Self-adaptive adjusting method and self-adaptive adjusting device of intelligent seat and intelligent seat

Abstract

The invention discloses a self-adaptive adjustment method and device of an intelligent seat and the intelligent seat, and belongs to the technical field of seat adjustment. The adaptive adjustment method comprises the following steps: acquiring multi-modal sensing data acquired by a sensor assembly, and performing fusion processing on the multi-modal sensing data to obtain a feature vector; determining a correction control parameter corresponding to a proportional-integral-differential control algorithm based on the feature vector; based on the corrected control parameters, a proportional-integral-differential control algorithm is adopted to determine a target adjusting quantity; adjusting the executing mechanism according to the target adjusting quantity; and determining a hierarchical adjustment strategy according to the feature vector, and performing hierarchical adjustment on the execution mechanism. Thus, dynamic self-adaptive adjustment of the intelligent seat is achieved, the intelligent seat can adapt to different users, adjustment response differences of the intelligent seat for different users are considered, the problems of overshoot or adjustment delay and the like are avoided, and the intelligent seat can be accurately adjusted for different users and in different states of the users.

Description

Technical Field

[0001] This application relates to the field of seat adjustment technology, and in particular to an adaptive adjustment method, adaptive adjustment device, smart seat, and computer-readable storage medium for a smart seat. Background Technology

[0002] Currently, most people spend long hours in the office and maintain a seated posture for extended periods. However, chairs lack dynamic adjustment capabilities and are difficult to adapt to different users. This can lead to problems such as excessive local pressure (e.g., in the buttocks and lower back) and uneven stress on the spine after prolonged sitting. Long-term use can cause chronic diseases such as lumbar muscle strain and lumbar disc herniation. Summary of the Invention

[0003] This application provides an adaptive adjustment method, adaptive adjustment device, smart seat, and computer-readable storage medium for a smart seat, in order to solve at least one of the aforementioned technical problems.

[0004] The adaptive adjustment method for a smart seat according to embodiments of this application, wherein the smart seat includes a sensor assembly and an actuator, the adaptive adjustment method comprising: The multimodal sensing data collected by the sensor components is acquired, and the multimodal sensing data is fused to obtain a feature vector. Based on the aforementioned feature vectors, determine the corrected control parameters corresponding to the proportional-integral-derivative control algorithm; The target adjustment amount is determined based on the modified control parameters using the proportional-integral-derivative control algorithm. Adjust the actuator according to the target adjustment amount; A hierarchical adjustment strategy is determined based on the feature vector, and the actuator is adjusted hierarchically.

[0005] In some embodiments, the multimodal sensing data includes any one or more of pressure sensing data, inertial sensing data, and millimeter-wave radar data; and / or The feature vector includes any one or more of the following: body shape vector, posture vector, and fatigue index.

[0006] In some implementations, the feature vector includes a body shape vector and a fatigue index, and the step of determining the corrected control parameters corresponding to the proportional-integral-derivative control algorithm based on the feature vector includes: The initial control parameters corresponding to the proportional-integral-derivative control algorithm are determined by querying the pre-stored body shape-parameter mapping table based on the body shape vector. The initial control parameters are fine-tuned based on the body shape vector and the fatigue index to obtain the corrected control parameters.

[0007] In some embodiments, the step of fine-tuning the initial control parameters based on the body shape vector and the fatigue index to obtain corrected control parameters includes: Based on the body shape vector and the fatigue index, determine the error vector between the target adjustment amount and the actual adjustment amount of the actuator; The initial control parameters are fine-tuned based on the error vector using a recursive least squares algorithm at preset intervals to obtain the corrected control parameters.

[0008] In some embodiments, the feature vector includes a fatigue index, and the step of determining a graded adjustment strategy based on the feature vector and performing graded adjustment on the actuator includes: If the fatigue index is determined to be less than a first threshold, the actuator is adjusted to the first level. If the fatigue index is determined to be greater than the first threshold and less than the second threshold, the actuator is subjected to secondary adjustment. If the fatigue index is determined to be greater than the second threshold and less than the third threshold, the actuator is adjusted in three levels. If the fatigue index is determined to be greater than the third threshold, the actuator is adjusted to level four.

[0009] In some embodiments, the first-level adjustment of the actuator includes: Control the actuator to maintain its current state; and / or The secondary adjustment of the actuator includes: Control a portion of the adjustment module in the actuator to make adjustments, and continue for a first duration; and / or The three-level adjustment of the actuator includes: Control a portion of the adjustment module in the actuator to make adjustments, and continue for a second duration; and / or The four-level adjustment of the actuator includes: The actuator controls all adjustment modules to adjust, and this continues for a third duration; Wherein, the first duration is less than the second duration, and the second duration is less than the third duration.

[0010] In some embodiments, the sensor assembly includes a pressure sensor, an inertial measurement unit, and a millimeter-wave radar. After determining the hierarchical adjustment strategy based on the feature vector and performing hierarchical adjustment on the actuator, the adaptive adjustment method further includes: The pressure of the smart seat is monitored in real time using the pressure sensor. If it is determined that the pressure is less than the pressure threshold and this condition persists for a fourth time, the actuator is controlled to reset, the inertial measurement unit and the millimeter-wave radar are controlled to shut down, and the pressure sensor is controlled to remain operational.

[0011] The adaptive adjustment device for a smart seat according to an embodiment of this application, wherein the smart seat includes a sensor assembly and an actuator, the adaptive adjustment device comprising: The data processing module is used to acquire multimodal sensing data collected by the sensor components and to perform fusion processing on the multimodal sensing data to obtain feature vectors. The first determining module is used to determine the corrected control parameters corresponding to the proportional-integral-derivative control algorithm based on the feature vector; The second determining module is used to determine the target adjustment amount based on the corrected control parameters using the proportional-integral-derivative control algorithm. The first adjustment module is used to adjust the actuator according to the target adjustment amount; The second adjustment module is used to determine a graded adjustment strategy based on the feature vector and to perform graded adjustment on the actuator.

[0012] The smart seat according to the embodiments of this application includes one or more processors and a memory, the memory storing a computer program, which, when executed by the processor, implements the adaptive adjustment method described in any of the above embodiments.

[0013] The computer-readable storage medium of the present application embodiment stores a computer program that, when executed by a processor, implements the adaptive adjustment method of any of the above embodiments.

[0014] The adaptive adjustment method, adaptive adjustment device, smart seat, and computer-readable storage medium of this application's embodiments determine feature vectors based on multimodal data, determine corrected control parameters corresponding to the proportional-integral-derivative (PID) control algorithm based on the feature vectors, and then determine the target adjustment amount using the PID control algorithm based on the corrected control parameters to adjust the actuator. Furthermore, a graded adjustment strategy is determined based on the feature vectors to perform graded adjustment of the actuator. In this way, dynamic adaptive adjustment of the smart seat is achieved, adapting to different users. It also considers the differences in adjustment response of the smart seat to different users, avoiding problems such as overshoot or adjustment delay, enabling the smart seat to achieve precise adjustment for different users and in different user states.

[0015] Additional aspects and advantages of embodiments of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of this application. Attached Figure Description

[0016] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, wherein: Figure 1 This is a flowchart illustrating the adaptive adjustment method of some embodiments of this application; Figure 2 This is a schematic diagram of a smart seat according to certain embodiments of this application; Figure 3 This is a flowchart illustrating the adaptive adjustment method of some embodiments of this application; Figure 4 This is a flowchart illustrating the adaptive adjustment method of some embodiments of this application; Figure 5 This is a flowchart illustrating the adaptive adjustment method of some embodiments of this application; Figure 6 This is a flowchart illustrating the adaptive adjustment method of some embodiments of this application; Figure 7 This is a flowchart illustrating the adaptive adjustment method of some embodiments of this application; Figure 8 This is a schematic diagram of the adaptive adjustment device according to certain embodiments of this application; Figure 9 This is a schematic diagram of a smart seat according to certain embodiments of this application; Figure 10 This is a schematic diagram illustrating the connection state between a computer-readable storage medium and a processor according to certain embodiments of this application. Detailed Implementation

[0017] The embodiments of this application will be further described below with reference to the accompanying drawings. The same or similar reference numerals in the drawings denote the same or similar elements or elements having the same or similar functions throughout. Furthermore, the embodiments of this application described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of this application, and should not be construed as limiting this application.

[0018] Please see Figures 1 to 3 This application provides an adaptive adjustment method for a smart seat 200. The smart seat 200 includes a sensor assembly 201 and an actuator 202. The adaptive adjustment method includes: 010: Acquire multimodal sensing data collected by sensor component 201, and perform fusion processing on the multimodal sensing data to obtain feature vectors; 020: Determine the corrected control parameters for the proportional-integral-derivative control algorithm based on eigenvectors; 030: The target adjustment amount is determined by using a proportional-integral-derivative control algorithm based on the modified control parameters; 040: Adjust actuator 202 according to the target adjustment amount; 050: Determine the hierarchical adjustment strategy based on the feature vector and perform hierarchical adjustment on the actuator 202.

[0019] In the adaptive adjustment method of this application, feature vectors are determined based on multimodal data. Corrected control parameters corresponding to the proportional-integral-derivative (PID) control algorithm are then determined based on the feature vectors. The target adjustment amount is then determined using the PID control algorithm based on the corrected control parameters to adjust the actuator 202. Furthermore, a tiered adjustment strategy is determined based on the feature vectors to perform tiered adjustment of the actuator 202. In this way, dynamic adaptive adjustment of the smart seat is achieved, adapting to different users. It also considers the differences in adjustment response of the smart seat 200 to different users, avoiding problems such as overshoot or adjustment delay, enabling the smart seat 200 to achieve precise adjustment for different users and in different user states.

[0020] Specifically, the smart seat 200 includes a sensor assembly 201, which comprises various sensors. The specific types of sensors can be determined based on the actual application. The sensor assembly 201 can collect status data of the smart seat 200 and the status data of the user sitting on the smart seat 200 to obtain multimodal sensing data.

[0021] The smart seat 200 may include a processing module. Multimodal data is transmitted to the processing module in real time via a Controller Area Network (CAN) bus. The processing module may employ a Central Processing Unit (CPU) to support parallel processing of sensor data from multiple modules. By fusing the multimodal sensor data through the processing module, feature vectors related to the seat state and user state can be obtained. Based on these feature vectors, the smart seat 200 can be adaptively adjusted to suit different users and different user states.

[0022] The smart seat 200 includes an actuator 202, and adjustment of the smart seat 200 is achieved by controlling the actuator 202. Based on eigenvectors, a proportional-integral-derivative (PID) control algorithm can be used to determine the target adjustment amount of the actuator 202. First, the corrected control parameters corresponding to the PID control algorithm can be determined based on the eigenvectors. These corrected control parameters include the proportional coefficient Kp, integral coefficient Ki, and derivative coefficient Kd. Then, based on the corrected control parameters, the PID control algorithm is used to determine the target adjustment amount. The actuator 202 can then be adjusted according to the target adjustment amount to achieve the adjustment of the smart seat 200.

[0023] After adjusting the actuator 202 according to the target adjustment amount, a hierarchical adjustment strategy can be determined based on the feature vector, and then the actuator 202 can be adjusted hierarchically using the determined hierarchical adjustment strategy. Hierarchical adjustment refers to dividing the actuator 202 into multiple adjustment levels, with higher levels resulting in greater adjustment, thereby adapting to different user needs.

[0024] In related technologies, PID controllers with fixed parameters do not take into account the differences in adjustment response among users of different body types, which can easily lead to overshoot, such as excessive lumbar support, or slow convergence, such as delayed seat depth adjustment.

[0025] In this embodiment, multimodal data is acquired, feature vectors are determined based on the multimodal data, and a PID control algorithm is used to determine the target adjustment amount to adjust the actuator 202 based on the feature vectors. Furthermore, the actuator 202 is subjected to tiered adjustment based on the feature vectors. The correction control parameters of the PID control algorithm are determined through the feature vectors, meaning that the correction control parameters differ for different users and for users in different states. Thus, considering the differences in adjustment response of the smart seat 200 to different users, while achieving adaptive adjustment of the smart seat 200, problems such as overshoot or adjustment delay are avoided, enabling the smart seat 200 to achieve precise adjustment for different users and in different user states.

[0026] In some implementations, the multimodal sensing data includes any one or more of pressure sensing data, inertial sensing data, and millimeter-wave radar data; and / or the feature vector includes any one or more of body shape vector, attitude vector, and fatigue index.

[0027] Specifically, sensor component 201 may include any one or more of pressure sensors, inertial sensors, and millimeter-wave radar. Correspondingly, multimodal sensing data may include any one or more of pressure sensing data, inertial sensing data, and millimeter-wave radar data. The feature vector obtained from processing the multimodal sensing data may include any one or more of body shape vectors, attitude vectors, and fatigue indices.

[0028] Of course, the sensor component 201 can also be equipped with other sensors according to actual application requirements, and the multimodal sensing data can also include other sensing data accordingly, so the feature vector can also include vectors that can represent other user features.

[0029] The specifications of the pressure sensor can be determined according to the actual application requirements, and there are no restrictions here.

[0030] In one example, the pressure sensor uses an area array pressure sensor, specifically a 32×32 flexible pressure array. The array covers the seat cushion (corresponding to the buttocks area) and backrest (corresponding to the waist and shoulder area) of the smart seat 200, and collects local pressure data P(x, y, t) at a frequency of 20Hz. The pressure sensing data includes local pressure data.

[0031] Pressure sensor data can be used to calculate a user's body shape parameters such as weight W, seat depth L, and hip width H. Weight W = total pressure / contact area × gravitational acceleration (g = 9.8 m / s²). 2 Seat depth L is the distance between the center of gravity at the front and rear of the seat cushion; hip width H is the distance between the peak pressure points in the left and right directions of the seat cushion. The body shape vector can be determined based on body weight W, seat depth L, and hip width H. The body shape vector B = [W, L, H], with accuracies of ±0.5kg, ±1mm, and ±1mm, respectively.

[0032] The specifications of inertial sensors can be determined according to the actual application requirements, and are not limited here.

[0033] In one example, the inertial sensor uses a 9-axis Inertial Measurement Unit (IMU). The smart seat 200 is equipped with two IMUs. The first IMU is mounted at the top of the backrest of the smart seat 200 to detect the user's torso pitch angle θ_pitch and torso roll angle θ_roll; the second IMU is mounted at the front of the seat cushion of the smart seat 200 to detect the user's pelvic rotation angle φ_obliquity; the sampling rate of both IMUs is 50Hz. The inertial sensing data includes the torso pitch angle θ_pitch, torso roll angle θ_roll, and pelvic rotation angle φ_obliquity.

[0034] Inertial sensing data can be used to determine the attitude vector. Specifically, the acquired data can be noise-reduced using Kalman filtering. The output torso pitch angle θ_pitch ranges from -15° to 15°, with forward tilt being positive and backward tilt being negative; the torso roll angle θ_roll ranges from -5° to 5°, with leftward roll being positive and rightward roll being negative; and the pelvic rotation angle φ_obliquity ranges from -3° to 3°, with forward tilt being positive and backward tilt being negative. The attitude vector A = [θ_pitch, θ_roll, φ_obliquity], with an angular accuracy of ±0.1°.

[0035] The specifications of millimeter-wave radar can be determined according to actual application requirements, and are not limited here.

[0036] In one example, a 60Hz millimeter-wave radar with a sampling rate of 10Hz is used. The millimeter-wave radar is installed in the center of the headrest of the smart seat 200 and can monitor the user's chest and abdominal micro-movements. Since these micro-movements reflect the user's respiratory fluctuations and heart rate fluctuations, the user's respiratory rate fR(t) and heart rate variability (HRV) can be extracted. The millimeter-wave radar data thus includes respiratory rate fR(t) and heart rate variability (HRV).

[0037] Millimeter-wave radar combined with pressure sensing data can be used to determine the fatigue index. Fatigue Index ;in, This represents the normalized value of heart rate variability. Measured value / The baseline value is 70ms. This is the normalized value of respiratory rate. = Measured value / standard respiratory rate at rest, where the standard respiratory rate is taken as 18 breaths / minute. This is the normalized value of the rate of change of pressure gradient. = Measured value / Threshold for pressure change after prolonged sitting; the threshold for pressure change after prolonged sitting is taken as 5 Pa / mm. Weighting coefficient. =0.4、 =0.3、 =0.3. The fatigue index F is normalized to 0~100 to facilitate subsequent graded adjustment.

[0038] In related technologies, the seats rely solely on pressure sensors to detect posture, failing to simultaneously acquire physiological fatigue signals (such as breathing and heart rate) and posture angles, resulting in a one-sided adjustment basis and difficulty in accurately matching user needs.

[0039] In this embodiment, through multimodal sensing data fusion, it can adapt to the differences in body shape and real-time posture changes of different users. The body shape adaptation accuracy is improved from 70% to 95%, and the posture angle detection error is <0.1°. It can accurately match users of different heights (150~190cm) and weights (40~100kg), expanding the applicability of the smart seat 200. It can also achieve precise adjustment for users of different body shapes, and can accurately match user needs.

[0040] Please see Figure 3 In some embodiments, before acquiring the multimodal sensing data collected by sensor component 201, the adaptive adjustment method further includes: After confirming that the smart seat 200 is powered on, initialize the smart seat 200. Control the pressure sensor to monitor pressure; When the pressure is determined to be greater than the preset value and continues for a fifth time, the control sensor assembly 201 is used to collect data.

[0041] Specifically, the initialization of the smart seat 200 includes a self-test of the sensor assembly 201 and a reset of the actuator 202. After initialization, the smart seat 200 enters a standby state, with the pressure sensor continuously operating to monitor pressure. If the pressure detected by the pressure sensor exceeds a preset value for five consecutive hours, it indicates that the user has sat down, and the smart seat 200 can then formally initiate the adjustment process. It controls the sensor assembly 201 to collect data, and after acquiring the multimodal sensing data collected by the sensor assembly 201, it can adjust the actuator 202.

[0042] The preset value and the fifth duration can be set according to actual application needs. In one example, the preset value is 50N and the fifth duration is 2s.

[0043] Please see Figure 3 and Figure 4 In some implementations, the eigenvector includes a body shape vector and a fatigue index. Determining the corrected control parameters (i.e., O2O) corresponding to the proportional-integral-derivative control algorithm based on the eigenvector includes: 021: Determine the initial control parameters corresponding to the proportional-integral-derivative control algorithm by querying the pre-stored body shape-parameter mapping table based on the body shape vector; 022: Fine-tuning the initial control parameters based on body shape vector and fatigue index to obtain corrected control parameters.

[0044] Specifically, a body shape-parameter mapping table between body shape parameters and control parameters of the PID control algorithm can be pre-stored in the intelligent seat 200. The body shape-parameter mapping table can be determined through experiments or calibration, and there are no restrictions here. Within six time intervals (e.g., 30 seconds) after the user is seated, the initial control parameters corresponding to the PID control algorithm can be determined by querying the body shape-parameter mapping table based on the currently determined body shape vector.

[0045] In this embodiment, the actuator 202 includes a lumbar support adjustment module, a seat depth adjustment module, and a shoulder support adjustment module. The lumbar support adjustment module includes a lumbar support motor, the seat depth adjustment module includes a seat depth motor, and the shoulder support adjustment module includes a shoulder support airbag. Three sets of corresponding initial control parameters can be obtained through querying. The first initial control parameter PID1 is the initial control parameter corresponding to the stroke L_lumbar of the lumbar support motor: proportional coefficient Kp1=2.5, integral coefficient Ki1=0.8, and derivative coefficient Kd1=0.3; The second initial control parameter PID2 is the initial control parameter corresponding to the seat depth motor stroke L_seat: proportional coefficient Kp2=3.0, integral coefficient Ki2=0.5, derivative coefficient Kd2=0.2; The third initial control parameter PID3 is the initial control parameter corresponding to the shoulder airbag pressure P_shoulder: proportional coefficient Kp3=1.8, integral coefficient Ki3=0.6, and derivative coefficient Kd3=0.1.

[0046] After determining the initial control parameters, they can be fine-tuned based on the body shape vector and fatigue index to obtain the corrected control parameters. The fine-tuning process is described in detail below.

[0047] Please see Figure 3 and Figure 5 In some implementations, the initial control parameters are fine-tuned based on the body shape vector and fatigue index to obtain the corrected control parameters (i.e., 022), including: 0221: Determine the error vector between the target value and the actual value of the actuator 202 based on the body shape vector and fatigue index; 0222: The initial control parameters are fine-tuned based on the error vector using a recursive least squares algorithm at preset intervals to obtain the corrected control parameters.

[0048] Specifically, the error vector between the target value and the actual value of the actuator 202 is determined based on the body shape vector and fatigue index. The error vector e(t) = [e1, e2, e3]. Here, e1 represents the error between the target value and the actual value of the lumbar support adjustment module, e1 = target lumbar support force - actual support force. The target lumbar support force is set based on the body weight W. For example, when W ≤ 50kg, the target lumbar support force is 30N; when W > 50kg, the target lumbar support force is 40N.

[0049] e2 represents the error between the target value and the actual value of the seat depth adjustment module. e2 = target seat depth - actual seat depth. Target seat depth = seat depth L × 0.8.

[0050] e3 represents the error between the target value and the actual value of the shoulder support adjustment module. e3 = target shoulder support air pressure - actual air pressure. The target shoulder support air pressure is set based on the fatigue index F. When F < 30, the target shoulder support air pressure is 0.5 kPa; when F ≥ 80, the target shoulder support air pressure is 1.2 kPa.

[0051] After determining the error vector, the initial control parameters are fine-tuned using the Recursive Least Squares (RLS) algorithm at preset intervals to obtain the corrected control parameters. The RLS algorithm corrects the control parameters of the PID control algorithm in real time by minimizing the sum of squared errors, thereby ensuring that the adjustment response time for users of different body types is ≤1s and the overshoot is <5%, avoiding abrupt adjustments that could lead to a poor user experience.

[0052] After determining the corrected control parameters for the PID control algorithm, the target adjustment amount of the actuator 202 can be determined based on the PID control algorithm and the corrected control parameters, and the actuator 202 can be adjusted according to the target adjustment amount. The adjustment process is described in detail below.

[0053] Please see Figure 3 In some embodiments, the actuator 202 includes a shoulder support adjustment module, a lumbar support adjustment module, and a seat depth adjustment module, and the target adjustment amount includes a shoulder support adjustment amount, a lumbar support adjustment amount, and a seat depth adjustment amount. Adjusting the actuator 202 (i.e., 040) according to the target adjustment amount includes: Adjust the shoulder support adjustment module according to the shoulder support adjustment amount; and / or Adjust the lumbar support adjustment module according to the lumbar support adjustment amount; and / or Adjust the seat depth adjustment module according to the seat depth adjustment amount.

[0054] Specifically, the actuator 202 includes a shoulder support adjustment module, a lumbar support adjustment module, and a seat depth adjustment module, which enable multi-dimensional adjustment. The lumbar support motor in the lumbar support adjustment module can be a set of silent linear motors, the seat depth motor in the seat depth adjustment module can be a set of silent linear motors, and the shoulder support airbags in the shoulder support adjustment module can include four low-pressure airbags, two for each shoulder support. Adjustment is performed based on the PID output signal, with specific parameters as follows: Lumbar support adjustment range: linear motor stroke 0-40mm, thrust 150N, speed ≤6mm / s, acceleration ≤0.8m / s². 2 The lumbar support motor stroke (i.e., lumbar support adjustment amount) L_lumbar is output based on the modified control parameters to adjust the lumbar support pushing depth in real time. In one example, when the fatigue index F=60, the pushing depth is 25mm.

[0055] Seat depth adjustment range: linear motor stroke 0-60mm, speed ≤5 mm / s, acceleration ≤0.8 m / s². 2 The seat depth motor stroke (i.e., seat depth adjustment amount) L_seat is output based on the corrected control parameters to adjust the fore-and-aft position of the seat cushion in real time. In one example, when the seat depth L=500mm, the target seat depth is 400mm.

[0056] Shoulder support adjustment range: bladder volume 3L, air pressure adjustment range 0~15kPa, air pressure change rate ≤2kPa / s (to avoid discomfort caused by excessive inflation). The shoulder support airbag pressure (i.e., shoulder support adjustment amount) P_shoulder_L / R is output according to the corrected control parameters to independently control the left and right shoulder support air pressures. In one example, when the torso tilt angle θ_roll = 3°, the left airbag pressure is 0.3kPa higher than the right.

[0057] In related technologies, the seats only support basic adjustments such as height adjustment and backrest tilt, lacking refined control over lumbar support and shoulder support. Furthermore, the acceleration of adjustment movements is unrestrained, which can easily cause abruptness. For example, excessively rapid airbag inflation can lead to discomfort.

[0058] In this embodiment, a multi-dimensional execution structure is set up. The execution mechanism 202 includes a shoulder support adjustment module, a lumbar support adjustment module, and a seat depth adjustment module, which can realize precise control of seat depth, lumbar support, and shoulder support. Moreover, the acceleration of the adjustment action is constrained to avoid abrupt adjustment that may cause a poor user experience.

[0059] Please see Figure 3 and Figure 6 In some implementations, the eigenvector includes a fatigue index. A graded adjustment strategy is determined based on the eigenvector, and the actuator 202 is adjusted in a graded manner (i.e., 050), including: 051: If the fatigue index is determined to be less than the first threshold, the actuator 202 shall be adjusted to the first level. 052: If the fatigue index is determined to be greater than the first threshold and less than the second threshold, perform secondary adjustment on the actuator 202; 053: If the fatigue index is determined to be greater than the second threshold and less than the third threshold, the actuator 202 shall be adjusted to level three. 054: If the fatigue index is determined to be greater than the third threshold, the actuator 202 shall be adjusted to level four.

[0060] In some embodiments, primary adjustment of the actuator 202 includes: Control actuator 202 to maintain the current state; and / or Secondary regulation of implementing agency 202, including: The control actuator 202 performs adjustments on a portion of its adjustment modules for a first duration; and / or The three-level adjustment of the implementing agency 202 includes: The control actuator 202 performs adjustments on a portion of its adjustment module for a second duration; and / or The four-level adjustment of the implementing agency 202 includes: All adjustment modules in the control actuator 202 are adjusted and this continues for a third duration; The first duration is shorter than the second duration, and the second duration is shorter than the third duration.

[0061] Specifically, a four-level adjustment strategy is triggered based on the value of the fatigue index F: First-level adjustment: When the fatigue index F is determined to be less than a first threshold, the actuator 202 is adjusted accordingly. The first threshold can be determined based on the actual application. In one example, the first threshold is 30, meaning that when F < 30, the actuator 202 is adjusted. First-level adjustment is a non-perceptible maintenance. In this case, the actuator 202 maintains its current state, only monitoring changes in posture through the pressure sensor, without actively adjusting, and is imperceptible to the user.

[0062] Secondary adjustment: When the fatigue index is determined to be greater than the first threshold and less than the second threshold, secondary adjustment is performed on actuator 202. In this case, some adjustment modules within actuator 202 participate in the adjustment and continue for a first duration. Both the second threshold and the first duration can be determined based on actual application conditions. In one example, the second threshold is 60, and the first duration is ≤3s, i.e., when 30≤F<60, secondary adjustment is performed on actuator 202. Secondary adjustment involves micro-pressure redistribution. For example, the lumbar support motor makes a micro-adjustment at a speed of 2mm / s (stroke ±5mm), and the shoulder support airbag slowly inflates and deflates at a speed of 0.5kPa / s to achieve localized pressure redistribution (e.g., a 10% reduction in peak hip pressure). The execution duration is ≤3s, and the user experiences only slight discomfort.

[0063] Level 3 Adjustment: When the fatigue index is determined to be greater than the second threshold and less than the third threshold, the actuator 202 undergoes level 3 adjustment. In this case, some adjustment modules within the actuator 202 participate in the adjustment and continue for a second duration, which is longer than the first duration. Both the third threshold and the second duration can be determined based on actual application conditions. In one example, the third threshold is 80, and the second duration is ≤10s, i.e., 60≤F<80. The actuator 202 undergoes level 3 adjustment, which is a coordinated relief mechanism. For example, the lumbar support motor activates a "push-relax" cycle (5s cycle, 30~40N push force), while the shoulder support airbags alternately inflate (left side → right side, 2s interval). Simultaneously, a 30s micro-massage pulse (2Hz frequency) can be added, with an execution duration ≤10s, resulting in noticeable user perception.

[0064] Level 4 Adjustment: When the fatigue index is determined to be greater than the third threshold, Level 4 adjustment is performed on actuator 202. At this time, all adjustment modules in actuator 202 participate in the adjustment and continue for a third duration, which is longer than the second duration. The third duration can be determined according to the actual application. In one example, when the third duration is ≤15s, that is, when F≤80, Level 4 adjustment is performed on actuator 202. Level 4 adjustment is a full-dimensional intervention. For example, the lumbar support motor actuates to increase the lumbar support pushing depth to 35-40mm, the seat depth motor actuates to adjust the seat depth backward by 5-10mm, and the shoulder support airbag pressure increases to 1.0~1.2kPa. At the same time, a voice reminder (such as "It is recommended to stand and move around for 1 minute") can be triggered. The execution duration is ≤15s, and the user perception is strong.

[0065] In related technologies, no correlation is established between fatigue level and adjustment strategy, making it impossible to dynamically switch the adjustment intensity according to the user's fatigue level, resulting in poor relief effect.

[0066] In this embodiment, a graded adjustment strategy is determined based on the fatigue index, which effectively alleviates user fatigue with good results. Studies have shown that it can reduce the subjective fatigue score after 4 hours of sitting from 6.8 / 10 to 4.2 / 10, improving the relief effect by 38%.

[0067] Please see Figure 3 and Figure 7 In some embodiments, sensor assembly 201 includes a pressure sensor, an inertial measurement unit, and a millimeter-wave radar. After determining a graded adjustment strategy based on the eigenvector and performing graded adjustment (i.e., 050) on actuator 202, the adaptive adjustment method further includes: 060: The pressure of the smart seat 200 is monitored in real time via a pressure sensor; 070: If it is determined that the pressure is less than the pressure threshold and continues for a fourth time, the control actuator 202 is reset, the control inertial measurement unit and millimeter-wave radar are turned off, and the pressure sensor is kept running.

[0068] Specifically, during the operation of the smart seat 200, the pressure sensor monitors the pressure on the smart seat 200 in real time. The pressure can be the sum of the pressure from the seat cushion and the backrest. If the pressure is less than the pressure threshold and remains below it for four hours, it indicates that the user has left the seat. At this time, the actuator 202 can be reset, the inertial measurement unit and millimeter-wave radar can be turned off, and the pressure sensor can be kept running. The reset of the actuator 202 includes zeroing the lumbar support motor, deflating the shoulder support airbag to 0 kPa, and restoring the seat depth to the initial position.

[0069] The pressure sensor remains operational to continuously monitor the pressure of the smart seat 200. This allows the sensor to detect when the user sits down again, return the multimodal sensing data collected by the sensor assembly 201, fuse the multimodal sensing data to obtain a feature vector, and then restart the adjustment process. Furthermore, the pressure sensor has low power consumption, less than 0.5W, which helps reduce the power consumption of the smart seat 200 by monitoring user occupancy.

[0070] Both the pressure threshold and the fourth duration can be determined based on the actual application. In one example, the pressure threshold is determined based on the weight of the currently seated user and is set to 10% of the user's weight, and the fourth duration is 5 seconds.

[0071] In the adaptive adjustment method of this application, a single microcontroller unit (MCU) + edge algorithm can be used, which eliminates the need for high-performance computing boards and cloud support, reduces system cost by 35%, and has a standby power consumption of <0.5W, meeting energy-saving standards.

[0072] The working process of the adaptive adjustment method of this application in two embodiments is described in detail below.

[0073] Example 1: Adult male user, weight 70kg, height 175cm, seat depth 520mm.

[0074] Data acquisition phase: The pressure sensor detected a total pressure of 686 N (i.e., 70 kg × 9.8), and calculated the hip width to be 420 mm; the inertial sensor output the trunk pitch angle θ_pitch = 5° (slight forward tilt) and the pelvic rotation angle φ_obliquity = 2°; the millimeter-wave radar detected a respiratory rate fR = 16 breaths / minute and a heart rate variability HRV = 65 ms.

[0075] Feature vector calculation: Body shape vector B = [70kg, 520mm, 420mm]; Posture vector A = [5°, 0°, 2°]; Fatigue index F = 0.4×(65 / 70) + 0.3×(16 / 18) + 0.3×(3 / 5) = 0.37 + 0.27 + 0.18 = 82.

[0076] PID tuning: The initial control parameters are obtained from the table as Kp1=2.8, Ki1=0.9, Kd1=0.35; after RLS fine-tuning, Kp1=2.9, Ki1=0.85, Kd1=0.32, and the error is e1=40N-35N=5N.

[0077] Execution adjustment (multi-dimensional execution structure adjustment + four-level adjustment): lumbar support push depth 38mm, speed 5mm / s, seat depth adjusted to 416mm (520mm×0.8), shoulder support airbag pressure 1.1kPa, change rate 1.8kPa / s, simultaneously triggering voice reminder, execution duration 12s.

[0078] Seat removal procedure: After the user leaves the seat, if the total pressure is less than 68.6N (70kg×10%×9.8) and lasts for 5 seconds, actuator 202 will reset and smart seat 200 will go into sleep mode.

[0079] Example 2: Adult female user, weight 50kg, height 160cm, seat depth 480mm.

[0080] Data acquisition phase: The pressure sensor detected a total pressure of 490N and calculated a hip width of 380mm; the inertial sensor output a trunk pitch angle θ_pitch=-3° (slight backward tilt) and a pelvic rotation angle φ_obliquity=-1°; the millimeter-wave radar detected a respiratory rate fR=18 breaths / minute and a heart rate variability HRV=72ms.

[0081] Feature vector calculation: Body shape vector B = [50kg, 480mm, 380mm]; Posture vector A = [-3°, 0°, -19°]; Fatigue index F = 0.4×(72 / 70) + 0.3×(18 / 18) + -0.3×(1 / 5) = 0.41 + 0.3 + 0.06 = 77.

[0082] PID tuning: Initial Kp1=2.4, Ki1=0.7, Kd1=0.25; after RLS fine-tuning, Kp1=2.3, Ki1=0.75, Kd1=0.28, error e1=30N-28N=2N.

[0083] Execution adjustment (multi-dimensional execution structure adjustment + three-level adjustment): lumbar support push depth 32mm, seat depth 400mm (480mm×0.83), shoulder support airbag pressure 0.9kpa, massage pulse 30s, execution duration 8s.

[0084] Please see Figure 3 and Figure 8 This application describes an adaptive adjustment device 100 for an intelligent seat 200. The intelligent seat 200 includes an actuator 202, and the adaptive adjustment device 100 includes a data processing module 10, a first determination module 20, a second determination module 30, a first adjustment module 40, and a second adjustment module 50. The data processing module 10 acquires multimodal sensing data collected by the sensor assembly 201 and performs fusion processing on the multimodal sensing data to obtain a feature vector. The first determination module 20 determines the corrected control parameters corresponding to the proportional-integral-derivative (PID) control algorithm based on the feature vector. The second determination module 30 determines the target adjustment amount using the PID control algorithm based on the corrected control parameters. The first adjustment module 40 adjusts the actuator 202 according to the target adjustment amount. The second adjustment module 50 determines a graded adjustment strategy based on the feature vector and performs graded adjustment on the actuator 202.

[0085] In some implementations, the feature vector includes a body shape vector and a fatigue index. The first determining module 20 is specifically used to query a pre-stored body shape-parameter mapping table based on the body shape vector to determine the initial control parameters corresponding to the proportional-integral-derivative control algorithm; and to fine-tune the initial control parameters based on the body shape vector and the fatigue index to obtain the corrected control parameters.

[0086] In some implementations, the first determining module 20 is specifically used to determine the error vector between the target adjustment amount and the actual adjustment amount of the actuator 202 based on the body shape vector and the fatigue index; and to fine-tune the initial control parameters according to the error vector using a recursive least squares algorithm at preset intervals to obtain the corrected control parameters.

[0087] In some implementations, the feature vector includes a fatigue index. The second adjustment module 50 is specifically configured to perform a first-level adjustment on the actuator 202 when the fatigue index is determined to be less than a first threshold; a second-level adjustment on the actuator 202 when the fatigue index is determined to be greater than the first threshold and less than a second threshold; a third-level adjustment on the actuator 202 when the fatigue index is determined to be greater than the second threshold and less than a third threshold; and a fourth-level adjustment on the actuator 202 when the fatigue index is determined to be greater than a third threshold.

[0088] In some embodiments, the second adjustment module 50 is specifically used to control the actuator 202 to maintain its current state; and / or the second adjustment module 50 is specifically used to control some adjustment modules in the actuator 202 to adjust for a first duration; and / or the second adjustment module 50 is specifically used to control some adjustment modules in the actuator 202 to adjust for a second duration; and / or the second adjustment module 50 is specifically used to control all adjustment modules in the actuator 202 to adjust for a third duration. Wherein, the first duration is shorter than the second duration, and the second duration is shorter than the third duration.

[0089] In some embodiments, sensor assembly 201 includes a pressure sensor, an inertial measurement unit, and a millimeter-wave radar. The adaptive adjustment device 100 also includes a reset / sleep module. After determining a graded adjustment strategy based on feature vectors and performing graded adjustment on actuator 202, the reset / sleep module monitors the pressure of the smart seat 200 in real time via the pressure sensor; if it determines that the pressure is below a pressure threshold for a sustained period of four hours, it controls actuator 202 to reset, controls the inertial measurement unit and millimeter-wave radar to shut down, and controls the pressure sensor to remain operational.

[0090] It should be noted that the explanation of the adaptive adjustment method in the foregoing embodiments also applies to the adaptive adjustment device 100 of the embodiments of this application, and will not be elaborated here.

[0091] Please see Figure 3 and Figure 9 This application also provides an intelligent seat 200. The intelligent seat 200 includes one or more processors 210 and a memory 220. The memory 220 stores a computer program, which, when executed by the processor 210, implements the adaptive adjustment method of any of the above embodiments.

[0092] For example, when a computer program is executed by processor 210, the following adaptive adjustment method is implemented: 010: Acquire multimodal sensing data collected by sensor component 201, and perform fusion processing on the multimodal sensing data to obtain feature vectors; 020: Determine the corrected control parameters for the proportional-integral-derivative control algorithm based on eigenvectors; 030: The target adjustment amount is determined by using a proportional-integral-derivative control algorithm based on the modified control parameters; 040: Adjust actuator 202 according to the target adjustment amount; 050: Determine the hierarchical adjustment strategy based on the feature vector and perform hierarchical adjustment on the actuator 202.

[0093] For example, when a computer program is executed by processor 210, the following adaptive adjustment method is implemented: 021: Determine the initial control parameters corresponding to the proportional-integral-derivative control algorithm by querying the pre-stored body shape-parameter mapping table based on the body shape vector; 022: Fine-tuning the initial control parameters based on body shape vector and fatigue index to obtain corrected control parameters.

[0094] It should be noted that the explanations of the adaptive adjustment method and adaptive adjustment device 100 in the foregoing embodiments also apply to the smart seat 200 of the embodiments of this application, and will not be elaborated here.

[0095] Please see Figure 10 This application also provides a computer-readable storage medium 300 storing a computer program 310 thereon. When the program is executed by the processor 320, it implements the adaptive adjustment method of any of the above embodiments.

[0096] For example, when the program is executed by processor 320, the following adaptive adjustment method is implemented: 010: Acquire multimodal sensing data collected by sensor component 201, and perform fusion processing on the multimodal sensing data to obtain feature vectors; 020: Determine the corrected control parameters for the proportional-integral-derivative control algorithm based on eigenvectors; 030: The target adjustment amount is determined by using a proportional-integral-derivative control algorithm based on the modified control parameters; 040: Adjust actuator 202 according to the target adjustment amount; 050: Determine the hierarchical adjustment strategy based on the feature vector and perform hierarchical adjustment on the actuator 202.

[0097] For example, when the program is executed by processor 320, the following adaptive adjustment method is implemented: 021: Determine the initial control parameters corresponding to the proportional-integral-derivative control algorithm by querying the pre-stored body shape-parameter mapping table based on the body shape vector; 022: Fine-tuning the initial control parameters based on body shape vector and fatigue index to obtain corrected control parameters.

[0098] It should be noted that the explanations of the adaptive adjustment method and adaptive adjustment device 100 in the foregoing embodiments also apply to the computer-readable storage medium 300 of the embodiments of this application, and will not be elaborated here.

[0099] In summary, the adaptive adjustment method, adaptive adjustment device 100, smart seat 200, and computer-readable storage medium 300 of the intelligent seat 200 in this application embodiment determine feature vectors based on multimodal data, determine corrected control parameters corresponding to the proportional-integral-derivative (PID) control algorithm based on the feature vectors, determine the target adjustment amount using the PID control algorithm based on the corrected control parameters, and adjust the actuator 202 accordingly. Furthermore, a graded adjustment strategy is determined based on the feature vectors to perform graded adjustment of the actuator 202. Thus, dynamic adaptive adjustment of the intelligent seat is achieved, adapting to different users. It also considers the differences in adjustment response of the intelligent seat 200 to different users, avoiding overshoot or adjustment delay, enabling the intelligent seat 200 to achieve precise adjustment for different users and in different user states.

[0100] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0101] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the function involved, as will be understood by those skilled in the art to which embodiments of this application pertain.

[0102] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, a computer-readable storage medium can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable storage medium could be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0103] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0104] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments. Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc.

[0105] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. An adaptive adjustment method for an intelligent seat, characterized in that, The smart seat includes sensor components and actuators, and the adaptive adjustment method includes: The multimodal sensing data collected by the sensor components is acquired, and the multimodal sensing data is fused to obtain a feature vector. Based on the aforementioned feature vectors, determine the corrected control parameters corresponding to the proportional-integral-derivative control algorithm; The target adjustment amount is determined based on the modified control parameters using the proportional-integral-derivative control algorithm. Adjust the actuator according to the target adjustment amount; A hierarchical adjustment strategy is determined based on the feature vector, and the actuator is adjusted hierarchically.

2. The adaptive adjustment method according to claim 1, characterized in that, The multimodal sensing data includes any one or more of pressure sensing data, inertial sensing data, and millimeter-wave radar data; and / or The feature vector includes any one or more of the following: body shape vector, posture vector, and fatigue index.

3. The adaptive adjustment method according to claim 1, characterized in that, The feature vector includes a body shape vector and a fatigue index. The step of determining the corrected control parameters corresponding to the proportional-integral-derivative control algorithm based on the feature vector includes: The initial control parameters corresponding to the proportional-integral-derivative control algorithm are determined by querying the pre-stored body shape-parameter mapping table based on the body shape vector. The initial control parameters are fine-tuned based on the body shape vector and the fatigue index to obtain the corrected control parameters.

4. The adaptive adjustment method according to claim 3, characterized in that, The process of fine-tuning the initial control parameters based on the body shape vector and the fatigue index to obtain corrected control parameters includes: The error vector between the target value and the actual value of the actuator is determined based on the body shape vector and the fatigue index. The initial control parameters are fine-tuned based on the error vector using a recursive least squares algorithm at preset intervals to obtain the corrected control parameters.

5. The adaptive adjustment method according to claim 1, characterized in that, The actuator includes a shoulder support adjustment module, a lumbar support adjustment module, and a seat depth adjustment module. The target adjustment amount includes a shoulder support adjustment amount, a lumbar support adjustment amount, and a seat depth adjustment amount. Adjusting the actuator according to the target adjustment amount includes: Adjust the shoulder support adjustment module according to the shoulder support adjustment amount; and / or Adjust the lumbar support adjustment module according to the lumbar support adjustment amount; and / or Adjust the seat depth adjustment module according to the seat depth adjustment amount.

6. The adaptive adjustment method according to claim 1, characterized in that, The feature vector includes a fatigue index. The step of determining a graded adjustment strategy based on the feature vector and performing graded adjustment on the actuator includes: If the fatigue index is determined to be less than a first threshold, the actuator is adjusted to the first level. If the fatigue index is determined to be greater than the first threshold and less than the second threshold, the actuator is subjected to secondary adjustment. If the fatigue index is determined to be greater than the second threshold and less than the third threshold, the actuator is adjusted in three levels. If the fatigue index is determined to be greater than the third threshold, the actuator is adjusted to level four.

7. The adaptive adjustment method according to claim 5, characterized in that, The first-level adjustment of the actuator includes: Control the actuator to maintain its current state; and / or The secondary adjustment of the actuator includes: Control a portion of the adjustment module in the actuator to make adjustments, and continue for a first duration; and / or The three-level adjustment of the actuator includes: Control a portion of the adjustment module in the actuator to make adjustments, and continue for a second duration; and / or The four-level adjustment of the actuator includes: The actuator controls all adjustment modules to adjust, and this continues for a third duration; Wherein, the first duration is less than the second duration, and the second duration is less than the third duration.

8. The adaptive adjustment method according to claim 1, characterized in that, The sensor assembly includes a pressure sensor, an inertial measurement unit, and a millimeter-wave radar. After determining the hierarchical adjustment strategy based on the feature vector and performing hierarchical adjustment on the actuator, the adaptive adjustment method further includes: The pressure of the smart seat is monitored in real time using the pressure sensor. If it is determined that the pressure is less than the pressure threshold and this condition persists for a fourth time, the actuator is controlled to reset, the inertial measurement unit and the millimeter-wave radar are controlled to shut down, and the pressure sensor is controlled to remain operational.

9. An adaptive adjustment device for an intelligent seat, characterized in that, The smart seat includes sensor components and actuators, and the adaptive adjustment device includes: The data processing module is used to acquire multimodal sensing data collected by the sensor components and to perform fusion processing on the multimodal sensing data to obtain feature vectors. The first determining module is used to determine the corrected control parameters corresponding to the proportional-integral-derivative control algorithm based on the feature vector; The second determining module is used to determine the target adjustment amount based on the corrected control parameters using the proportional-integral-derivative control algorithm. The first adjustment module is used to adjust the actuator according to the target adjustment amount; The second adjustment module is used to determine a graded adjustment strategy based on the feature vector and to perform graded adjustment on the actuator.

10. A smart seat, characterized in that, The intelligent seat includes one or more processors and a memory, the memory storing a computer program that, when executed by the processor, implements the adaptive adjustment method according to any one of claims 1-8.

Images