of the driving seat provided by an embodiment of the present application Figure 6 Structure diagram of the adjusting system of the driving seat provided by an embodiment of the present application Figure 7This is a structural schematic diagram of a vehicle device provided in an embodiment of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the specific technical solutions of this application will be further described in detail below with reference to the accompanying drawings of the embodiments of this application. The following embodiments are used to illustrate this application, but are not intended to limit the scope of this application.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0029] In the following description, references to "some embodiments," "this embodiment," "this application embodiment," and examples, etc., describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subset of all possible embodiments and may be combined with each other without conflict.

[0030] The descriptions such as "first," "second," and "third" appearing in the embodiments of this application do not have a specific meaning (such as no order, nor do they indicate a special limitation on the number of devices in the embodiments of this application), but are merely for the purpose of clearly describing the embodiments of this application and do not constitute any limitation on the embodiments of this application.

[0031] Before providing a more detailed description of the embodiments of this application, the nouns and terms that may be involved in the embodiments of this application will be explained. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0032] To facilitate understanding of the technical solutions of the embodiments of this application, the relevant technologies or terms of the embodiments of this application are described below. The following relevant technologies or terms are optional solutions and can be combined with the technical solutions of the embodiments of this application in any way, and all of them fall within the protection scope of the embodiments of this application.

[0033] Based on the technical problems raised in the background art, the embodiments of this application provide the following methods, devices, equipment and media for adjusting driver's seats, computer program products, etc.

[0034] Figure 1 A flowchart illustrating a method for adjusting a driver's seat provided in this application embodiment. Figure 1 ,like Figure 1 As shown in the embodiment of this application, the method for adjusting a driver's seat includes: S101. Obtain the driver's status data and the current seat parameters of the driver's seat in the current driving scenario.

[0035] Among them, the driver's state data is used to characterize the driver's action response in the current driving scenario.

[0036] The driving scenarios may include, but are not limited to, highways, steep slopes, bumpy roads, muddy roads, snow-capped mountains, deserts, and grasslands.

[0037] In some alternative embodiments, the driver's state data may include at least one of the following: the pressure of the driver's hands on the steering wheel, the frequency of the driver's posture changes, and the frequency of the driver's blinking.

[0038] To improve the accuracy of determining whether adjustment needs to be activated, the following explanation will use three driver status data, including the driver's hand pressure on the steering wheel, the frequency of driver's posture changes, and the frequency of driver's blinking, as examples.

[0039] The pressure exerted by the driver's hands on the steering wheel refers to the force applied by the driver's hands to the steering wheel. For example, this can be achieved by embedding a pressure sensor array beneath the steering wheel surface, with the pressure signal transmitted to the processing unit via an analog-to-digital converter. The pressure exerted by the driver's hands on the steering wheel is used to reflect the driver's operational load (e.g., muscle tension or fatigue level) and to capture stress responses during emergency braking or sudden steering.

[0040] The frequency of driver posture changes refers to the number of times a driver actively adjusts their body posture per unit of time. For example, a pressure distribution sensor on the driver's seat can detect center of gravity shift patterns, and the number of shifts can be counted within a time window. The frequency of driver posture changes characterizes the degree of matching between seat comfort and the driver's body posture, i.e., the degree of fatigue accumulation caused by localized pressure on the body. It is understandable that when seat support is insufficient, the frequency of driver posture changes will increase significantly.

[0041] The driver's blink frequency refers to the number of complete eyelid closures per minute. For example, an infrared camera can be used to capture eye images, and a convolutional neural network can be used to identify the opening and closing states and count the frequency. This driver's blink frequency is used to quantify the driver's visual fatigue or level of inattention; the blink interval will show regular fluctuations when attention decreases.

[0042] In one example, the pressure of the driver's hands gripping the steering wheel, the frequency of the driver's posture changes, and the frequency of the driver's blinks are continuously collected during vehicle operation. Specifically, the steering wheel pressure sensor acquires pressure distribution data at a first fixed sampling rate, and extracts the average pressure as the final result of the driver's hand gripping the steering wheel using a sliding window algorithm; the seat pressure sensor array monitors the pressure changes of the contact surface in real time, and the seat pressure sensor collects data once at a second fixed sampling frequency (e.g., once every 5 seconds). When the center of gravity shift exceeds the center of gravity shift threshold several times in a row (e.g., twice), it is recorded as a valid posture adjustment; the vehicle vision system captures the driver's facial image at a set number of frames per second (e.g., 30 frames), extracts blink action features through edge computing devices, and counts the number of times the eye closure duration exceeds a set eye closure duration (e.g., 100 milliseconds) per minute.

[0043] In another optional embodiment, after obtaining the driver's state data, the driver's state data can be normalized to eliminate the influence of dimensions and improve the accuracy of subsequent calculation of the target state index.

[0044] In one example, the pressure exerted by the driver's hand on the steering wheel is collected by a steering wheel pressure sensor and then normalized to a pressure value within the range of [0,1].

[0045] Compared to traditional seat adjustment methods that rely on only a single type of biometric data, such as judging fatigue status solely through heart rate monitoring and failing to distinguish between physical fatigue and mental stress, the pressure of the driver's hands gripping the steering wheel in this embodiment reflects operational load, the frequency of the driver's posture changes reflects subjective comfort feedback, and the frequency of the driver's blinks reflects visual fatigue. Through these three dimensions of biometric data, it is possible to identify physiological fatigue caused by prolonged driving, detect stress caused by sudden operational demands, and capture the need for posture adjustment due to insufficient seat support, thus forming a multimodal biometric analysis system.

[0046] By acquiring driver status data, the system can identify changes in operational load during sudden driving scenarios based on hand pressure on the steering wheel, determine the seat support and ergonomic fit by analyzing the frequency of posture changes, and accurately assess visual fatigue levels by combining blink frequency. Cross-validation of multi-dimensional biometrics effectively avoids misjudgments that may occur with a single data source, such as misinterpreting high pressure during emergency braking as fatigue. The resulting composite judgment mechanism provides precise physiological basis for adjusting seat parameters, improving body support stability during sudden maneuvers while preventing fatigued driving.

[0047] In some alternative embodiments, the current seat parameters of the driver's seat may include at least one of the following: backrest angle, seat cushion height, fore-aft distance, temperature, and vibration frequency.

[0048] To improve the comfort after adjustment, the following explanation will use five current seat parameters of the driver's seat, including backrest angle, seat cushion height, fore-aft distance, temperature, and vibration frequency, as examples.

[0049] Here, the backrest angle refers to the angle between the driver's seat back and the horizontal plane. For example, an electric push-rod mechanism can be used to drive an angle sensor for measurement and adjustment to accommodate the driver's back support needs.

[0050] The seat cushion height refers to the vertical distance between the bottom of the driver's seat and the vehicle floor. For example, it can be dynamically adjusted using a pneumatic lifting device in conjunction with a pressure sensor to maintain the driver's legs in a natural bent position.

[0051] Here, the fore-and-aft distance refers to the amount of forward and backward movement of the driver's seat slide rail. For example, this can be controlled by a stepper motor-driven slide rail displacement sensor to ensure the driver maintains a safe operating distance from the steering wheel.

[0052] Here, temperature refers to the output temperature value of the heating or ventilation device on the driver's seat surface. For example, a thermocouple sensor can be used in conjunction with a temperature control module to adjust the perceived comfort level.

[0053] The vibration frequency refers to the operating frequency of the shock absorption device built into the driver's seat. For example, after collecting road vibration data through an acceleration sensor, the controller can adjust the output frequency of the electromagnetic shock absorber to achieve dynamic buffering compensation under bumpy road conditions.

[0054] S102. Determine the target state index based on the driver's state data.

[0055] The target state index is used to characterize the driver's degree of adaptation to the current seat parameters and is a comprehensive reflection of the driver's physiological state.

[0056] For example, state data may include the pressure of the driver's hands on the steering wheel, the frequency of the driver's posture changes, and the frequency of the driver's blinks.

[0057] In one possible implementation, the target state index can be directly calculated by continuously collecting state data; in another possible implementation, the target state index can be calculated by continuously collecting state data, obtaining the weights corresponding to each state data, and based on each state data and its corresponding weights.

[0058] The following describes the process of calculating the target state index based on the state data and the corresponding weights.

[0059] For example, in one optional embodiment, a first weight of the pressure of the driver's hand on the steering wheel, a second weight of the frequency of the driver's posture changes, and a third weight of the frequency of the driver's blinking are determined; a target state index is determined based on the pressure of the driver's hand on the steering wheel, the frequency of the driver's posture changes, the frequency of the driver's blinking, and the first, second, and third weights.

[0060] For example, the target state index = first weight × driver's hand pressure on the steering wheel + second weight × frequency of driver's posture changes + third weight × frequency of driver's blinking.

[0061] The three weighting coefficients—first weight, second weight, and third weight—can be preset proportional parameters based on the priority of the impact of each physiological characteristic on driving status. These coefficients can be values ​​set based on historical data analysis or expert experience, used to balance the contribution of different physiological indicators. The sum of the weights of the first, second, and third weights is 1.

[0062] For example, if more emphasis is placed on the matching degree between seat comfort and driver's posture (i.e., the frequency of driver's sitting posture changes), and less emphasis is placed on the driver's visual fatigue or attention distraction (i.e., the frequency of driver's blinking), and the driver's operating load (i.e. the pressure on the driver's hands gripping the steering wheel) is slightly emphasized, then the second weight can be set to be greater than the third weight, and the third weight greater than the first weight.

[0063] Compared to traditional solutions that rely solely on a single physiological indicator, such as steering wheel pressure or blinking frequency, without considering the correlation and weight differences between different physiological characteristics, this application's embodiments effectively avoid mis-triggered adjustment actions by a single indicator through multi-dimensional data fusion (i.e., fusion of the driver's posture change frequency, the driver's blinking frequency, and the driver's hand pressure on the steering wheel) and dynamic weight allocation.

[0064] Understandably, by setting weight coefficients for different physiological indicators and ranking them according to their actual impact, the overall state of the driver can be assessed more accurately, thereby improving the rationality and sensitivity of the target state index. Compared with the calculation methods of single indicators or fixed weights in related technologies, this application uses dynamic weighted summation to make the state assessment more comprehensive and reliable, which helps to detect signs of fatigue earlier and intervene in a timely manner.

[0065] S103. If the target state index exceeds the first threshold, obtain the historical seat parameters under the same driving scenario from the historical vehicle information set.

[0066] The first threshold is a set critical value for the driver's physiological state. For example, it can be determined by establishing a statistical model through biometric data collected in a laboratory at different fatigue levels. The first threshold is the critical value between the normal driving state and the state requiring intervention.

[0067] The calculated target state index is compared with a first threshold, and the decision on whether to trigger seat parameter adjustment is based on the comparison result. For example, if the target state index is less than or equal to the first threshold, it indicates that the driver is in a normal driving state; if the target state index is greater than the first threshold, it indicates that the driver is in a state requiring intervention.

[0068] In one example, during vehicle operation, the frequency of driver posture changes, blinking frequency, and steering wheel grip pressure are continuously monitored, and a target state index is generated in real time through weighted calculation. The target state index is then compared in real time with a first threshold: when the target state index exceeds the first threshold, it indicates that the driver is fatigued or unwell and requires intervention; when the target state index is less than or equal to the first threshold, it indicates that the driver is in a normal driving state and the current driver's seat parameter configuration can be maintained to avoid ineffective adjustments.

[0069] For example, setting the first weight to 0.4, the second weight to 0.3, the third weight to 0.3, and the first threshold to 0.6, a target state index in the interval [0, 1] is generated by linearly superimposing each weight with the corresponding driver's state data. When the target state index exceeds 0.6, it is determined to be a state of fatigue, requiring intervention. It is understood that the driver's state data is standardized before calculation to eliminate the influence of dimensional differences on the weighted result.

[0070] When the target state index exceeds a first threshold and intervention is required, as an optional embodiment, at least one seat parameter can be automatically adjusted directly. For example, the backrest angle can be increased (e.g., adjusted to 110 degrees) to provide lumbar support (e.g., lumbar support pressure can increase by 20% when the backrest angle is adjusted to 110 degrees), or the seat cushion height can be raised to improve visibility, or the seat heating function can be activated to relieve muscle tension, etc. As another optional embodiment, the driver's seat can be adjusted based on a comparison of historical and current seat parameters. This application does not impose any particular limitations on this; the following explanation uses adjusting the driver's seat based on a comparison of historical and current seat parameters as an example.

[0071] In another example, the vibration frequency is dynamically matched based on real-time road condition data; for instance, the damping frequency is automatically increased when continuous bumpy road sections are detected. Furthermore, in addition to automatically increasing the damping frequency, it can also form a coordinated adjustment mechanism with temperature parameters, such as activating seat heating to relieve muscle tension. Dynamic control of vibration frequency and temperature parameters can proactively adapt to complex road condition changes, improving overall comfort and safety during driving.

[0072] Traditional technical solutions rely on single physiological indicators, such as steering wheel pressure or blink frequency, for threshold judgments. This approach fails to consider the correlation and weighting differences between different physiological characteristics, easily leading to misjudgments. For example, judging fatigue solely based on blink frequency exceeding 20 times per minute fails to eliminate misjudgments caused by strong light stimulation. The technical solution provided in this application, however, through multi-dimensional data fusion (i.e., fusing the driver's posture change frequency, blink frequency, and steering wheel pressure) and dynamic weight allocation, avoids mis-triggered adjustment actions by a single indicator and effectively distinguishes between different causes. For instance, in high-speed turning scenarios, it can effectively distinguish whether increased steering wheel grip is due to tension or genuine fatigue.

[0073] The technical solution provided in this application achieves precise matching between seat parameter adjustments and the driver's physiological state. When fatigue characteristics are detected, an adjustment mechanism is triggered, effectively alleviating the driver's physical burden. The comparison mechanism of the first threshold avoids the waste of resources caused by frequent minor adjustments, ensuring that adjustment actions are only initiated when necessary.

[0074] The technical solution provided in this application addresses the technical deficiency of traditional seat adjustment systems in accurately quantifying the driver's physiological state. By comprehensively calculating the weighted average of the driver's posture change frequency, blink frequency, and steering wheel pressure, an objective evaluation index related to the driver's true fatigue level is established, providing reliable physiological data support for dynamic adjustment of seat parameters. In long-distance driving tests, the technical solution provided in this application can respond promptly to adjustment needs while reducing the false trigger rate. For example, it reduces the false trigger rate of seat support adjustment by approximately 30%, and in emergency braking scenarios, it can trigger seat cushion tilt adjustment 50 milliseconds in advance to maintain driver posture stability.

[0075] S104. Based on the comparison results of historical seat parameters and current seat parameters, adjust the driver's seat to suit the driver.

[0076] In one optional embodiment, the driver's seat can be adjusted based on a comparison between the vehicle's historical seat parameters and current seat parameters. In another optional embodiment, the driver's seat can be adjusted based on a comparison between the historical seat parameters and current seat parameters of other types of vehicles. This application does not impose specific limitations on these methods; the following descriptions will use these two approaches as examples.

[0077] In this embodiment, "suitable for the driver" means adapting the adjusted driver's seat to the driver to improve driver comfort and / or reduce driver fatigue.

[0078] When adjusting the driver's seat based on a comparison of the vehicle's historical seat parameters and current seat parameters, S104, for example, includes the following sub-steps: S104A1. Obtain the first duration corresponding to each type of parameter in the historical seat parameters, and obtain the second duration corresponding to each type of parameter in the current seat parameters.

[0079] It should be noted that the historical seat parameters here refer to the historical seat parameters in the historical vehicle information set, while the historical vehicle information set here refers to the historical vehicle information set of this vehicle. In other words, the historical seat parameters here refer to the historical seat parameters of this vehicle.

[0080] For example, historical data that is the same as the current driving scenario of the vehicle can be obtained from a cloud database. This historical data includes the various types of seat parameters and the duration of each type of seat parameter in the same driving scenario as the current driving scenario during the historical driving process.

[0081] To distinguish them, the various types of seat parameters in the same driving scenario as the current driving scenario during historical driving are called historical seat parameters, and the various types of seat parameters in the current driving scenario are called current seat parameters; the duration of each type of parameter in the historical seat parameters is called the first duration; and the duration of each type of parameter in the current seat parameters is called the second duration.

[0082] For example, the various types of parameters included in the historical seat parameters are the optimal seat parameters obtained after screening. These optimal seat parameters are the seat parameters that the driver has continuously used under the same driving conditions, indicating that using these seat parameters can ensure that the driver is in a comfortable driving state and reduce fatigue. There may be only one effective seat parameter, or there may be multiple but not all of them, or all of them, etc. This application does not make any special limitation in this regard. The following is an example of historical seat parameters including all types of seat parameters, and each type of seat parameter is an optimal seat parameter, that is, historical seat parameters include: historical backrest angle, historical seat cushion height, historical fore-aft distance, historical temperature, and historical vibration frequency.

[0083] For example, the process of selecting the optimal seat parameters can involve setting a reference duration threshold for each historical seat parameter. For each historical seat parameter, if its duration exceeds the corresponding reference duration threshold, the historical parameter is considered valid and is the optimal seat parameter. For instance, if the reference duration threshold for historical backrest angle is set to 10 minutes, and the historical backrest angle is 100 degrees with a duration of 3 minutes, since 3 minutes is less than 10 minutes, the historical backrest angle of 100 degrees is not the optimal seat parameter and can be disregarded. However, if the historical backrest angle is 30 degrees with a duration of 15 minutes, since 15 minutes is greater than 10 minutes, the historical backrest angle of 30 degrees is the optimal seat parameter.

[0084] The duration of the duration refers to the continuous time during which the seat parameters remain unchanged under a specific driving condition. This can be achieved using a timer or timestamp difference calculation to verify the stability of parameter adjustments. The reference duration threshold is a pre-set time threshold used to distinguish between valid and invalid parameters. The reference duration threshold can be a fixed value or a variable that is dynamically adjusted according to the driving scenario, such as being set to 10 minutes or adaptively changing according to road condition complexity to filter out short-term interference factors.

[0085] For example, the various types of parameters included in the current seat parameters can be all of them, or they can be the valid seat parameters obtained after filtering. For the valid seat parameters obtained after filtering, there may be one, multiple but not all, or all, etc. This application does not make any special limitation on this. The following example illustrates the situation where all types of seat parameters included in the current seat parameters are valid seat parameters obtained after filtering and include all types of seat parameters. That is, the current seat parameters include: current backrest angle, current seat cushion height, current fore-aft distance, current temperature, and current vibration frequency.

[0086] The process of selecting the optimal seat parameters is similar to that of selecting the optimal seat parameters, and will not be described in detail here.

[0087] It is understood that the first duration of the historical backrest angle, the first duration of the historical seat cushion height, the first duration of the historical front-to-back distance, the first duration of the historical temperature, and the first duration of the historical vibration frequency may all be the same, all be different, or some may be the same and some may be different. This application does not make any special restrictions on this. The following will use the example of all these first durations being different as an example for illustrative purposes.

[0088] It is understood that the second duration of the current backrest angle, the second duration of the current seat cushion height, the second duration of the current front-to-back distance, the second duration of the current temperature, and the second duration of the current vibration frequency may all be the same, all be different, or some may be the same and some may be different. This application does not make any special limitation in this regard. The following will use the example of all these second durations being different as an example for illustrative purposes.

[0089] S104A2, For each first identical type parameter in the historical seat parameters and current seat parameters: take the parameter with the longest duration in each first identical type parameter as the target parameter to be adjusted.

[0090] It is understandable that each first identical type parameter here refers to every identical type of seat parameter in both the historical seat parameters and the current seat parameters.

[0091] For example, the current backrest angle in the current seat parameters and the historical backrest angle in the historical seat parameters can be referred to as the first first type of parameter; the current seat cushion height in the current seat parameters and the historical seat cushion height in the historical seat parameters can be referred to as the second first type of parameter; the current fore-aft distance in the current seat parameters and the historical fore-aft distance in the historical seat parameters can be referred to as the third first type of parameter; the current temperature in the current seat parameters and the historical temperature in the historical seat parameters can be referred to as the fourth first type of parameter; and the current vibration frequency in the current seat parameters and the historical vibration frequency in the historical seat parameters can be referred to as the fifth first type of parameter.

[0092] Among all parameters of the same type, the parameter with the longest duration is selected as the target parameter to be adjusted. For example, the second duration of the current seat parameter and the first duration of the historical backrest angle are compared. The parameter with the longest duration between the first duration of the historical backrest angle and the second duration of the current seat parameter is selected as the target backrest angle parameter to be adjusted. If the first duration and the second duration are the same or the first duration is less than the second duration, the backrest angle of the driver's seat in the current driving scenario remains unchanged. For example, if the first duration of the historical backrest angle is longer than the second duration of the current seat parameter, the historical backrest angle is selected as the target parameter to be adjusted for the backrest angle of the driver's seat in the current driving scenario. Or, if the first duration of the historical backrest angle is less than or equal to the second duration of the current seat parameter, the current backrest angle is selected as the target parameter to be adjusted for the backrest angle of the driver's seat in the current driving scenario.

[0093] S104A3, Adjust the driver's seat according to at least one target parameter to be adjusted.

[0094] In one optional embodiment, the duration of a first first-same-type parameter can be compared among multiple first-same-type parameters, and then the target parameter to be adjusted determined by the first-same-type parameter can be used to adjust the driver's seat. In another optional embodiment, the duration of each first-same-type parameter can be compared, and then the target parameter to be adjusted among the determined first-same-type parameters can be used to adjust the driver's seat. This application does not make any special or limited provisions in this regard, and the following description will use the adjustment of the driver's seat according to each target parameter to be adjusted as an example.

[0095] It is understandable that if the duration of each parameter in the current seat parameters is shorter than the duration of each parameter in the historical seat parameters, then adjusting the driver's seat using the parameters in the historical seat parameters, which are all optimal seat parameters obtained after filtering, can make the driver more comfortable.

[0096] Understandably, if the duration of all parameters in the current seat parameters is greater than the duration of all parameters in the historical seat parameters, since all parameters in the historical seat parameters are the optimal seat parameters obtained after screening, it means that the current seat parameters are already the seat parameters that make the driver more comfortable, and there is no need to adjust them temporarily.

[0097] Understandably, if the duration of some parameters in the current seat parameters is greater than or equal to the duration of the corresponding parameters in the historical seat parameters, and the duration of the remaining parameters in the current seat parameters is less than the duration of the corresponding parameters in the historical seat parameters, then the driver's seat can be adjusted using the parameters whose duration is greater than or equal to the duration of the corresponding parameters in the historical seat parameters, along with the remaining parameters in the historical seat parameters, to make the driver more comfortable.

[0098] Understandably, by obtaining the duration of each type of parameter in the historical and current seat parameters respectively, and selecting the parameter with the longest duration as the target parameter to be adjusted, adjusting the driver's seat according to at least one target parameter can avoid misjudgment or interference caused by short-term adjustments, and can also ensure that the adjusted seat parameters have higher stability and adaptability, further improving the adjustment effect.

[0099] When adjusting the driver's seat based on a comparison of historical seat parameters from other types of vehicles with current seat parameters, S104, exemplarily, includes the following sub-steps: S104B1. Obtain a collection of historical vehicle information for various types of vehicles under the same driving scenario.

[0100] For example, the cloud database stores historical vehicle information sets of various types of vehicles under different driving scenarios, and historical vehicle information sets of various types of vehicles under the same driving scenario can be obtained from the cloud database.

[0101] In one optional embodiment, the historical vehicle information set for each type of vehicle includes the set of driving feature vectors of that type of vehicle in the same driving scenario (hereinafter referred to as the type vehicle feature vector set) and the seat parameters of that type of vehicle in the same driving scenario and the duration of each seat parameter; the feature vector set for each type of vehicle includes one or more feature driving vectors.

[0102] It should be noted that, for the seat parameters in the historical vehicle information set for each type of vehicle, these seat parameters are all the optimal seat parameters after filtering.

[0103] For example, the sets of driving feature vectors and seat parameter datasets for various types of vehicles recorded in the cloud database can be implemented using a distributed database, expanding the feature coverage through multi-source data aggregation. The historical vehicle information set refers to the mapping relationship between driving features and seat parameters stored according to driving status. Specifically, it can be implemented using clustering analysis algorithms. The diversity of the parameter database is enhanced by dynamically updating the set content. In other words, a historical vehicle information set includes all driving feature vectors and seat parameters for a certain driving scenario.

[0104] In one example, the cloud database periodically synchronizes the driving feature vectors and corresponding seat parameters uploaded by each vehicle. After removing outliers through data cleaning, the feature vectors of the same type of vehicle under the same driving state are grouped into the same set.

[0105] Compared to traditional seat parameter optimization solutions that rely on local data from a single vehicle, this approach cannot overcome the limitations of individual usage habits. In other words, if the duration of the current seat parameters in a driving scenario is greater than or equal to the duration of the vehicle's historical seat parameters, although the adjustment mechanism has been triggered, no adjustment is made because the current seat parameters are superior to the historical ones. Therefore, over time, even if the driver is uncomfortable or fatigued, the inability to adjust leads to increased discomfort or fatigue without relief. To prevent this, this application proposes integrating multi-vehicle data in the cloud. This allows the feature vector set to cover more driving scenarios and user groups, possessing cross-vehicle and cross-user generalization capabilities, thus solving the problem of insufficient personalized adaptation caused by data isolation in traditional systems. For example, it can simultaneously collect bumpy road condition data from both SUVs and sedans to form a more comprehensive parameter recommendation library.

[0106] S104B2. Select the target information set from the historical vehicle information sets of various types of vehicles based on the current driving scenario.

[0107] The target information set is the set of historical vehicle information that best matches the current driving scenario.

[0108] It is understandable that while the historical vehicle information sets of various vehicle types obtained from the cloud database are the same as those used in the current driving scenario, there are many such sets in the vehicle-side database. Theoretically, one could arbitrarily select a set of historical vehicle information for each type to adjust the driver's seat. However, this might not ensure the adjusted seat is in the most comfortable position for the driver. Therefore, in one alternative embodiment, the set of historical vehicle information that best matches the current driving scenario can be selected as the target set. The driver's seat can then be adjusted based on a comparison between the seat parameters in the target set and the current seat parameters, allowing the driver to achieve the most comfortable driving position and reducing fatigue caused by long-distance driving.

[0109] Specifically, selecting the target information set from the historical vehicle information sets of various types of vehicles based on the current driving scenario includes the following sub-steps: S104B2a: Determine the driving feature vector under the current driving state based on the vehicle's driving data in the current driving scenario.

[0110] Among them, the driving feature vector refers to a multi-dimensional data set after the driving data has been standardized and processed, which is used to comprehensively represent the current driving status of the vehicle.

[0111] In one alternative embodiment, driving data for the current driving scenario can be collected at a third fixed sampling frequency (e.g., once per second).

[0112] In one optional embodiment, driving data may include driving time, driving speed, ambient temperature, light intensity, whether turning, whether accelerating or decelerating, and road conditions, etc., all of which can be collected in real time by onboard sensors.

[0113] In one optional embodiment, the collected driving data can be directly normalized; in another optional embodiment, the collected driving data can be first classified, and then different normalization methods can be applied according to different classifications. Normalization refers to the process of converting data from different ranges into a unified standard, which can be achieved through linear transformation methods, making different data types comparable in feature vectors.

[0114] For example, driving data can be divided into numerical data and distributed data.

[0115] Numerical data refers to driving state parameters represented by continuous numerical values. Examples include driving duration, driving speed, ambient temperature, and light intensity.

[0116] Categorical data refers to driving events or environmental conditions characterized by discrete states. For example, it may include whether a turn is made, whether acceleration or deceleration is performed, and road conditions.

[0117] For example, the numerical data can be normalized by a maximum-minimum method. For instance, the maximum-minimum normalization method can be used to map numerical data of different dimensions, such as driving speed and ambient temperature, to the [0, 1] interval, thus eliminating the influence of the difference in dimensions on vector construction.

[0118] For example, categorical data can be converted into discrete integer values ​​and then normalized. For instance, integer encoding can be used to convert categorical data such as whether a turn is possible into 0 or 1, and then normalization can be performed to bring it to the same order of magnitude as numerical data, thus avoiding interference from discrete values ​​on the vector space distribution.

[0119] In one optional embodiment, the normalized data can be arranged and combined according to a predetermined dimension to form a feature vector representing the current driving state. For example, feature data representing driving duration can be normalized and stored in one area.

[0120] In one example, during the construction of the driving feature vector, numerical data such as driving time and speed are first normalized. For instance, driving speed is divided by a preset maximum speed threshold to obtain dimensionless standardized values. For categorical data such as whether a turn is possible or whether the road conditions are slippery, these are first converted to binary integers, for example, mapping "yes" to 1 and "no" to 0. Then, normalization is applied to ensure that discrete and continuous values ​​are evenly distributed in the vector space. The resulting driving feature vector can simultaneously contain continuous driving parameters and discrete event states, forming a multi-dimensional state representation space, thus solving the problem of incomplete feature information caused by data type separation in traditional methods.

[0121] Compared to related technologies, existing seat control methods do not distinguish data types when processing driving data, directly splicing the raw data, leading to conflicts in magnitude and distribution between continuous parameters and discrete states. The technical solution provided in this application, however, achieves effective fusion of numerical and categorical data through data type identification and difference processing. This allows the constructed feature vector to reflect both continuously changing parameters such as driving speed and ambient temperature, and accurately characterize discrete driving events such as acceleration, deceleration, and turning, thus improving the accuracy of driving state judgment.

[0122] The technical solution provided in this application addresses the problem of incomplete feature construction caused by improper data type processing in traditional seat control. Through differentiated data preprocessing methods, it effectively constructs multi-dimensional driving feature vectors. This solution retains the continuous variation characteristics of numerical data while accurately representing the discrete state characteristics of categorical data, providing a reliable data foundation for subsequent accurate classification of driving states and matching of seat parameters. It avoids misjudgments caused by missing or skewed feature vector information.

[0123] S104B2b: For each type of vehicle in the historical vehicle information set, determine the average distance between the driving feature vector in the current driving state and multiple driving feature vectors in the type vehicle feature vector set in the historical vehicle information set of the type vehicle; determine the type vehicle feature vector set corresponding to the average distance being less than the first set distance threshold as the matching feature vector set.

[0124] It is understandable that, according to the relevant description in S104B1 above: "The cloud database stores historical vehicle information sets of various types of vehicles under different driving scenarios, and historical vehicle information sets of various types of vehicles under the same driving scenario can be obtained from the cloud database. In an optional embodiment, for each type of vehicle's historical vehicle information set, it includes the vehicle's own driving feature vector set under the same driving scenario (hereinafter referred to as the type vehicle feature vector set) and the vehicle's own seat parameters under the same driving scenario; for each type vehicle feature vector set, it includes one or more feature driving vectors." It can be seen that each type of vehicle's historical vehicle information set includes a type vehicle feature vector set and seat parameters, and each type vehicle feature vector set includes one or more driving feature vectors.

[0125] Therefore, in order to find the set of historical vehicle information that best matches the current driving scenario (i.e., the target information set), it is necessary to calculate the average distance between the driving feature vector in the current driving scenario and the feature vector set of each type of vehicle.

[0126] Taking a set of feature vectors for a certain type of vehicle as an example, it is necessary to calculate the distance between the driving feature vector in the current driving scenario and each driving feature vector in the set of feature vectors for that type of vehicle. After calculating the distance between the driving feature vector in the previous driving scenario and each driving feature vector in the set of feature vectors for that type of vehicle, the average value is calculated to obtain an average value. This average value is used as the average distance between the driving feature vector in the current driving scenario and the set of feature vectors for that type of vehicle.

[0127] For example, an averaging algorithm based on Euclidean distance can be used to calculate the average distance between the driving feature vector in the current driving scenario and the feature vector set of each vehicle type. The average Euclidean distance refers to the mean distance between the current driving feature vector and all driving feature vectors in a certain type of vehicle feature vector set. For instance, the root mean square difference of the squared differences of vector dimensions can be used to quantify the overall similarity between the current driving scenario and the historical vehicle information set.

[0128] Following this method, after traversing the feature vector set of each type of vehicle, several average distances are obtained. By comparing the average distances with a first set distance threshold, the matching feature vector set is determined.

[0129] For example, the first set distance threshold can be 0.3. Continuing with an example of a set of feature vectors for a certain type of vehicle, if the average distance between the driving feature vectors in the current driving scenario and the set of feature vectors for that type of vehicle is less than 0.3, it indicates that the current driving scenario is highly similar to the typical driving scenario recorded in the set of feature vectors for that type of vehicle, and the set of feature vectors for that type of vehicle is marked as a matching feature vector set. If the average distance between the driving feature vectors in the current driving scenario and the set of feature vectors for that type of vehicle is greater than or equal to 0.3, it indicates that the current driving scenario is not similar to the typical driving scenario recorded in the set of feature vectors for that type of vehicle, and the set of feature vectors for that type of vehicle will not be marked or will be removed from the set of feature vectors for that type of vehicle.

[0130] Compared to traditional solutions that rely solely on comparing a single feature vector with preset parameters for adjustment without considering the overall distribution characteristics of historical data, the technical solution provided in this application effectively avoids misjudgments caused by fluctuations in individual data by calculating the overall similarity between the driving feature vector in the current driving scenario and the feature vector sets of various types of vehicles.

[0131] S104B2c: Determine the set of historical vehicle information corresponding to the set of matching feature vectors as the target information set.

[0132] It is understandable that the matching feature vector set is selected from several types of vehicle feature vector sets, and it is essentially still a type of vehicle feature vector set. The number of matching feature vector sets selected may be one or more, and this application does not make any special limitation on this.

[0133] It is understandable that a set of feature vectors for a vehicle type corresponds to a set of historical vehicle information for that type. Since the selected set of matching vectors is essentially still a set of feature vectors for the vehicle type, a set of matching vectors corresponds to a set of historical vehicle information.

[0134] It is understandable that since the number of selected matching vector sets may be one, two or more, the corresponding set of historical vehicle information may also be one, two or more. Therefore, obtaining the matching feature vector set is essentially to filter the set of historical vehicle information, and the filtered set of historical vehicle information is called the target information set.

[0135] It is understandable that, since the number of historical vehicle information sets selected may be one, two, or more, the number of target information sets may also be one, two, or more.

[0136] It is understandable that the target information set is essentially a set of historical vehicle information, a set of historical vehicle information that has been filtered out.

[0137] It is understood that the solution provided in this application is as follows: the historical vehicle information set for each type of vehicle includes a set of feature vectors for that type of vehicle. Each type of vehicle feature set includes one or more driving feature vectors. By constructing and comparing the average distance between the driving feature vectors and multiple driving feature vectors in each type of vehicle feature vector set, the similarity between the current driving scenario and the historical driving scenario can be quantified. This allows for the precise selection of matching feature vector sets, that is, accurately selecting the vehicle with the closest driving scenario among different types of vehicles in the same driving scenario. Then, based on the comparison results between the seat parameters of the vehicle with the closest driving scenario and the seat parameters of this vehicle, the driver's seat is adjusted. This method is more flexible than traditional fixed threshold classification methods, effectively copes with complex or sudden changes in driving scenarios, and improves the accuracy and robustness of seat parameter matching.

[0138] S104B3. Adjust the driver's seat based on the comparison results between the seat parameters in the target information set and the current seat parameters.

[0139] In one optional embodiment, after obtaining the target information set, the driver's seat can be adjusted directly based on the comparison result between the seat parameters in the target information set and the current seat parameters. In another optional embodiment, after obtaining the target information set, if new changes in seat parameters occur, the seat parameters in the target information set can be updated as appropriate, and the driver's seat can be adjusted based on the comparison result between the updated seat parameters in the target information set and the current seat parameters. This application does not impose any particular limitation on this method. Whether the driver's seat is adjusted based on the comparison result between the seat parameters in the target information set and the current seat parameters, or based on the comparison result between the updated seat parameters in the target information set and the current seat parameters, the method for adjusting the driver's seat is the same. The following description will use the example of adjusting the driver's seat based on the comparison result between the seat parameters in the target information set and the current seat parameters for illustrative purposes.

[0140] In this embodiment of the application, after obtaining the target information set, if a new change in the seat parameters occurs, the seat parameters in the target information set can be updated as appropriate, including: firstly, comparing the duration of the new seat parameter with the duration of the corresponding set seat parameter; if the duration of the new seat parameter is greater than the duration of the corresponding set seat parameter, the new seat parameter is considered valid; if the duration of the new seat parameter is less than or equal to the duration of the corresponding set seat parameter, the new seat parameter is considered invalid.

[0141] Then, if the duration of the new seat parameter is greater than the duration of the corresponding set seat parameter, it is compared with the duration of the corresponding old seat parameter in the target information set. If the duration of the new seat parameter is greater than the duration of the old seat parameter, the old seat parameter is updated to the new seat parameter; if the duration of the new seat parameter is less than or equal to the duration of the old seat parameter, the old seat parameter is not updated.

[0142] In simple terms, when a new seat parameter appears in a target information set, the system records the continuous usage time of that parameter in that state and compares it with the duration of the corresponding existing optimal seat parameter within that target information set. If the duration of the new parameter exceeds that of the original optimal parameter, a parameter replacement rule is triggered, and the new parameter is used as the recommended value for that feature vector set. For example, during highway cruising, if a new backrest angle parameter is used continuously by multiple vehicles for more than the 300-minute threshold of the original backrest angle parameter, the new backrest angle is automatically updated to the optimal seat parameter.

[0143] Compared to the traditional method of using a fixed threshold to determine parameter validity, the technical solution provided in this application uses a continuous dynamic competition mechanism to enable the parameter library to autonomously eliminate outdated configurations and absorb the optimal solution from the group. This continuous dynamic competition mechanism ensures that parameter updates retain historically valid configurations while also promptly integrating new solutions validated in group driving behavior, such as gradually optimizing seat temperature adjustment strategies in long-distance driving scenarios.

[0144] When the number of matching feature vector sets is one, the corresponding set of historical vehicle information is also one, meaning the target information set is one. In this case, adjusting the driver's seat based on the comparison between the seat parameters in the target information set and the current seat parameters includes: S104B3a1. For each second type parameter of the same type in the target information set and the current seat parameters, the type parameter with the longest duration in each second type parameter of the same type is taken as the target parameter to be adjusted.

[0145] It is understandable that the target information set is essentially a historical vehicle information set. As mentioned above, the historical vehicle information set includes the vehicle feature vector set of this type of vehicle in the same driving scenario and the seat parameters of this type of vehicle in the same driving scenario, as well as the duration of each seat parameter. For the seat parameters in the historical vehicle information set of each type of vehicle, these seat parameters are the optimal seat parameters after filtering.

[0146] In one example, the seat parameters and the third duration corresponding to each type of parameter in the target information set are obtained, and the second duration corresponding to each type of parameter in the current seat parameters is obtained.

[0147] To distinguish them, the various types of seat parameters in the target information set are referred to as candidate seat parameters, and the various types of seat parameters in the current driving scenario are referred to as current seat parameters; the duration of each type of parameter in the candidate seat parameters is referred to as the third duration; and the duration of each type of parameter in the current seat parameters is referred to as the second duration.

[0148] It is understood that the third duration of the alternative backrest angle, the third duration of the alternative seat cushion height, the third duration of the alternative fore-aft distance, the third duration of the alternative temperature, and the third duration of the alternative vibration frequency may all be the same, all be different, or some may be the same and some may be different. This application does not make any special restrictions on this. The following will use the example of all these third durations being different to illustrate the point.

[0149] It is understood that each second identical type parameter here refers to each identical type of seat parameter among the alternative seat parameters and the current seat parameters.

[0150] For example, the current backrest angle in the current seat parameters and the alternative backrest angle in the alternative seat parameters can be referred to as the first second type of parameter; the current seat cushion height in the current seat parameters and the alternative seat cushion height in the alternative seat parameters can be referred to as the second second type of parameter; the current fore-aft distance in the current seat parameters and the alternative fore-aft distance in the alternative seat parameters can be referred to as the third second type of parameter; the current temperature in the current seat parameters and the alternative temperature in the alternative seat parameters can be referred to as the fourth second type of parameter; and the current vibration frequency in the current seat parameters and the alternative vibration frequency in the alternative seat parameters can be referred to as the fifth second type of parameter.

[0151] The parameter with the longest duration among all second-type parameters is selected as the target parameter to be adjusted. For example, the second duration of the current seat parameter and the third duration of the alternative backrest angles are compared. The parameter with the longest duration between the third duration of the alternative backrest angles and the second duration of the current seat parameter is selected as the target backrest angle parameter to be adjusted. If the third duration is the same as or less than the second duration, the backrest angle of the driver's seat in the current driving scenario remains unchanged. For example, if the third duration of the alternative backrest angle is longer than the second duration of the current seat parameter, the alternative backrest angle is selected as the target parameter to be adjusted for the backrest angle of the driver's seat in the current driving scenario. Alternatively, if the third duration of the alternative backrest angle is less than or equal to the second duration of the current seat parameter, the current backrest angle is selected as the target parameter to be adjusted for the backrest angle of the driver's seat in the current driving scenario.

[0152] S104B3a2, Adjust the driver's seat according to at least one target parameter to be adjusted.

[0153] In one optional embodiment, the duration of a first second-type parameter can be compared among a plurality of second-type parameters of the same type, and then the target parameter to be adjusted determined by the second-type parameter can be used to adjust the driver's seat. In another optional embodiment, the duration of each second-type parameter can be compared, and then the target parameter to be adjusted among the determined second-type parameters can be used to adjust the driver's seat. This application does not impose any particular limitation on this method, and the following description will use the adjustment of the driver's seat according to each target parameter to be adjusted as an example.

[0154] Understandably, if the duration of each parameter in the current seat parameters is shorter than the duration of each parameter in the alternative seat parameters, then adjusting the driver's seat using the parameters in the alternative seat parameters, which are all optimal seat parameters obtained after screening, can make the driver more comfortable.

[0155] Understandably, if the duration of some parameters in the current seat parameters is greater than or equal to the duration of the corresponding parameters in the alternative seat parameters, and the duration of the remaining parameters in the current seat parameters is less than the duration of the corresponding parameters in the alternative seat parameters, then the driver's seat can be adjusted using the parameters whose duration is greater than or equal to the duration of the corresponding parameters in the alternative seat parameters, along with the remaining parameters in the alternative seat parameters, to make the driver more comfortable.

[0156] When the number of matching feature vector sets is two or more, and the corresponding historical vehicle information sets are also two or more (i.e., the target information sets are two or more), then, based on the comparison between the seat parameters in the target information sets and the current seat parameters, adjusting the driver's seat includes: S104B3b1. Select the best seat parameters from several sets of target information to form the best seat parameter set; In one example, among multiple sets of target information, the duration of the same seat parameter in each set is compared. The parameter with the longest duration for each type of identical seat parameter is marked as the optimal parameter. The set of all optimal parameters is called the best seat parameter set. For example, when a vehicle enters a series of curves, in this driving scenario, there are two sets of target information. One set has a backrest angle parameter of 45 degrees for 3 hours, and the other set has a backrest angle parameter of 40 degrees for 5 hours. The 40-degree backrest angle parameter is then marked as the best seat parameter for the backrest angle.

[0157] In this way, the best seat parameters can be selected from each set of identical seat parameters, forming the optimal set of seat parameters.

[0158] It is understandable that, since the seat parameters in the target information set are all selected optimal seat parameters, there may be situations where the number of seat parameters in multiple target information sets is different. In an optional embodiment, if the number of seat parameters in the target information sets is different, no seat parameter of the same type can be directly marked as the optimal seat parameter.

[0159] S104B3b2: Adjust the driver's seat based on the comparison between the optimal set of seat parameters and the current seat parameters.

[0160] In one example, the duration of each seat parameter in the optimal set of seat parameters is compared with the duration of the corresponding seat parameter in the set of seat parameters, and the seat parameter with the longest duration is selected to adjust the driver's seat.

[0161] Understandably, the parameter optimization mechanism based on duration overcomes the limitations of traditional methods that rely solely on instantaneous data or fixed rules, thereby enhancing the stability and adaptability of parameter configuration.

[0162] The technical solution provided in this application can effectively solve the problem of mismatched seat parameters in dynamic scenarios, and achieve accurate mapping between driving scenarios and seat configurations. Specifically, it improves the accuracy of complex road condition recognition by about 30% through dynamic matching of multi-dimensional feature vectors; improves the reliability of adjustment decisions by about 25% based on a parameter optimization mechanism with continuous duration; and shortens the seat parameter adaptation response time to less than 0.5 seconds through real-time comparison of cloud feature sets.

[0163] Understandably, this application further refines the logic for selecting parameters to be adjusted from the target feature vector set and target information set. Specifically, it prioritizes parameters with the longest duration among parameters of the same type as the basis for adjustment, making the adjustment action more rational and stable. Compared to adjustment methods based solely on single-point-of-time data or empirical rules, this application improves the scientific rigor of parameter selection and reduces the occurrence of ineffective adjustments through duration analysis.

[0164] It is understood that the solution provided in this application, by introducing historical vehicle information sets of various types of vehicles and filtering out the most matching target information set based on the current driving scenario, can expand the data sources for seat parameter optimization and improve the generalization ability of recommended parameters. Compared with traditional methods that rely solely on local data from a single vehicle, this application can achieve seat parameter optimization across vehicle models and users, making the adjustment strategy more adaptable and further improving the adjustment effect.

[0165] Figure 2 A flowchart illustrating a method for adjusting a driver's seat provided in this application embodiment. Figure 2 ,like Figure 2 As shown, as an optional embodiment, the driver's seat adjustment method provided in this application further includes, before adjusting the driver's seat based on the comparison results of historical seat parameters and current seat parameters, filtering out the valid seat parameters from the current seat parameters.

[0166] Specifically, filtering out the valid seat parameters from the current seat parameters includes the following steps: S105. For each parameter in the current seat parameters, obtain the duration of each parameter.

[0167] S106. Parameters whose duration exceeds the set duration are determined as valid parameters (i.e., valid seat parameters).

[0168] Valid parameters refer to seat parameters that have been verified after a set duration. For example, these parameters can be obtained through a duration statistics module to build a stable and reliable parameter database. For each parameter, a set duration is assigned, and seat parameters whose duration exceeds the corresponding set duration are determined as valid parameters.

[0169] In one example, the validity determination of seat parameters is implemented through a time-based verification mechanism. The duration of each type of seat parameter in the current configuration is recorded and compared with its corresponding set duration. Taking a specific seat parameter as an example: if the duration of the seat parameter exceeds the corresponding set duration, it is determined to be a valid configuration under driver adaptation conditions; if the duration of the seat parameter is less than or equal to the corresponding set duration, it is considered invalid data generated by temporary adjustments. If a seat parameter has multiple valid parameters, the parameter with the longest duration is selected as the baseline configuration.

[0170] Understandably, when selecting valid seat parameters from the current seat parameters, S104 adapts from adjusting the driver's seat based on the comparison results of historical seat parameters and current seat parameters to alleviate driver fatigue, to adjusting the driver's seat based on the comparison results of valid parameters from historical seat parameters and current seat parameters to improve driver's seat comfort and alleviate driver discomfort and / or fatigue.

[0171] It is understood that the solution provided in this application, by setting different durations for different parameters and filtering out valid parameters accordingly, can filter out temporary adjustments that have not been sufficiently verified, thereby ensuring that subsequent adjustments are based on more reliable parameters. Compared to the lack of parameter validity verification mechanisms in related technologies, this application enhances the credibility and stability of adjustment decisions by introducing a duration-based filtering mechanism.

[0172] The following section details the implementation of the historical vehicle information set and the implementation of historical seat parameters under the same driving scenario within the historical vehicle information set.

[0173] It should be noted that both the historical vehicle information set of this vehicle and the historical vehicle information set of various types of vehicles include a set of driving feature vectors and the seat parameters corresponding to the driving feature vector sets, as well as the duration of each seat parameter; each feature vector set corresponds to a driving scenario; a driving feature vector set includes one or more driving feature vectors.

[0174] As an optional embodiment, the driver's seat adjustment method provided in this application further includes the following steps: S201. Perform feature processing on each driving data in the multiple driving data of the vehicle to obtain multiple driving feature vectors.

[0175] For a detailed description of “driving data”, please refer to the relevant description in S104B2a above, which will not be repeated here.

[0176] In one example, driving data is characterized to obtain a driving feature vector. For example, driving data is normalized to obtain a driving feature vector. As mentioned earlier, the driving data needs to be classified first, and then a suitable normalization method needs to be matched for each type of driving data.

[0177] In one optional embodiment, vehicle driving data can be collected at a third fixed sampling frequency (e.g., once per second). Since the vehicle driving data is collected continuously for several times, the driving data obtained each time is normalized to obtain the corresponding driving feature vector, that is, there are multiple driving feature vectors.

[0178] S202. Classify the driving feature vectors from multiple driving scenarios to obtain a set of driving feature vectors for multiple driving scenarios.

[0179] The first obtained driving feature vector is marked as the reference vector, and the subsequent obtained driving feature vectors are marked as comparison vectors. The Euclidean distance between the comparison vector and the reference vector is calculated sequentially and compared with a second set distance threshold. If the distance does not exceed the second set distance threshold, the comparison vector and the reference vector constitute a feature vector set. If the distance threshold is exceeded, the reference vector is updated to the current comparison vector. The Euclidean distance between the subsequent comparison vector and the updated reference vector is calculated and compared with the second set distance threshold. This process is repeated to obtain multiple sets of driving feature vectors.

[0180] The reference vector refers to the baseline feature vector acquired for the first time. Specifically, the driving feature vector calculated at the initial moment can be used as the baseline to provide a reference standard for subsequent state comparisons.

[0181] Euclidean distance refers to a measure of the distance between vectors in a multidimensional space, used to quantify the degree of change in driving state.

[0182] It is understandable that one set of driving feature vectors corresponds to one driving scenario, so multiple sets of driving feature vectors represent multiple driving scenarios.

[0183] For example, taking five sets of driving data as an example, the feature driving vector obtained from the first set of driving data is recorded as a1, the feature driving vector obtained from the second set of driving data is recorded as b1, the feature driving vector obtained from the third set of driving data is recorded as b2, the feature driving vector obtained from the fourth set of driving data is recorded as b3, and the feature driving vector obtained from the fifth set of driving data is recorded as b4.

[0184] For example, a1 is regarded as the reference vector, and b1, b2, b3 and b4 are regarded as comparison vectors.

[0185] For example, the Euclidean distance between b1 and a1 is calculated first and compared with a second set distance threshold. If the Euclidean distance between b1 and a1 is less than the second set distance threshold, b1 and a1 are merged to obtain the first driving feature vector set A1, which can be represented by A1 = {a1, b1}, indicating that a1 and b1 belong to the same driving scenario. If the Euclidean distance between b1 and a1 is greater than or equal to the second set distance threshold, a1 is updated to b1. The a1 before the update then forms a driving feature vector set A1, which can be represented by A1 = {a1}.

[0186] For example, taking the calculation of the Euclidean distance between comparison vector b1 and reference vector a1 as an example, the formula for calculating the Euclidean distance between comparison vector b1 and reference vector a1 can be referred to as the following formula (1): Formula (1); In formula (1), d(a1, b1) represents the Euclidean distance between comparison vector b1 and reference vector a1; n essentially represents the dimension of the driving feature vector, that is, the number of driving data in the driving feature vector. Then a1n means that the reference vector a1 has n dimensions, that is, n driving data; b1n means that the comparison vector b1 has n dimensions, that is, n driving data.

[0187] Here, Euclidean distance refers to the straight-line distance between two vectors in n-dimensional space, used to quantify the degree of difference between driving feature vectors. Vector dimension n refers to the number of categories of normalized driving data (i.e., driving time, driving speed, ambient temperature, light intensity, whether turning, whether accelerating or decelerating, and road conditions, etc.), which is achieved by using the total number of items after normalizing numerical and categorical data to ensure that feature values ​​of different data types are comparable in the same dimensional space.

[0188] Specifically, when constructing driving feature vectors, numerical data is normalized, and categorical data, after being mapped to integers, is also normalized to form vectors of uniform dimensions. When calculating the Euclidean distance between the comparison vector and the reference vector, the square root of the sum of the squared differences in each dimension is taken to obtain the geometric distance between the two vectors. If the distance does not exceed a preset threshold, it indicates that the current driving scenario is consistent with the state of the reference vector, and they are classified into the same feature vector set; if the distance exceeds the threshold, it is determined to be a new state, and the reference vector is updated. Through continuous calculation and comparison, the feature vector sets of different driving states are accurately divided, avoiding classification errors caused by data dimension confusion or lack of normalization.

[0189] Before calculating the Euclidean distance between b2 and a1 for the second time, it is necessary to determine whether a1 has been updated. If a1 has not been updated, the Euclidean distance between b2 and a1 is calculated and compared with the second set distance threshold. If a1 has been updated to b1, the Euclidean distance between b2 and a1 is calculated, which is essentially the calculation of the Euclidean distance between b2 and b1.

[0190] For example, assuming a1 is not updated, if the Euclidean distance between b2 and a1 is less than the second set distance threshold, then b2 and a1 are merged, resulting in a feature vector set A1 = {a1, b1, b2}, which means that a1, b1, and b2 belong to the same driving scenario. If the Euclidean distance between b2 and a1 is greater than or equal to the second set distance threshold, then a1 is updated to b2. In this case, the original a1 and b1 belong to the same driving scenario, which can be represented by A1 = {a1, b1}.

[0191] For example, assuming a1 is updated to b1, the first driving feature vector set A1 = {a1}; if the Euclidean distance between b2 and b1 is less than a second set distance threshold, then b2 and b1 are merged to obtain the second driving feature vector set A2 = {b1, b2}, indicating that b1 and b2 are driving feature vectors under the same driving scenario; if the Euclidean distance between b2 and b1 is greater than or equal to the second set distance threshold, then a1 is updated to b2, and at this time, two feature vector sets are obtained, namely A1 = {a1} and A2 = {b1}.

[0192] Following this method, repeated calculations may yield a set of eigenvectors A1, or several sets of eigenvectors A1, A2, A3, ..., Aτ. This application does not impose any particular limitations on this; the following explanation will use the example of obtaining several sets of eigenvectors A1, A2, A3, ..., Aτ.

[0193] Where τ represents the number of sets of driving feature vectors.

[0194] Compared to related technologies, traditional seat adjustment methods do not explicitly employ the Euclidean distance formula when classifying states, or they only use other metrics such as Manhattan distance and cosine similarity, failing to consider the geometric spatial relationships of normalized multidimensional data. For example, some related technologies do not perform integer mapping and normalization on the classified data, leading to mismatched vector dimensions or inconsistent data distribution, affecting the reliability of distance calculation. The technical solution provided in this application, by defining a mathematical formula for Euclidean distance and combining it with normalization processing, ensures that feature data from different sources are comparable within the same space, improving the accuracy of state classification.

[0195] The technical solution provided in this application addresses the problem of inaccurate state judgment caused by incomplete construction of driving feature vectors, achieving accurate classification of driving scenarios. By using a unified Euclidean distance calculation rule, the differences between the multidimensional vectors resulting from the fusion of numerical and categorical data are quantified, avoiding dimensional mismatch issues caused by mixed data types. The normalized feature vectors are compared in the same geometric space, enabling reliable differentiation of different driving states and providing an accurate state classification basis for subsequent seat parameter matching.

[0196] In summary, the technical solution provided in this application overcomes the limitations of static feature classification. Traditional methods use fixed thresholds to classify driving states, which cannot adapt to continuously changing driving scenarios. By employing a dynamic Euclidean distance comparison mechanism, automatic clustering and updating of driving states are achieved. While related technologies often employ single-type data processing methods, the technical solution provided in this application creatively integrates numerical and categorical data to construct a multi-dimensional driving feature vector.

[0197] The technical solution provided in this application effectively solves the problem of coordinating dynamic scene recognition and seat parameter optimization. Multi-dimensional data fusion processing enhances the representational capability of driving feature vectors, making driving scene classification more accurate. A dynamic Euclidean distance comparison mechanism enables automatic segmentation and tracking of driving scenes, adapting to sudden changes in road conditions.

[0198] S203. For each driving scenario's feature vector set in the set of driving feature vectors for multiple driving scenarios, determine the historical seat parameters for the same driving scenario in the historical vehicle information set based on the seat parameters corresponding to each driving feature vector in the set of feature vectors for each driving scenario.

[0199] It is understandable that multiple sets of driving feature vectors are obtained, each set corresponding to a driving scenario. Each set of driving feature vectors includes one or more driving feature vectors, and each driving feature vector corresponds to seat parameters.

[0200] For example, five sets of driving data were collected, resulting in three feature vector sets: A1 = {a1, b1}, A2 = {b2, b3}, and A3 = {b4, b5}. This indicates that there are three driving scenarios, with A1 representing one driving scenario, A2 representing another, and A3 representing yet another.

[0201] For each of the three driving scenarios, the seat parameters corresponding to each driving scenario need to be determined by the seat parameters corresponding to the driving feature vectors in each driving scenario.

[0202] Taking driving scenario A1 as an example, driving feature vector a1 corresponds to seat parameters, and driving feature vector b1 corresponds to seat parameters. For example, the duration of the same seat parameters in driving feature vector a1 and driving feature vector b1 can be compared. The seat parameter with the longest duration among the same seat parameters is marked as the first seat parameter. After comparing the duration of each same seat parameter, the first seat parameter among each same seat parameter can be obtained. Therefore, the seat parameters of driving scenario A1 are the first seat parameters among each same seat parameter.

[0203] In an optional embodiment, the seat parameters obtained by this method for each driving scenario can be stored in a cloud database along with the driving feature vector set and the corresponding seat parameters for each driving scenario. The feature vector set and the corresponding seat parameters for a driving scenario can form a vehicle information set, which is then stored in the cloud database to form a historical vehicle information set.

[0204] Compared to the traditional fixed parameter adjustment mode, this solution introduces a continuous duration verification mechanism to ensure the practicality and stability of the recommended parameters.

[0205] It is understood that the solution provided in this application, by performing feature processing on multiple driving data and classifying them into sets of driving feature vectors for multiple driving scenarios, helps to establish richer scenario models, thereby improving the accuracy of seat parameter matching. Compared with the shortcomings of traditional methods that rely on manually defined scenarios or limited data, this application realizes automatic clustering and updating of driving scenarios, which allows the driver's seat adjustment method to no longer be limited to a single driving scenario, but to adapt to different driving scenarios, thus improving the generalization ability of the driver's seat adjustment method.

[0206] The following examples illustrate possible implementation schemes of the lane-changing control method described in one or more of the above embodiments.

[0207] With the rapid development of technology and the continuous improvement of people's living standards, automobiles have become an indispensable means of transportation, and comfort has become an important factor for people when choosing a car. In particular, for drivers, the comfort of the seat directly affects their physical condition, thereby indirectly affecting driving safety. With the rapid development of automotive intelligence and autonomous driving technologies, the complexity and dynamism of driving scenarios have significantly increased. Traditional fixed or manually adjustable seat systems can no longer meet drivers' needs for comfort, safety, and adaptability. Therefore, adaptive adjustment of car seats has become increasingly important.

[0208] Current mainstream seat control system adjustment technologies mainly rely on preset parameters or simple environmental perception. However, the lack of in-depth analysis of dynamic scenarios and driver states in these technologies leads to the following problems with the adjustment results: 1. Traditional methods fail to fully integrate driving data and driver physiological data, making it difficult to achieve real-time optimization of seat parameters in complex road conditions or sudden driving behaviors; 2. Most related technologies rely on a single sensor, such as pressure distribution, without considering external factors such as ambient light and road conditions, affecting the accuracy of judgment; 3. Traditional seat adjustment parameters are mostly based on fixed thresholds or user preferences, lacking continuous learning of dynamic scenarios and unable to update parameters through big data to adapt to personalized needs.

[0209] Based on this, in order to solve the above-mentioned technical problems, this embodiment proposes a seat control method and system that can adaptively adjust according to the scenario.

[0210] Example 1: The driver's seat adjustment method provided in this embodiment can be referred to... Figure 3 , Figure 3 A flowchart illustrating a method for adjusting a driver's seat provided in this application embodiment. Figure 3 ,like Figure 4 As shown, the solution in this embodiment is as follows: S301. Continuously acquire actual driving data of the vehicle at a fixed frequency, and calculate the driving feature vector based on the actual driving data.

[0211] S302. Mark the first acquired driving feature vector as the reference vector, and mark the subsequently acquired driving feature vectors as comparison vectors.

[0212] S303. Calculate the Euclidean distance between the comparison vector and the reference vector in sequence, and compare it with the second set distance threshold: if it does not exceed the second set distance threshold, the comparison vector and the reference vector constitute a feature vector set, and the feature vector set corresponds one-to-one with the driving state of the car (i.e., the driving scenario); if it exceeds the second set distance threshold, the reference vector is updated to the current comparison vector.

[0213] S304. Repeatedly calculate the Euclidean distance and compare it with the second set distance threshold, update the reference vector, and obtain multiple feature vector sets.

[0214] That is, the multiple feature vector sets are A1, A2, A3, ..., Aτ.

[0215] S305. Obtain driver data (i.e., driver status data) and seat parameters when the car is in motion. Calculate the driver index based on the driver data. Determine whether to change the seat parameters based on the driver index. Obtain the duration corresponding to different seat parameters under the same driving condition. Extract the first seat parameter under the same driving condition based on the duration.

[0216] S306. Based on the historical vehicle information set of each type of vehicle under the same driving condition, update the first seat parameters corresponding to the historical vehicle information set under the same driving condition; when the car is driving, match the updated optimal seat parameters based on the driving feature vector under the current driving condition, and adaptively adjust the seat according to the first seat parameters.

[0217] The following is combined with Figure 4 This embodiment will be described in detail. Figure 4 The method for adjusting the driver's seat.

[0218] Figure 5 A flowchart illustrating a method for adjusting a driver's seat provided in this application embodiment. Figure 5 ,like Figure 5 As shown, the method for adjusting the driver's seat includes the following steps: S401. Obtain actual driving data at a fixed frequency.

[0219] In one example, actual driving data is collected in real time by onboard sensors, including dynamic parameters such as vehicle speed, steering angle, and road vibration. In another example, actual driving data includes data collected in real time by onboard sensors, such as driving duration, driving speed, ambient temperature, light intensity, whether turning, whether accelerating or decelerating, and road conditions. This embodiment will subsequently use actual driving data including data collected in real time by onboard sensors, such as driving duration, driving speed, ambient temperature, light intensity, whether turning, whether accelerating or decelerating, and road conditions, as an example for illustrative explanation.

[0220] S402, Calculate the driving feature vector.

[0221] Among them, the driving feature vector refers to a multi-dimensional data set after standardization of various driving data. Specifically, various driving data can be classified first, and then a normalization calculation method can be matched to combine the normalized numerical data and the classified data mapped to integers to form a comprehensive representation of the current driving status.

[0222] In the data preprocessing stage, the maximum and minimum values ​​of numerical data are normalized, and categorical data are converted into discrete integer values ​​and then normalized.

[0223] Most related technologies employ a single-type data processing approach. This solution creatively integrates numerical and categorical data to construct a multi-dimensional feature vector.

[0224] This embodiment further proposes a calculation process for the driving feature vector, including: the types of actual driving data include numerical data and categorical data; numerical data includes driving time, driving speed, ambient temperature, and light intensity; categorical data includes whether turning, whether accelerating or decelerating, and road conditions; the type of actual driving data is determined: if the actual driving data is numerical data, normalization is performed; if the actual driving data is categorical data, the categorical data is mapped to integers before normalization is performed; the normalized actual driving data constitutes the driving feature vector.

[0225] Numerical data refers to driving state parameters represented by continuous numerical values. Specifically, the maximum-minimum normalization method can be used to map the values ​​of different dimensions such as driving speed and ambient temperature to the [0, 1] interval, thereby eliminating the influence of dimensional differences on vector construction.

[0226] Categorical data refers to driving events or environmental conditions represented by discrete states. Specifically, integer encoding can be used to convert whether a turn is made into 0 or 1, and then normalization can be applied to bring it to the same order of magnitude as numerical data, avoiding interference from discrete values ​​on the vector space distribution. Normalization is the process of converting data from different ranges into a unified standard, which can be achieved through linear transformation methods, making different data types comparable in feature vectors.

[0227] Specifically, in the process of constructing driving feature vectors, numerical data such as driving time and speed are first normalized. For example, driving speed is divided by a preset maximum speed threshold to obtain dimensionless standardized values. For categorical data such as whether a turn is possible or whether the road conditions are slippery, they are first converted into binary integers, for example, mapping "yes" to 1 and "no" to 0. Then, normalization is performed to ensure that discrete and continuous values ​​are evenly distributed in the vector space. The resulting driving feature vector can simultaneously contain continuous driving parameters and discrete event states, forming a multi-dimensional state representation space, which solves the problem of incomplete feature information caused by data type separation in traditional methods.

[0228] Compared to related technologies, existing seat control methods do not differentiate between data types when processing driving data, directly splicing the raw data, leading to conflicts in magnitude and distribution between continuous parameters and discrete states. This embodiment, however, achieves effective fusion of numerical and categorical data through data type identification and difference processing. This allows the constructed feature vector to reflect both continuously changing parameters such as driving speed and ambient temperature, and accurately characterize discrete driving events such as acceleration, deceleration, and turning, thus improving the accuracy of driving state judgment.

[0229] Through the above technical solution, this embodiment solves the technical problem of incomplete feature construction caused by improper data type processing in traditional seat control. By using differentiated data preprocessing methods, it achieves the effective construction of multi-dimensional driving feature vectors. This solution retains the continuous variation characteristics of numerical data while accurately expressing the discrete state characteristics of categorical data, providing a reliable data foundation for subsequent accurate classification of driving states and matching of seat parameters, and avoiding misjudgment problems caused by missing feature vector information or distribution deviations.

[0230] S403. Determine whether the driving feature vector is being acquired for the first time. If yes, proceed to S404; otherwise, proceed to S406.

[0231] S404, marked as the reference vector.

[0232] The first acquired driving feature vector is marked as the reference vector.

[0233] The reference vector refers to the baseline feature vector acquired for the first time. Specifically, the driving feature vector calculated at the initial moment can be used as the baseline to provide a reference standard for subsequent state comparisons.

[0234] S405. Initialize the feature vector set.

[0235] It is understandable that, since this is the first driving data, there is only the first driving feature vector a1 at this time. At this time, a1 can temporarily become a feature vector set, that is, A1={a1}.

[0236] S406, marked as a comparison vector.

[0237] The subsequently obtained driving feature vectors are labeled as comparison vectors b1, b2, ..., bμ.

[0238] Where μ represents the number of comparison vectors.

[0239] S407. Calculate the Euclidean distance between the comparison vector and the reference vector.

[0240] The Euclidean distance between the comparison vector and the reference vector will be calculated sequentially.

[0241] Euclidean distance refers to a measure of the distance between vectors in a multidimensional space, used to quantify the degree of change in driving state.

[0242] S408. Determine whether the Euclidean distance exceeds the second set distance threshold. If yes, execute S411; otherwise, execute S409.

[0243] S409. Add to the current feature vector set.

[0244] The Euclidean distance between the comparison vector and the reference vector is calculated sequentially and compared with a second set distance threshold. If the distance does not exceed the second set distance threshold, the comparison vector and the reference vector constitute a feature vector set. A feature vector set corresponds to a driving state of the car. In other words, the feature vector set and the driving state of the car are in one-to-one correspondence.

[0245] Taking reference vector a1 and the first comparison vector b1 as an example, the Euclidean distance between b1 and a1 is calculated and compared with a second set distance threshold. If the Euclidean distance between b1 and a1 is less than the second set distance threshold, b1 and a1 are merged to obtain the first driving feature vector set A1, which can be represented by A1 = {a1, b1}, indicating that a1 and b1 belong to the same driving scenario. If the Euclidean distance between b1 and a1 is greater than or equal to the second set distance threshold, a1 is updated to b1. The a1 before the update forms a driving feature vector set A1, which can be represented by A1 = {a1}.

[0246] S410, Generate multiple feature vector sets.

[0247] S411, Update the reference vector.

[0248] Understandably, S407~S411 need to be considered together, that is: calculate the Euclidean distance between the comparison vector and the reference vector in sequence, and compare it with the second set distance threshold; if it does not exceed the second set distance threshold, the comparison vector and the reference vector constitute a feature vector set, and the feature vector set corresponds one-to-one with the same driving state of the car; if it exceeds the second set distance threshold, update the reference vector to the current comparison vector; by repeatedly calculating the Euclidean distance and comparing it with the second set distance threshold, and updating the reference vector, multiple driving feature vector sets A1, A2, A3...Aτ can be obtained.

[0249] Where τ represents the number of sets of driving feature vectors.

[0250] Each set of driving feature vectors corresponds to a set of first seat parameters.

[0251] S412, Obtain driver data and seat parameters.

[0252] It is understandable that the driver data and seat parameters here are the driver data and seat parameters under the current driving conditions.

[0253] S413, Calculate the driver index.

[0254] S414. Determine whether the first threshold is exceeded. If yes, execute S415; otherwise, execute S419.

[0255] Understandably, determining whether the first threshold has been exceeded involves judging whether the driver index (i.e., the target state index) exceeds the first threshold.

[0256] S415, Adjust seat parameters.

[0257] When the driver index exceeds the first threshold, it indicates that adjusting the seat parameters is permissible. It's understood that "adjusting seat parameters" here means the option to do so is available, but no actual adjustments have been made yet.

[0258] S416. Extract the first seat parameters corresponding to the feature vector set.

[0259] S417, Update the parameters of the first seat.

[0260] S418, Perform adaptive seat adjustment.

[0261] It is understandable that when seat parameters need to be adjusted, adaptive seat adjustment means adjusting according to the updated first seat parameters; when seat parameters do not need to be adjusted, adaptive seat adjustment means maintaining the current seat parameters.

[0262] S419, Keep current seat parameters.

[0263] The specific implementation methods of S412~S418 are described below.

[0264] The driver index is a comprehensive indicator reflecting the driver's physiological state. Specifically, it is calculated using a weighted summation method, taking into account factors such as steering wheel pressure, frequency of posture changes, and blinking frequency. This index is used to determine whether seat parameters need adjustment. The primary seat parameters are the effective adjustment parameters verified over a prolonged period. These parameters are typically those of the seat configuration used for the longest time under the same driving conditions, ensuring the stability of the adjustment strategy.

[0265] The feature vector construction module arranges and combines the processed data according to predetermined dimensions to form feature vectors representing the current driving state. The dynamic classification module continuously calculates the Euclidean distance between the new feature vector and the reference vector. When the Euclidean distance value is lower than a second preset distance threshold, the current feature set is expanded; when it exceeds the second preset distance threshold, a new feature set is established. The physiological monitoring module synchronously collects the driver's biometric data and calculates the driver index using preset weight coefficients. When the driver index exceeds a first threshold, seat parameter adjustments are triggered. The parameter optimization module records the usage time of seat parameters corresponding to different feature sets and selects the effective parameters with the longest usage time as the recommended configuration. The system obtains historical driving data through a cloud data platform and continuously updates the correspondence between feature sets and optimal parameters, forming an adaptive optimization closed loop.

[0266] Compared to related technologies, this solution overcomes the limitations of static feature classification. Traditional methods use fixed thresholds to classify driving states, which cannot adapt to continuously changing driving scenarios. By employing a dynamic Euclidean distance comparison mechanism, automatic clustering and updating of driving states are achieved. Compared to traditional fixed parameter adjustment modes, this solution introduces a continuous duration verification mechanism to ensure the practicality and stability of recommended parameters. Furthermore, cross-vehicle feature set optimization is achieved through cloud data sharing, forming a swarm intelligence optimization mechanism.

[0267] Through the above technical solutions, this embodiment effectively solves the coordination problem between dynamic scene recognition and seat parameter optimization. Multi-dimensional data fusion processing enhances the representational capability of feature vectors, making driving state classification more accurate. A dynamic Euclidean distance comparison mechanism enables automatic segmentation and tracking of driving scenes, adapting to sudden changes in road conditions. A dual verification mechanism based on driver index and duration ensures that seat adjustments respond to immediate physiological states while maintaining long-term usage preferences. Cloud-based data collaborative optimization enables the system to continuously learn, constantly improving the accuracy and personalization of parameter recommendations.

[0268] This embodiment further proposes a formula for calculating the Euclidean distance between the comparison vector and the reference vector. For example, taking the calculation of the Euclidean distance between the comparison vector b1 and the reference vector a1 as an example, the formula for calculating the Euclidean distance between the comparison vector b1 and the reference vector a1 can be referred to as the following formula (1): Formula (1); In formula (1), d(a1, b1) represents the Euclidean distance between comparison vector b1 and reference vector a1; n essentially represents the dimension of the driving feature vector, that is, the number of driving data in the driving feature vector. Then a1n means that the reference vector a1 has n dimensions, that is, n driving data; b1n means that the comparison vector b1 has n dimensions, that is, n driving data.

[0269] Here, Euclidean distance refers to the straight-line distance between two vectors in n-dimensional space, used to quantify the degree of difference between driving feature vectors. The vector dimension n refers to the number of categories of the normalized actual driving data, which is achieved by using the total number of items after normalizing numerical and categorical data to ensure that feature values ​​of different data types are comparable in the same dimensional space.

[0270] Specifically, when constructing driving feature vectors, numerical data is normalized, and categorical data, after being mapped to integers, is also normalized, forming vectors of uniform dimensions. When calculating the Euclidean distance between the comparison vector and the reference vector, the square root of the sum of the squared differences in each dimension yields the geometric distance (i.e., Euclidean distance) between the two vectors. If the Euclidean distance does not exceed a second preset distance threshold, it indicates that the current driving state is consistent with the state of the reference vector, and they are classified into the same feature vector set; if it exceeds the second preset distance threshold, it is determined to be a new state, and the reference vector is updated. Through continuous calculation and comparison, the feature vector sets of different driving states are accurately divided, avoiding classification errors caused by data dimension confusion or lack of normalization.

[0271] Compared to related technologies, traditional seat adjustment methods do not explicitly employ the Euclidean distance formula when classifying states, or they only use other metrics such as Manhattan distance and cosine similarity, failing to consider the geometric spatial relationships of multidimensional data after normalization. For example, some related technologies do not perform integer mapping and normalization on the classified data, leading to mismatched vector dimensions or inconsistent data distribution, affecting the reliability of distance calculation. This solution defines a mathematical formula for Euclidean distance and combines it with normalization processing to ensure that feature data from different sources are comparable within the same space, improving the accuracy of state classification.

[0272] Through the above technical solution, this embodiment solves the problem of inaccurate state judgment caused by incomplete construction of driving feature vectors, and achieves accurate classification of driving states. By using a unified Euclidean distance calculation rule, the differences between the multidimensional vectors after fusing numerical and categorical data are quantified, avoiding dimensional mismatch problems caused by mixed data types. The normalized feature vectors are compared in the same geometric space, enabling different driving states to be reliably distinguished, providing an accurate state classification basis for subsequent seat parameter matching.

[0273] This embodiment further proposes driver data during vehicle operation, including steering wheel grip pressure, frequency of posture changes, and blink frequency. Steering wheel grip pressure refers to the force exerted by the driver's hands on the steering wheel, which can be achieved by embedding a pressure sensor array under the steering wheel surface. The pressure signal is transmitted to the processing unit via an analog-to-digital converter. This parameter reflects the driver's muscle tension and operational load, capturing stress responses during emergency braking or sudden steering. Posture change frequency refers to the number of times the driver actively adjusts their body posture per unit time. This can be achieved by detecting center of gravity movement patterns using a seat pressure distribution sensor and counting the number of movements within a time window. This parameter characterizes the degree of matching between seat comfort and the driver's posture; the frequency increases significantly when seat support is insufficient. Blink frequency refers to the number of complete eyelid closures per minute. This can be achieved by using an infrared camera to capture eye images, identifying the opening and closing states using a convolutional neural network, and counting the frequency. This parameter quantifies the driver's visual fatigue level; the blink interval shows regular fluctuations when attention decreases.

[0274] Specifically, while the vehicle is in motion, the steering wheel pressure sensor continuously collects the three types of data mentioned above. It acquires pressure distribution data at a fixed sampling rate and extracts the average pressure as a dynamic indicator using a sliding window algorithm. The seat pressure sensor array monitors changes in contact surface pressure in real time, and records a valid seating posture adjustment when two consecutive center-of-gravity shifts exceed a threshold. The onboard vision system captures the driver's facial images at 30 frames per second, extracts blinking features using edge computing, and counts the number of times the eye-closing duration exceeds 100 milliseconds per minute. These three types of data are normalized and then input into a weighted calculation module to form a comprehensive driver index.

[0275] Compared to related technologies, traditional seat adjustment systems rely on only a single type of biometric data, such as judging fatigue solely through heart rate monitoring, and cannot distinguish between physical fatigue and mental stress. This solution, however, integrates biometric data from three dimensions: operational load, subjective comfort feedback, and visual fatigue. This allows it to identify physiological fatigue caused by prolonged driving, detect stress states resulting from sudden operational demands, and capture postural adjustment needs triggered by insufficient seat support, forming a multimodal biometric analysis system.

[0276] Through the above technical solution, this embodiment can identify changes in operational load during sudden driving scenarios based on hand pressure on the steering wheel, determine the matching degree of seat support and ergonomics by the frequency of posture changes, and accurately assess the degree of visual fatigue by combining blink frequency. Cross-validation of multi-dimensional biometrics effectively avoids misjudgments that may occur from a single data source, such as misjudging the high pressure during emergency braking as a state of fatigue. The resulting composite judgment mechanism can provide accurate physiological basis for adjusting seat parameters, improving body support stability during sudden operations while preventing fatigued driving.

[0277] This embodiment further proposes the calculation process of the driver index, which includes: weighting and summing the pressure of the hand gripping the steering wheel, the frequency of changing the sitting posture, and the frequency of blinking to obtain the driver index, with the weight coefficients being predefined; the specific calculation formula is: driver index = first preset coefficient × pressure of the hand gripping the steering wheel + second preset coefficient × frequency of changing the sitting posture + third preset coefficient × frequency of blinking, where the sum of the first preset coefficient, the second preset coefficient, and the third preset coefficient is 1.

[0278] Among these metrics, steering wheel grip pressure refers to the pressure applied by the driver to the steering wheel surface, which can be measured using a steering wheel grip force sensor to reflect the driver's level of tension or muscle fatigue. Posture change frequency refers to the number of times the driver adjusts their body posture per unit of time, which can be detected using a seat pressure distribution sensor or a vision sensor to characterize the degree of fatigue accumulation caused by localized pressure on the body. Blinking frequency refers to the number of times the driver closes their eyelids per minute, which can be statistically analyzed using an infrared camera combined with image recognition algorithms to assess visual fatigue or the degree of distraction. Weighting coefficients are pre-set proportional parameters based on the priority of each physiological characteristic's impact on driving performance; these can be values ​​set based on historical data analysis or expert experience to balance the contribution of different physiological indicators.

[0279] Specifically, the steering wheel pressure signal is collected in real time by sensors and normalized to a pressure value within the range of 0-1. The seat pressure distribution sensor detects the number of posture changes every 5 seconds and converts them into frequency values. The vision sensor counts the number of blinks per minute and calculates the frequency per minute. By linearly superimposing weighted coefficients of 0.4 for steering wheel pressure, 0.3 for posture change frequency, and 0.3 for blink frequency, a driver index in the 0-1 range is generated. When the index exceeds 0.6, it is determined to be a state of fatigue, triggering an adjustment strategy that adjusts the seat back angle to 110 degrees and increases lumbar support pressure by 20%. All index data are standardized before calculation to eliminate the influence of dimensional differences on the weighted results.

[0280] Compared to related technologies, traditional methods rely solely on single physiological indicators such as steering wheel pressure or blink frequency for threshold judgments, failing to consider the correlations and weighting differences between different physiological characteristics. For example, related technologies might determine fatigue simply by monitoring blink frequency exceeding 20 times per minute, but this doesn't eliminate false positives caused by strong light stimulation. This solution, through multi-dimensional data fusion and dynamic weight allocation, can effectively distinguish between increased steering wheel grip due to tension and true fatigue in high-speed cornering scenarios, avoiding mis-triggered adjustments based on a single indicator.

[0281] Through the above technical solution, this embodiment addresses the technical deficiency of traditional seat adjustment systems in accurately quantifying the driver's physiological state. By establishing an objective evaluation index related to the driver's true fatigue level through a weighted calculation model that integrates steering wheel pressure, posture changes, and blink frequency, reliable physiological data support is provided for the dynamic adjustment of seat parameters. In long-distance driving tests, this solution reduced the false trigger rate of seat support force adjustment by approximately 30%, and in emergency braking scenarios, it can trigger seat cushion tilt adjustment 50 milliseconds in advance to maintain driver posture stability.

[0282] This embodiment further proposes to set a composite parameter system including backrest angle, seat height, fore-aft distance, temperature and vibration frequency during seat control. By comparing the real-time calculated driver index with a preset threshold, the system determines whether to trigger seat parameter adjustment based on the comparison result.

[0283] The backrest angle refers to the angle between the seat back and the horizontal plane. This is measured and adjusted using an electric push-rod mechanism driven by an angle sensor to accommodate the driver's back support needs. The seat cushion height is the vertical distance between the bottom of the seat and the vehicle floor. This is dynamically adjusted using a pneumatic lifting device and a pressure sensor to maintain the driver's natural leg curvature. The fore-aft distance refers to the amount of forward and backward movement of the seat rails. This is controlled by a stepper motor driven by a rail displacement sensor to ensure a safe operating distance between the driver and the steering wheel. The temperature refers to the output temperature of the seat surface heating or ventilation device. This is achieved using a thermocouple sensor linked to a temperature control module to adjust the perceived comfort level. The vibration frequency refers to the operating frequency of the seat's built-in shock absorbers. This is achieved by collecting road vibration data using an accelerometer, and then adjusting the output frequency of the electromagnetic shock absorbers by a controller to achieve dynamic buffering compensation under bumpy road conditions. The driver index threshold is a pre-set physiological state threshold. This is determined by establishing a statistical model based on biometric data collected in a laboratory at different fatigue levels to distinguish between normal driving conditions and intervention-required conditions.

[0284] Specifically, during vehicle operation, the system continuously monitors biometric data such as steering wheel pressure, frequency of posture changes, and blinking frequency, generating a real-time driver index through weighted calculation. This index is then compared in real-time with preset thresholds: when the index exceeds the threshold, it indicates driver fatigue or discomfort, at which point the system automatically adjusts at least one seat parameter, such as increasing the backrest angle to provide lumbar support, raising the seat cushion height to improve visibility, or activating the seat heating function to relieve muscle tension; when the index does not exceed the threshold, the current parameter configuration is maintained to avoid ineffective adjustments. Vibration frequency parameters are dynamically matched based on real-time road condition data; for example, the system automatically increases the damping frequency when continuous bumpy road sections are detected, forming a coordinated adjustment mechanism with temperature parameters. The threshold comparison process employs a dual-state decision logic, initiating adjustments only when intervention is truly necessary, ensuring both responsiveness and reduced system energy consumption.

[0285] Compared to related technologies, traditional seat adjustment systems often employ fixed parameter combinations or single-dimensional adjustments. For example, they may adjust the seat position solely based on user presets or rely on ambient temperature to control the heating function, lacking a multi-parameter coordination mechanism. Related technologies often rely on subjective settings to assess the driver's physiological state, lacking quantitative indicators and dynamic threshold judgment mechanisms, leading to inaccurate adjustment timing. This solution utilizes a composite parameter system to cover both ergonomic support and environmental adaptation needs, combining precise comparisons of driver indices and dynamic thresholds to form a data-driven adjustment decision-making mechanism.

[0286] Through the above technical solution, this embodiment achieves precise matching between seat parameter adjustments and the driver's physiological state. When fatigue characteristics are detected, multi-dimensional adjustment actions are automatically triggered, effectively alleviating the driver's physical burden. The threshold comparison mechanism avoids the waste of resources caused by frequent minor adjustments, ensuring that adjustment actions are initiated only when necessary. Dynamic control of vibration frequency and temperature parameters can proactively adapt to complex road condition changes, improving overall comfort and safety during driving.

[0287] This embodiment further proposes a process for extracting the first seat parameter under the same driving state, including: comparing the duration with a set duration threshold; if the duration exceeds the set duration threshold, the seat parameter is determined to be a valid parameter; if the duration does not exceed the duration threshold, the seat parameter is determined to be an invalid parameter; when there are multiple valid parameters among the same seat parameters, the valid parameter with the longest duration can be marked as the first seat parameter under the same driving state.

[0288] The duration refers to the continuous time during which seat parameters remain unchanged under a specific driving condition. This can be achieved using a timer or timestamp difference calculation to verify the stability of parameter adjustments. The duration threshold is a pre-set time threshold used to distinguish between valid and invalid parameters. This can be a fixed value or a dynamically adjusted variable based on the driving scenario, such as a 10-minute threshold or an adaptive value based on road complexity, used to filter short-term interference factors. Valid parameters refer to seat parameter configurations verified by the duration threshold. These can be obtained through a duration statistics module to build a stable and reliable parameter database. The first seat parameter refers to the seat parameter marked as the optimal configuration under the same driving condition. This can be determined using a priority ranking algorithm, such as using duration as a weighting indicator, and serves as the benchmark parameter for adaptive seat adjustment.

[0289] Specifically, the validity assessment of seat parameter adjustments is achieved through a time-based verification mechanism. When the same driving state is detected, the system records the duration of the current seat parameter and compares it with a preset threshold. If the duration exceeds the threshold, the parameter is determined to be a valid configuration under the driver's adaptation state; if it does not exceed the threshold, it is considered invalid data generated by temporary adjustments. Among multiple valid parameters, the system selects the parameter with the longest duration as the baseline configuration and associates it with the feature vector set of the corresponding driving state. For example, in a smooth highway driving scenario, if the seat back angle is maintained at 30 degrees for more than 15 minutes, this angle is marked as the first seat parameter in that scenario.

[0290] Compared to related technologies, traditional seat parameter validity assessments rely solely on a single point in time or simple threshold triggers, lacking continuous verification of parameter stability. This solution introduces a filtering mechanism combining duration and dynamic thresholds to effectively distinguish between temporary adjustments and stable states, preventing erroneous parameter updates due to brief operations. For example, related technologies may incorrectly record parameters due to temporary changes in driver posture, while this solution eliminates such interference through a duration-based filtering mechanism.

[0291] Through the above technical solution, this embodiment solves the problem of inaccurate judgment of seat parameter validity in dynamic driving scenarios. The duration threshold verification mechanism effectively filters invalid parameters generated by sudden driving behaviors or environmental changes, ensuring the reliability of parameter extraction. Simultaneously, using the longest-lasting valid parameter as the benchmark configuration improves the matching degree between seat adjustment and the driver's actual needs, avoiding interference from invalid data to the adaptive system.

[0292] This embodiment further proposes to obtain historical driving data containing historical driving feature vectors of different vehicles and corresponding seat parameters from a cloud database to form a feature vector set; when a new seat parameter in the feature vector set has a duration that exceeds the duration of the existing first seat parameter, the new parameter is marked as the first seat parameter.

[0293] Historical driving data refers to a dataset of driving feature vectors and seat parameters recorded by multiple vehicles under different driving conditions, stored in the cloud. This can be implemented using a distributed database, expanding feature coverage through multi-source data aggregation. The duration comparison mechanism verifies validity based on the cumulative time parameters have been used continuously under the same driving conditions. This can be implemented using a timestamp accumulation algorithm, with duration threshold filtering to ensure parameter stability. The feature vector set refers to the mapping relationship between driving features and seat parameters, categorized and stored according to driving conditions. This can be implemented using clustering analysis algorithms, with dynamic updates to the set enhancing the diversity of the parameter database.

[0294] Specifically, the cloud database periodically synchronizes driving feature vectors and corresponding seat parameters uploaded by each vehicle. After data cleaning to remove outliers, feature vectors of the same type of vehicle under the same driving state are grouped into the same set. When a new seat parameter appears in a feature vector set, the system records the continuous usage time of that parameter in that state and compares it with the duration of the existing best parameter in the set. If the cumulative usage time of the new parameter exceeds that of the original best parameter, a parameter replacement rule is triggered, and the new parameter is adopted as the recommended value for that feature vector set. For example, during highway cruising, if a new type of seat tilt angle parameter is used by multiple vehicles for more than 300 minutes, it is automatically upgraded to the first seat parameter.

[0295] Compared to related technologies, traditional seat parameter optimization relies on local data from a single vehicle, failing to overcome the limitations of individual usage habits. This solution, however, integrates multi-vehicle data in the cloud, enabling the feature vector set to cover more driving scenarios and user groups. For example, it simultaneously collects bumpy road condition data from both SUVs and sedans, forming a more comprehensive parameter recommendation library. While related technologies use fixed thresholds to determine parameter validity, this solution employs a continuous, dynamic competition mechanism, allowing the parameter library to autonomously eliminate outdated configurations and absorb the optimal solution from the group.

[0296] Through the above technical solution, this embodiment achieves dynamic incremental optimization of the seat parameter library, solving the problem of insufficient personalized adaptation caused by data isolation in traditional systems. By fusing multi-source data in the cloud, parameter recommendations possess cross-vehicle and cross-user generalization capabilities; for example, automatically calling safe seating posture parameters verified by multiple vehicles in sharp turning scenarios. A continuous competition mechanism ensures that parameter updates retain historically valid configurations while also promptly integrating new solutions verified in group driving behavior, such as gradually optimizing seat temperature adjustment strategies in long-distance driving scenarios.

[0297] This embodiment further proposes a process for matching seat parameters based on a set of feature vectors, including: acquiring driving feature vectors of the vehicle at a fixed frequency; calculating the average Euclidean distance between the driving feature vectors and all feature vector sets; if the average Euclidean distance is less than a preset distance threshold, then marking the feature vector set as a matching set; acquiring the first seat parameters corresponding to all matching sets, and setting the first seat parameter with the longest duration as the seat parameter in this driving state.

[0298] The driving feature vector is a vector formed by normalizing numerical and categorical data. Specifically, numerical data such as driving speed and ambient temperature are linearly transformed to the 0-1 range, and categorical data such as whether a turn occurs are mapped to integers and then normalized. This feature vector comprehensively represents the multi-dimensional characteristics of the driving state. The average Euclidean distance is the average distance between the current driving feature vector and all vectors in a given feature vector set. This can be calculated using the root mean square difference of the squared differences of vector dimensions, used to quantify the overall similarity between the current state and the historical set. The longest-lasting first seat parameter refers to the parameter configuration that has been verified as effective and used for the longest time in the same driving state. This can be achieved by statistically analyzing the cumulative effective time of seat parameters in this state, prioritizing reliable parameters that have been validated over a long period.

[0299] Specifically, during vehicle operation, a current driving feature vector is collected at specific time intervals (e.g., once per second). This vector contains normalized data such as speed and steering status. This vector is then compared with multiple feature vector sets stored in the cloud: for each set, the mean Euclidean distance between the current vector and all historical vectors within that set is calculated. When this mean is less than a preset threshold (e.g., 0.3), it indicates that the current driving state is highly similar to the typical state recorded in that set, and the set is marked as a matching set. Among multiple matching sets, the optimal seat parameters corresponding to each set are extracted, and the parameter with the longest cumulative usage time is ultimately selected as the adjustment basis. For example, when the vehicle enters a series of curves, if the average distance value of the curve scene set matched by the system is 0.25, and the backrest angle parameters corresponding to this set include a 45-degree configuration (cumulative usage of 3 hours) and a 40-degree configuration (cumulative usage of 5 hours), then the 40-degree parameter will be prioritized for automatic adjustment.

[0300] Compared to related technologies, traditional methods adjust parameters by simply comparing a single feature vector with preset parameters, failing to consider the overall distribution characteristics of historical data. This scheme, however, effectively avoids misjudgments caused by fluctuations in individual data by calculating the overall similarity between the current state and the feature set. Furthermore, the parameter optimization mechanism based on duration overcomes the limitations of traditional methods that rely solely on instantaneous data or fixed rules, enhancing the stability and adaptability of parameter configuration.

[0301] Through the above technical solution, this embodiment can effectively solve the problem of mismatched seat parameters in dynamic scenarios, and achieve accurate mapping between driving status and seat configuration. Specifically, it improves the accuracy of complex road condition recognition by about 30% through dynamic matching of multi-dimensional feature vectors; improves the reliability of adjustment decisions by about 25% based on the parameter optimization mechanism based on continuous duration; and shortens the seat parameter adaptation response time to less than 0.5 seconds through real-time comparison of cloud feature sets.

[0302] Example 2: The driver's seat adjustment device provided in this embodiment can be referred to as follows. Figure 6 , Figure 6 This is a schematic diagram of the structure of a driver's seat adjustment device provided in an embodiment of this application, as shown below. Figure 6 As shown, the solution in this embodiment is as follows: The driver's seat adjustment device 50 includes: Driving feature extraction module 501: used to continuously acquire actual driving data of the vehicle at a fixed frequency, and calculate driving feature vector based on the actual driving data.

[0303] Feature analysis module 502: It is used to mark the first acquired driving feature vector as the reference vector and the subsequently acquired driving feature vector as the comparison vector; it calculates the Euclidean distance between the comparison vector and the reference vector in turn and compares it with a preset distance threshold: if the distance threshold is not exceeded, the comparison vector and the reference vector constitute a feature vector set A1, and the feature vector set corresponds one-to-one with the same driving state of the car; if the distance threshold is exceeded, the reference vector is updated to the current comparison vector; it repeatedly calculates the Euclidean distance and compares it with the distance threshold, updates the driving feature reference vector, and obtains the feature vector sets A2, A3, ..., Aτ.

[0304] Seat parameter extraction module 503: acquires driver data (i.e., driver status data) and seat parameters when the car is in motion, calculates driver index based on driver data, determines whether to change seat parameters based on driver index, and acquires the duration corresponding to different seat parameters under the same driving state, and extracts the first seat parameter under the same driving state based on the duration.

[0305] Seat parameter optimization module 504: Based on the historical vehicle information set of various types of vehicles under the same driving conditions, it updates the first seat parameters corresponding to the historical vehicle information set under the same driving conditions; when the car is driving, it matches the updated optimal seat parameters based on the driving feature vector of the current driving conditions, and adaptively adjusts the seat according to the first seat parameters.

[0306] This embodiment has the following beneficial effects: 1. By integrating multi-dimensional driving environment data with driver physiological characteristics, the system achieves refined perception of complex driving scenarios, effectively enhancing its adaptability to dynamic road conditions. 2. By using a continuous duration analysis and big data learning mechanism, the system automatically filters and updates the optimal adjustment parameters to ensure that the seat configuration continues to evolve with driving scenarios and driver needs; 3. Through multi-dimensional biometric collaborative monitoring, it accurately captures changes in the driver's physiological state and triggers seat parameter adjustments in advance to alleviate fatigue and improve comfort; 4. By adopting a data cleaning and dynamic matching mechanism, noise interference and abnormal operation are effectively filtered out, ensuring the reliability of seat adjustment decisions and the consistency of scenario adaptation.

[0307] It should be noted that although the steps of the method in this application are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps; or steps from different embodiments may be combined into a new technical solution.

[0308] Based on the same inventive concept as the foregoing embodiments, this application provides a driver's seat adjustment system, which can be referred to... Figure 7 .

[0309] Figure 7 This is a schematic diagram of the structure of a driver's seat adjustment system provided in an embodiment of this application, as shown below. ​ As shown, there is an acquisition module 601, a determination module 602, and an adjustment module 603.

[0310] The acquisition module 601 is configured to acquire the driver's state data and the current seat parameters of the driver's seat in the current driving scenario; the state data is used to characterize the driver's action response in the current driving scenario.

[0311] The determination module 602 is configured to determine a target state index based on the driver's state data; the target state index is used to characterize the driver's degree of adaptation to the current seat parameters.

[0312] The adjustment module 603 is configured to, when the target state index exceeds a first threshold, obtain historical seat parameters under the same driving scenario from the historical vehicle information set; and adjust the driver's seat to suit the driver based on the comparison result of the historical seat parameters and the current seat parameters.

[0313] In an optional embodiment, the adjustment module 603 is further configured to: obtain a first duration corresponding to each type of parameter in the historical seat parameters; obtain a second duration corresponding to each type of parameter in the current seat parameters; for each first same type parameter in the historical seat parameters and the current seat parameters: take the parameter with the longest duration among each first same type parameter as the target parameter to be adjusted; and adjust the driver's seat according to at least one target parameter to be adjusted.

[0314] In an optional embodiment, the adjustment module 603 is further configured to: acquire a set of historical vehicle information for various types of vehicles under the same driving scenario; select a target information set from the set of historical vehicle information for various types of vehicles based on the current driving scenario; the target information set is the set of historical vehicle information that best matches the current driving scenario; and adjust the driver's seat based on the comparison result between the seat parameters in the target information set and the current seat parameters.

[0315] In an optional embodiment, the adjustment module 603 is further configured to: determine the driving feature vector in the current driving state based on the driving data of the vehicle in the current driving scenario; for each type of vehicle in the historical vehicle information set, determine the average distance between the driving feature vector in the current driving state and multiple driving feature vectors in the type vehicle feature vector set in the historical vehicle information set of the type vehicle; determine the type vehicle feature vector set whose average distance is less than a first set distance threshold as the matching feature vector set; and determine the historical vehicle information set corresponding to the matching feature vector set as the target information set.

[0316] In an optional embodiment, the adjustment module 603 is further configured to: for each second identical type parameter in the target information set and the current seat parameters, take the type parameter with the longest duration in each second identical type parameter as the target parameter to be adjusted; and adjust the driver's seat according to at least one target parameter to be adjusted.

[0317] In an optional embodiment, the adjustment module 603 is further configured to: obtain the duration of each parameter in the current seat parameters; determine the parameters whose duration exceeds a set duration as valid parameters; assign a set duration to each parameter; and adjust the driver's seat based on the comparison result between the historical seat parameters and the valid parameters in the current seat parameters.

[0318] In an optional implementation, the determining module 602 is further configured to: perform feature processing on each driving data in the multiple driving data of the vehicle to obtain multiple driving feature vectors; classify the multiple driving feature vectors to obtain a set of driving feature vectors for multiple driving scenarios; and for each driving scenario's driving feature vector set in the set of driving feature vectors for multiple driving scenarios, determine the historical seat parameters in the historical vehicle information set for the same driving scenario based on the seat parameters corresponding to each driving feature vector in the set of driving feature vectors for each driving scenario.

[0319] In an optional implementation, the determining module 602 is further configured to: determine a target state index based on the driver's state data, including: the pressure of the driver's hand on the steering wheel, the frequency of the driver's posture changes, and the frequency of the driver's blinking; determine a first weight of the pressure of the driver's hand on the steering wheel, a second weight of the frequency of the driver's posture changes, and a third weight of the frequency of the driver's blinking; determine the target state index based on the pressure of the driver's hand on the steering wheel, the frequency of the driver's posture changes, and the frequency of the driver's blinking, as well as the first weight, the second weight, and the third weight; wherein the second weight is greater than the third weight, and the third weight is greater than the first weight.

[0320] The descriptions of the above device embodiments are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0321] It should be noted that the module division in the embodiments of this application is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, exist as separate physical units, or have two or more units integrated into one unit. The integrated units can be implemented in hardware, as software functional units, or a combination of software and hardware.

[0322] It should be noted that, in the embodiments of this application, if the above-described methods are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause the vehicle equipment to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0323] This application provides a vehicle device. ​ This is a schematic diagram of the structure of a vehicle device provided in an embodiment of this application, such as... ​As shown, the vehicle device 70 includes a memory 701 and a processor 702. The memory 701 stores a computer program that can run on the processor 702. When the processor 702 executes the program, it implements the steps in the method provided in the above embodiments.

[0324] It should be noted that the memory 701 is configured to store instructions and applications executable by the processor 1102, and can also cache data to be processed or already processed (e.g., image data, audio data, voice communication data and video communication data) in the processor 702 and various modules in the vehicle equipment 70, which can be implemented by flash memory or random access memory (RAM).

[0325] This application also provides a computer-readable storage medium for storing computer programs.

[0326] Optionally, the computer-readable storage medium can be applied to the vehicle equipment in the embodiments of this application, and the computer program causes the processor or vehicle equipment to perform the various methods of the embodiments of this application, which will not be described in detail here for the sake of brevity.

[0327] This application also provides a computer program product, including computer program instructions.

[0328] Optionally, the computer program product can be applied to the vehicle equipment in the embodiments of this application, and the computer program instructions cause the processor or vehicle equipment to execute the various methods of the embodiments of this application. For the sake of brevity, they will not be described in detail here.

[0329] This application also provides a computer program.

[0330] Optionally, the computer program can be applied to the vehicle equipment in the embodiments of this application. When the computer program runs on the processor or vehicle equipment, it causes the processor or vehicle equipment to execute the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0331] It should be noted that the descriptions of the vehicle equipment, storage medium, computer program product, and computer program embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the vehicle equipment, storage medium, computer program product, and computer program embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0332] It should be understood that the phrases "one embodiment," "an embodiment," or "some embodiments" mentioned throughout the specification mean that a specific feature, structure, or characteristic related to an embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment," "in one embodiment," or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The descriptions of the various embodiments above tend to emphasize the differences between the various embodiments; their similarities or commonalities can be referred to mutually, and for the sake of brevity, they will not be repeated here.

[0333] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three kinds of relationships. For example, object A and / or object B can represent three situations: object A exists alone, object A and object B exist simultaneously, and object B exists alone.

[0334] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0335] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple modules or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or modules can be electrical, mechanical, or other forms.

[0336] The modules described above as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules. They may be located in one place or distributed across multiple network units. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0337] In addition, each functional module in the various embodiments of this application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the integrated modules can be implemented in hardware or in the form of hardware plus software functional units.

[0338] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0339] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause the vehicle equipment to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0340] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0341] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0342] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0343] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for adjusting a driver's seat, characterized in that, Includes the following steps: Acquire the driver's state data and the current seat parameters of the driver's seat in the current driving scenario; the state data is used to characterize the driver's action response in the current driving scenario; Determine the target state index based on the driver's state data; The target state index is used to characterize the driver's degree of adaptation to the current seat parameters; If the target state index exceeds the first threshold, obtain historical seat parameters for the same driving scenario from the historical vehicle information set; Based on the comparison between the historical seat parameters and the current seat parameters, the driver's seat is adjusted to suit the driver.

2. The method according to claim 1, characterized in that, Adjusting the driver's seat based on a comparison of the historical seat parameters and the current seat parameters includes: Obtain the first duration corresponding to each type of parameter in the historical seat parameters; Obtain the second duration corresponding to each type of parameter in the current seat parameters; For each first identical type parameter in the historical seat parameters and the current seat parameters: the parameter with the longest duration in each first identical type parameter is taken as the target parameter to be adjusted; The driver's seat is adjusted according to at least one target parameter to be adjusted.

3. The method according to claim 1, characterized in that, The method involves obtaining historical seat parameters under the same driving scenario from the historical vehicle information set. Based on the comparison between the historical seat parameters and the current seat parameters, the driver's seat is adjusted, including: Obtain a collection of historical vehicle information for various types of vehicles under the same driving scenario; Based on the current driving scenario, a target information set is selected from the historical vehicle information sets of various types of vehicles; the target information set is the historical vehicle information set that best matches the current driving scenario. The driver's seat is adjusted based on the comparison between the seat parameters in the target information set and the current seat parameters.

4. The method according to claim 3, characterized in that, The step of selecting a target information set based on the historical vehicle information set of various types of vehicles according to the current driving scenario includes: Determine the driving feature vector of the current driving state based on the vehicle's driving data in the current driving scenario; For each type of vehicle in the historical vehicle information set, determine the average distance between the driving feature vector in the current driving state and multiple driving feature vectors in the type vehicle feature vector set in the historical vehicle information set of the type vehicle. The set of vehicle feature vectors corresponding to types with an average distance less than a first set distance threshold is determined as the matching feature vector set; The set of historical vehicle information corresponding to the set of matching feature vectors is determined as the target information set.

5. The method according to claim 3, characterized in that, Adjusting the driver's seat based on the comparison result between the seat parameters in the target information set and the current seat parameters includes: For each second identical type parameter in the target information set and the current seat parameters, the type parameter with the longest duration in each second identical type parameter is taken as the target parameter to be adjusted. The driver's seat is adjusted according to at least one target parameter to be adjusted.

6. The method according to any one of claims 1 to 5, characterized in that, Before adjusting the driver's seat based on the comparison result of the historical seat parameters and the current seat parameters, the method further includes: For each parameter in the current seat parameters, obtain the duration of each parameter; Parameters whose duration exceeds a set duration are determined to be valid parameters; each parameter corresponds to a set duration. The driver's seat is adjusted based on a comparison of the valid parameters in the historical seat parameters and the current seat parameters.

7. The method according to any one of claims 1 to 5, characterized in that, The method further includes: Each driving data point in multiple vehicle driving data sets is characterized separately to obtain multiple driving feature vectors; The driving feature vectors are classified multiple times to obtain a set of driving feature vectors for multiple driving scenarios; For each driving feature vector set in a set of driving feature vectors for multiple driving scenarios, the historical seat parameters for the same driving scenario in the historical vehicle information set are determined based on the seat parameters corresponding to each driving feature vector in the set of driving feature vectors for each driving scenario.

8. The method according to any one of claims 1 to 5, characterized in that, When the driver's state data includes: the pressure of the driver's hands on the steering wheel, the frequency of the driver's posture changes, and the frequency of the driver's blinking, determining the target state index based on the driver's state data includes: The first weight of the pressure of the driver's hand on the steering wheel, the second weight of the frequency of the driver's posture changes, and the third weight of the frequency of the driver's blinking are determined. The target state index is determined based on the pressure of the driver's hands on the steering wheel, the frequency of the driver's posture changes, and the frequency of the driver's blinking, as well as the first weight, the second weight, and the third weight. Wherein, the second weight is greater than the third weight, and the third weight is greater than the first weight.

9. A driver's seat adjustment system, characterized in that, It includes: an acquisition module, a determination module, and an adjustment module; The acquisition module is configured to acquire the driver's status data and the current seat parameters of the driver's seat in the current driving scenario; The state data is used to characterize the driver's action response in the current driving scenario; The determining module is configured to determine a target state index based on the driver's state data. The target state index is used to characterize the driver's degree of adaptation to the current seat parameters; The adjustment module is configured to obtain historical seat parameters under the same driving scenario from the historical vehicle information set when the target state index exceeds the first threshold. Based on the comparison between the historical seat parameters and the current seat parameters, the driver's seat is adjusted to suit the driver.

10. A vehicle device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 8.

Images