A spinal curvature monitoring method of an intelligent seat and a vehicle-mounted health cloud system

Abstract

This invention discloses a method for monitoring spinal curvature in a smart seat and an in-vehicle health cloud system. The method includes: real-time acquisition of spinal curvature data of an occupant sitting in the seat and vehicle acceleration during vehicle movement; demodulation and curvature reconstruction of the real-time acquired spinal curvature data to obtain a three-dimensional spinal curvature curve of the occupant; and, based on a vibration-posture coupled fatigue algorithm and combined with the real-time acquired vehicle acceleration, determining the risk level of spinal injury to the occupant during vehicle movement; and triggering an intervention strategy corresponding to the risk level. This invention constructs a three-dimensional spinal curvature curve based on the occupant's spinal curvature data, accurately identifies the occupant's sitting posture, and, combined with vehicle acceleration during vehicle movement, calculates the cumulative fatigue caused by the superposition of vehicle vibration and poor posture in real time. This effectively identifies the risk of chronic injury caused by the long-term superposition of vehicle vibration and poor posture, achieving accurate predictive risk assessment before pain occurs.

Description

Technical Field

[0001] This invention relates to the field of smart seat technology, and in particular to a method for monitoring spinal curvature in a smart seat and an in-vehicle health cloud system. Background Technology

[0002] With the development of intelligent vehicles and people's travel needs, long-distance driving has become a common application scenario, and the requirements for the comfort of car seats are getting higher and higher. Existing car seats have functions such as heating, ventilation, massage, and memory, and are equipped with basic lumbar support adjustment.

[0003] While existing smart car seats have introduced posture monitoring methods to determine the type of occupant's sitting posture or identify fatigue, they are still limited to the macroscopic posture recognition level, such as leaning forward, leaning back, and leaning to the side, and have not achieved quantitative detection of the physiological curvature of the human spine. Moreover, existing fatigue monitoring is mostly based on single-dimensional monitoring of driving behavior analysis or physiological characteristic information. During long-term driving, given the diversity of occupant sitting postures, it is difficult to form an effective means of predicting the cumulative musculoskeletal damage caused by the combination of long-term driving and poor sitting posture, and it is difficult to intervene before spinal pain occurs. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a method for monitoring spinal curvature in an intelligent seat and an in-vehicle health cloud system.

[0005] This invention is achieved by the following technical solution:

[0006] A method for monitoring spinal curvature in a smart chair, comprising:

[0007] Real-time data collection of spinal curvature and vital signs of occupants while seated, as well as vehicle acceleration during vehicle movement;

[0008] The real-time acquired spinal curvature data is demodulated and reconstructed to obtain the three-dimensional curvature curve of the occupant's spine.

[0009] Based on the three-dimensional curvature curve of the occupant's spine and the real-time collected vehicle acceleration, the risk level of spinal injury to the occupant during vehicle operation is determined using a vibration-attitude coupled fatigue algorithm.

[0010] Based on the risk level of spinal injury to the occupant, an intervention strategy corresponding to that risk level is triggered.

[0011] By adopting the above technical solution, a three-dimensional curvature curve of the spine is constructed based on the occupant's spinal curvature data to obtain the actual state of the occupant's spine in the sagittal and coronal planes. This accurately identifies the occupant's sitting posture and, combined with the vehicle acceleration during vehicle movement, uses a vibration-posture coupled fatigue algorithm to calculate the cumulative fatigue caused by the superposition of vehicle vibration and poor sitting posture in real time. This effectively identifies the risk of chronic injury caused by the long-term superposition of vehicle vibration and poor sitting posture, and achieves accurate predictive risk assessment before pain occurs.

[0012] As described above, a method for monitoring spinal curvature in a smart seat, wherein demodulating and reconstructing the real-time acquired spinal curvature data to obtain the occupant's three-dimensional spinal curvature curve includes:

[0013] The real-time acquired spinal curvature data is demodulated using a demodulator to calculate the physical strain value of each discrete monitoring point.

[0014] Based on the physical strain values ​​at each discrete monitoring point, the curvature data of the spine support region of the seat is derived using the Euler-Bernoulli beam theory.

[0015] Based on the curvature data of the spine support area of ​​the seat, spatial curve integration and fitting are performed to obtain a continuous and differentiable spatial curve.

[0016] Based on the occupant's vital signs, the normal vector of the continuous differentiable spatial curve is offset to obtain the occupant's unique three-dimensional spinal curvature curve.

[0017] The spinal curvature monitoring method for a smart chair, as described above, involves deriving curvature data of the spinal support region of the chair based on the Euler-Bernoulli beam theory, using the physical strain values ​​at each discrete monitoring point.

[0018] The curvature of each discrete monitoring point is calculated using the following formula:

[0019] ;

[0020] In the formula, Let be the curvature of the i-th discrete monitoring point. Let h be the strain value at the i-th discrete monitoring point, and h be the vertical distance between the fiber core and the bending neutral layer.

[0021] Based on the curvature of each discrete monitoring point, the tangent vector and normal vector of each discrete monitoring point in three-dimensional space are recursively calculated using the Flyner-Serrey formula. The specific calculation formula is as follows:

[0022] ;

[0023] In the formula, Let be the principal normal vector of the i-th discrete monitoring point. Let be the tangent vector of the i-th discrete monitoring point. Let be the binormal vector of the i-th discrete monitoring point. Let be the torque of the i-th discrete monitoring point.

[0024] As described above, a method for monitoring spinal curvature in a smart chair, wherein the step of integrating and fitting a spatial curve based on the curvature data of the spinal support area of ​​the chair to obtain a continuously differentiable spatial curve includes:

[0025] Based on the curvature data, the change in tangential angle of each spinal support area of ​​the seat is calculated. The specific calculation formula is as follows:

[0026] ;

[0027] In the formula, This represents the change in pitch angle and tangent angle. The rate of change of yaw angle and tangent angle. Let S be the curvature of the spinal support area S in the seat;

[0028] The tangent of each spinal support region is transformed by coordinate integration with the two-dimensional coordinates of each discrete monitoring point within the corresponding region to obtain the three-dimensional coordinates of each discrete monitoring point in the three-dimensional coordinate system. The specific integration transformation process is as follows:

[0029] ;

[0030] In the formula, The pitch angle corresponds to the sagittal curvature of the spine. This is the yaw angle, corresponding to the coronal scoliosis of the spine;

[0031] By using cubic spline interpolation, the three-dimensional coordinates of each discrete monitoring point are smoothly fitted to generate a continuous and differentiable spatial curve.

[0032] As described above, a method for monitoring spinal curvature in a smart seat, wherein the risk level of spinal injury to the occupant during vehicle operation is determined based on the occupant's three-dimensional spinal curvature curve and real-time vehicle acceleration, using a vibration-attitude coupled fatigue algorithm, includes:

[0033] The median filtering method was used to remove the impulse noise caused by instantaneous body shaking from the three-dimensional curvature curve of the occupant's spine, resulting in a smooth three-dimensional curvature curve of the spine. A low-pass filter was used to remove the high-frequency mechanical noise of the vehicle from the vehicle acceleration.

[0034] The vehicle acceleration after noise filtering is frequency-weighted to obtain frequency-weighted acceleration;

[0035] Based on the three-dimensional curvature curve of the occupant's spine and the frequency-weighted acceleration after noise filtering, calculate the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment.

[0036] Based on the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment, the risk level of spinal injury to the occupant during vehicle operation is determined.

[0037] As described above, the method for monitoring spinal curvature in a smart seat, wherein the step of performing frequency-weighted processing on the vehicle acceleration after noise filtering to obtain frequency-weighted acceleration includes:

[0038] The time-domain data of vehicle acceleration after noise filtering is converted into frequency-domain data by short-time Fourier transform.

[0039] Based on the ISO 2631 standard for evaluating whole-body vibration, a weighting function is generated, and the maximum value of the weighting function is taken when the vibration frequency is between 4Hz and 8Hz.

[0040] The frequency-weighted acceleration is obtained by weighting the vehicle's acceleration based on the frequency domain data and a weighting function. The specific calculation process is as follows:

[0041] ;

[0042] In the formula, For weighted functions, This is the frequency domain data for vehicle acceleration.

[0043] The spinal curvature monitoring method for a smart seat, as described above, involves calculating the comprehensive pressure on the occupant's spine and the cumulative spinal fatigue index at the current moment based on the three-dimensional spinal curvature curve and frequency-weighted acceleration of the occupant after noise filtering. This includes:

[0044] Based on the three-dimensional curvature curve of the occupant's spine after noise filtering, the angle deviation from the ideal spinal reference angle is calculated to obtain the angle attenuation coefficient. The specific calculation process is as follows:

[0045] ;

[0046] In the formula, in the formula, The angular attenuation coefficient, This is the attitude sensitivity coefficient. It is a non-linear amplification index. This is the angular deviation value;

[0047] Among them, the angle deviation value for ;

[0048] In the formula, The time series of each spinal segment containing the three-dimensional curvature of the spine; The ideal spinal reference angle;

[0049] Based on the frequency-weighted acceleration and the angle attenuation coefficient, the instantaneous comprehensive pressure on the occupant's spine at the current moment is calculated. The specific calculation process is as follows:

[0050] ;

[0051] In the formula, For instantaneous comprehensive pressure, Static load weight, Dynamic shearing weights;

[0052] Based on the instantaneous comprehensive pressure on the occupant's spine at the current moment, and combined with biological metabolic recovery factors, the cumulative fatigue index of the occupant's spine at the current moment is calculated. The specific calculation process is as follows:

[0053] ;

[0054] In the formula, The cumulative spinal fatigue index of the occupants at the current moment. For attenuation term, This represents the muscle recovery time constant.

[0055] As described above, a method for monitoring spinal curvature in a smart seat, wherein determining the risk level of spinal injury to the occupant during vehicle operation based on the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative spinal fatigue index of the occupant at the current moment includes:

[0056] By using a preset fatigue index threshold, the continuously changing cumulative spinal fatigue index of the occupant at the current moment is mapped to a discrete risk level, which is divided into an extremely high risk level, a moderate risk level, and a safe level.

[0057] If the instantaneous combined pressure on the occupant's spine at the current moment is greater than or equal to the instantaneous danger threshold, the risk level of injury to the occupant's spine at this moment is determined to be extremely high risk.

[0058] If the instantaneous comprehensive pressure on the occupant's spine at the current moment is less than or equal to the instantaneous danger threshold, but greater than or equal to the instantaneous warning threshold, the risk level of spinal injury to the occupant at this moment is determined to be moderate risk.

[0059] If the instantaneous comprehensive pressure on the occupant's spine at the current moment is less than the instantaneous warning threshold, the risk level of spinal injury to the occupant at this moment is determined to be a safe level.

[0060] As described above, a method for monitoring spinal curvature in a smart seat, wherein triggering an intervention strategy corresponding to the risk level of spinal injury to the occupant, includes:

[0061] When the risk level of spinal injury to the occupant is extremely high, vibration reduction intervention is performed on the vehicle and the support force is increased in the spinal support area of ​​the seat.

[0062] When the risk level of spinal injury to the occupant is at the moderate risk level, an early warning will be issued, and the support force in the spinal support area of ​​the seat will be increased.

[0063] Furthermore, to achieve the above objectives, the present invention also provides an in-vehicle health cloud system, which is used to implement the spinal curvature monitoring method described above, and the in-vehicle health cloud system includes:

[0064] The acquisition module includes a flexible fiber optic grating sensor array disposed inside the seat back for real-time acquisition of spinal curvature data of the occupant sitting on the seat. The acquisition module also includes an inertial measurement unit fixed to the seat frame for real-time acquisition of vehicle acceleration during vehicle movement.

[0065] The vehicle-mounted health brain module includes a modeling unit, a spinal injury judgment unit, and an execution unit.

[0066] The modeling unit is used to demodulate and reconstruct the curvature of the spine in real time to obtain the three-dimensional curvature curve of the occupant's spine.

[0067] The spinal injury assessment unit is used to determine the risk level of spinal injury to the occupant during vehicle operation based on the three-dimensional curvature curve of the occupant's spine and the real-time collected vehicle acceleration, using a vibration-attitude coupled fatigue algorithm.

[0068] The execution unit is used to trigger an intervention strategy corresponding to the risk level of spinal injury suffered by the occupant.

[0069] Compared with existing technologies, the spinal curvature monitoring method for smart seats and the in-vehicle health cloud system proposed in this invention have the following beneficial effects:

[0070] 1. The intelligent seat spinal curvature monitoring method proposed in this invention constructs a three-dimensional spinal curvature curve based on the occupant's spinal curvature data to obtain the actual state of the occupant's spine in the sagittal and coronal planes, accurately identify the occupant's sitting posture, and, combined with the vehicle acceleration during vehicle movement, calculates the cumulative fatigue caused by the superposition of vehicle vibration and poor sitting posture in real time through a vibration-posture coupled fatigue algorithm. This effectively identifies the risk of chronic injury caused by the long-term superposition of vehicle vibration and poor sitting posture, and achieves accurate predictive risk assessment before pain occurs. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0072] Figure 1 This is a flowchart of the spinal curvature monitoring method of the present invention;

[0073] Figure 2 for Figure 1 Flowchart of the specific method for step S20;

[0074] Figure 3 for Figure 2 Flowchart of the specific method for step S22;

[0075] Figure 4 for Figure 2 Flowchart of the specific method for step S23;

[0076] Figure 5 for Figure 1 Flowchart of the specific method for step S30;

[0077] Figure 6 for Figure 5 Flowchart of the specific method for step S32;

[0078] Figure 7 for Figure 5 Flowchart of the specific method for step S33;

[0079] Figure 8 for Figure 5 Flowchart of the specific method for step S34;

[0080] Figure 9 for Figure 1 Flowchart of the specific method for step S40;

[0081] Figure 10 This is a structural block diagram of the in-vehicle health cloud system of the present invention;

[0082] Figure 11This is an internal layout diagram of a flexible fiber optic grating curvature sensor array and a distributed capacitive micro-motion sensor array for an automotive seat.

[0083] Figure 12 This is a layout diagram of the inertial measurement unit for a car seat.

[0084] In the figure: 1. Flexible fiber optic grating curvature sensor array; 2. Distributed capacitive micro-motion sensor array; 3. Inertial measurement unit. Detailed Implementation

[0085] To make the technical problems solved, technical solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0086] While existing smart car seats have introduced posture monitoring methods to determine the type of occupant's sitting posture or identify fatigue, they are still limited to the macroscopic posture recognition level, such as leaning forward, leaning back, and leaning to the side, and have not achieved quantitative detection of the physiological curvature of the human spine. Moreover, existing fatigue monitoring is mostly based on single-dimensional monitoring of driving behavior analysis or physiological characteristic information. During long-term driving, given the diversity of occupant sitting postures, it is difficult to form an effective means of predicting the cumulative musculoskeletal damage caused by the combination of long-term driving and poor sitting posture, and it is difficult to intervene before spinal pain occurs.

[0087] To address the aforementioned technical problems, this invention proposes a solution: real-time acquisition of spinal curvature data and vital signs of occupants while seated, as well as vehicle acceleration during vehicle movement; demodulation and curvature reconstruction of the real-time acquired spinal curvature data to obtain the occupant's three-dimensional spinal curvature curve; based on the occupant's three-dimensional spinal curvature curve and the real-time acquired vehicle acceleration, a vibration-attitude coupled fatigue algorithm is used to determine the risk level of spinal injury to the occupant during vehicle movement; finally, based on the risk level of spinal injury to the occupant, an intervention strategy corresponding to that risk level is triggered.

[0088] The above scheme constructs a three-dimensional spinal curvature curve based on the occupant's spinal curvature data to obtain the actual state of the occupant's spine in the sagittal and coronal planes, accurately identifying the occupant's sitting posture. Combined with the vehicle's acceleration during vehicle movement, a vibration-posture coupled fatigue algorithm is used to calculate the cumulative fatigue caused by the superposition of vehicle vibration and poor sitting posture in real time. This effectively identifies the risk of chronic injury caused by the long-term superposition of vehicle vibration and poor sitting posture, achieving accurate predictive risk assessment before pain occurs.

[0089] Based on the above, please refer to Figure 1As shown in the embodiments of this specification, a method for monitoring spinal curvature of a smart chair is proposed, including steps S10-S40, wherein:

[0090] S10 collects real-time data on the spinal curvature of occupants while seated, their vital signs, and the vehicle's acceleration during operation.

[0091] The occupants include drivers and non-drivers, and the spinal curvature data includes curvature change data in the sagittal plane (anterior-posterior curvature) and curvature change data in the coronal plane (lateral curvature).

[0092] In this embodiment, when the occupant is sitting in the seat, the flexible fiber optic grating sensor array inside the seat back collects the occupant's spinal curvature data in real time, and the inertial measurement unit fixed on the seat frame collects the vehicle acceleration during the vehicle's movement in real time. This embodiment provides a comprehensive and reliable data foundation for subsequent comprehensive judgment of the degree of damage to the occupant's spine caused by the superposition of road impact and poor sitting posture during vehicle movement, through spinal curvature data and vehicle acceleration.

[0093] It should be noted that, please refer to Figure 11 and Figure 12 As shown, an exemplary arrangement of a flexible fiber Bragg grating sensor array and an inertial measurement unit (IMU) is provided. To achieve bidirectional monitoring of the sagittal and coronal planes of the spine, this embodiment employs two polymer-encapsulated fiber Bragg gratings symmetrically distributed along the center line of the seat back (corresponding to the position of the human spine) in a continuous S-shaped pattern. Each flexible fiber Bragg grating has multiple FBG grating points to form a monitoring range covering from the sacrum to the cervical spine. Each FBG grating point refers to a tiny sensing area with specific wavelength-selective reflection function formed on the fiber by a fiber Bragg grating. The IMU, in addition to collecting vehicle acceleration, can also collect dynamic information such as vehicle angular velocity. Using an IMU to measure dynamic information such as acceleration and angular velocity is a conventional measurement method, and this application will not elaborate further.

[0094] In addition, a distributed array of capacitive micro-motion sensors is deployed beneath the surface layer of the seat to form a highly sensitive capacitive sensing network layer, which is used to sense occupant vital signs such as changes in surface pressure, subtle muscle tremors, and trends in center of gravity shift.

[0095] S20 demodulates and reconstructs the real-time acquired spinal curvature data to obtain the three-dimensional curvature curve of the occupant's spine.

[0096] Among them, the three-dimensional curvature curve of the spine refers to a continuous curve model that can accurately represent the actual morphological characteristics of the occupant's spine in three-dimensional space through sensor measurement and mathematical reconstruction technology, and is used to quantitatively assess whether there are abnormalities in the occupant's spinal posture.

[0097] In this embodiment, by demodulating the real-time acquired spinal curvature data through filtering, noise reduction, and signal separation, interference from environmental factors such as the inherent noise of the flexible fiber Bragg grating sensor and slight body swaying is eliminated. Effective feature signals directly related to spinal morphology are accurately extracted, generating a three-dimensional spinal curvature curve that can truly reflect the real-time spinal morphology of the occupant. Compared with existing solutions that only detect the two-dimensional state of the spine in a single direction, the three-dimensional spinal curvature curve can comprehensively cover the sagittal and coronal planes of the spine, thereby enabling more accurate identification of the occupant's sitting posture and providing a reliable data foundation for subsequent spinal injury assessment.

[0098] S30 uses the three-dimensional curvature curve of the occupant's spine and the real-time collected vehicle acceleration to determine the risk of spinal injury to the occupant during vehicle operation based on the vibration-attitude coupled fatigue algorithm.

[0099] Among them, the vibration-posture coupled fatigue algorithm is a comprehensive evaluation algorithm that quantitatively analyzes the cumulative fatigue damage to spinal tissues (muscles, intervertebral discs, etc.) caused by the synergistic effect of vibration load and dynamic posture of the spine.

[0100] Existing solutions typically treat poor posture and vehicle vibration as independent risk factors to analyze the degree of risk of spinal damage separately, while ignoring the fact that the combined effect of poor posture and vehicle vibration is more harmful to the spine. For example, when the spine is in a flexed (forward) state, its ability to withstand vertical vibration will decrease significantly, and even small bumps will be converted into huge shear forces on the intervertebral discs.

[0101] This embodiment uses a vibration-posture coupled fatigue algorithm to calculate the cumulative fatigue caused by the superposition of vehicle vibration and poor sitting posture in real time. This avoids problems such as one-sided spinal health assessment and deviation from the actual spinal injury state caused by single-factor analysis. It effectively identifies the risk of chronic injury caused by the long-term superposition of vehicle vibration and poor sitting posture, and achieves accurate predictive risk assessment before pain occurs.

[0102] S40 triggers an intervention strategy corresponding to the risk level of spinal injury to the occupant.

[0103] The risk assessment results of occupant spinal injury are transmitted to the vehicle controller via CAN bus or vehicle Ethernet communication network. The vehicle controller will dynamically adjust the damping mode of the vehicle chassis, the multi-airbag array of the car seat or the flexible electric drive frame according to the risk assessment results to intervene in the stress state of the occupant's spine, reduce the rate of fatigue injury accumulation, and remind the occupant of the current spinal injury risk through voice prompts, pop-up prompts on the vehicle display screen and other warning methods.

[0104] It is worth noting that the damping modes of the aforementioned vehicle chassis refer to the damping modes of continuously variable damping (CDC), which are generally divided into comfort damping mode, standard damping mode, and sport damping mode. The comfort damping mode has a smaller damping force, which can significantly filter out minor road bumps and reduce the transmission of low-frequency vibrations to the occupant's spine. The standard damping mode has a moderate damping force, balancing driving stability and vibration filtering effect. The sport damping mode has a larger damping force, which can reduce the inertial load on the occupant's spine caused by large changes in vehicle posture during rapid acceleration, sudden braking, or sharp turns.

[0105] In addition, the aforementioned multi-airbag array is used to adjust the curvature of the lumbar support and the height of the support protrusion, while the flexible electric drive frame is used to adjust the curvature of the cervical and thoracic spine support areas. Based on the risk level assessment results, the system dynamically adjusts the different spinal support areas of the car seat to form protrusions or depressions, thus intervening and adjusting the stress state of the occupant's spine in advance and avoiding long-term fatigue accumulation that could damage the occupant's spine.

[0106] Optional, please refer to Figure 2 As shown, in step S20, the real-time acquired spinal curvature data is demodulated and reconstructed to obtain the three-dimensional curvature curve of the occupant's spine. This also includes steps S21-S24, where:

[0107] S21, the real-time acquired spinal curvature data is demodulated by a demodulator to calculate the physical strain value of each discrete monitoring point;

[0108] S22, based on the physical strain values ​​of each discrete monitoring point, the curvature data of the spine support area of ​​the seat is derived using the Euler-Bernoulli beam theory;

[0109] S23. Based on the curvature data of the spinal support area of ​​the seat, perform spatial curve integration and fitting to obtain a continuous and differentiable spatial curve.

[0110] S24. Based on the occupant's vital signs, the normal vector of the continuous differentiable spatial curve is shifted to obtain the occupant's unique three-dimensional spinal curvature curve.

[0111] The demodulator, specifically designed for fiber optic Bragg grating (FBG) sensing systems, is a photoelectric signal processing instrument whose core function is to precisely convert the wavelength shift of the reflected FBG light signal into the corresponding physical strain. In this embodiment, when an occupant's back rests against the seat back, the physiological curvature of the spine forces the seat back cushion to deform, causing the integrated FBG grating dots to slightly bend, resulting in axial strain. This causes a linear wavelength shift. The demodulator will collect the linear wavelength shift of each FBG grating point at a frequency of 1000Hz, thereby calculating the physical strain value of each FBG grating point.

[0112] It is worth noting that the aforementioned discrete monitoring points refer to multiple FBG grating points deployed within the flexible fiber optic grating.

[0113] Optional, please refer to Figure 3 As shown, in step S22, based on the physical strain values ​​of each discrete monitoring point, the curvature data of the seat spinal support region is derived according to the Euler-Bernoulli beam theory. This also includes steps S221-S222, where:

[0114] S221, calculate the curvature of each discrete monitoring point. The specific calculation formula is as follows:

[0115] ;

[0116] In the formula, Let be the curvature of the i-th discrete monitoring point. Let h be the strain value at the i-th discrete monitoring point, and h be the vertical distance between the fiber core and the bending neutral layer.

[0117] S222, based on the curvature of each discrete monitoring point, the tangent vector and normal vector of each discrete monitoring point in three-dimensional space are recursively calculated using the Freyner-Serrey formula. The specific calculation formula is as follows:

[0118] ;

[0119] In the formula, Let be the principal normal vector of the i-th discrete monitoring point. Let be the tangent vector of the i-th discrete monitoring point. Let be the binormal vector of the i-th discrete monitoring point. Let be the torque of the i-th discrete monitoring point.

[0120] Euler-Bernoulli beam theory is a classic theory in solid mechanics used to analyze the bending deformation and mechanical response of slender beams under transverse loads. In this embodiment, the spinal support area of ​​the seat is equivalent to an elastic beam structure, and the deformation of the occupant's spine on the seat is equivalent to the deformation of the elastic beam structure. This is combined with the core formula of Euler-Bernoulli beam theory. This allows us to derive the curvature of each discrete monitoring point. The Fryner-Serrey formula is a set of differential equations in differential geometry that describes the geometric properties of a smooth space curve. It is used to establish the differential relationship between the unit tangent vector, normal vector, and binormal vector of a smooth space curve at a certain point and the curve curvature and torsion.

[0121] Specifically, the physical strain values ​​of each discrete monitoring point are substituted into... By combining the vertical distance between the fiber core and the bending neutral layer, which is determined by the encapsulation structure of the flexible fiber grating, the curvature of each discrete monitoring point can be obtained. When recursively calculating the tangent and normal vectors of each discrete monitoring point, the bottom of the seat (i.e., the sacrum position) is first set as the origin (0, 0, 0), and the initial tangent vector direction is vertically upward. Based on the curvature of each discrete monitoring point... and torque By substituting into the Flyner-Serrey formula, the tangent vector of each discrete monitoring point in three-dimensional space can be calculated recursively. and normal vector ( , ).

[0122] Optional, please refer to Figure 4 As shown, in step S23, based on the curvature data of the spine support area of ​​the seat, spatial curve integration and fitting are performed to obtain a continuously differentiable spatial curve. This also includes steps S231-S233, where:

[0123] S231, based on curvature data, calculate the change in shear angle in each spinal support area of ​​the seat. The specific calculation formula is as follows:

[0124] ;

[0125] In the formula, This represents the change in pitch angle and tangent angle. The rate of change of yaw angle and tangent angle. Let S be the curvature of the spinal support area S in the seat;

[0126] S232, the tangent of each spinal support region is transformed by coordinate integration with the two-dimensional coordinates of each discrete monitoring point within the corresponding region to obtain the three-dimensional coordinates of each discrete monitoring point in the three-dimensional coordinate system. The specific integration transformation process is as follows:

[0127] ;

[0128] In the formula, The pitch angle corresponds to the sagittal curvature of the spine. This is the yaw angle, corresponding to the coronal scoliosis of the spine;

[0129] S233 uses cubic spline interpolation to smoothly fit the three-dimensional coordinates of each discrete monitoring point, generating a continuous and differentiable spatial curve.

[0130] The pitch angle refers to the rotation angle of the upper body (or torso) relative to the seat reference coordinate system (the midpoint of the front end of the seat cushion as the origin, the longitudinal axis as the X-axis, the transverse axis as the Y-axis, and the vertical axis as the Z-axis) in the sagittal plane around the y-axis. When the upper body is tilted forward relative to an upright sitting posture, the pitch angle is positive; when the upper body is tilted backward, the pitch angle is negative. The yaw angle refers to the rotation angle of the upper body (or torso) relative to the seat reference coordinate system in the coronal plane around the x-axis. It is used to characterize the degree of bending and offset of the spine in the left and right directions. Generally, when the upper body shifts to the left, the yaw angle is defined as negative; when it shifts to the right, the yaw angle is defined as positive. The spinal support area is divided into the cervical spine support area, the thoracic spine support area, and the lumbar spine support area. The spinal support area S represents one of the three support areas: cervical spine support area, thoracic spine support area, and lumbar spine support area.

[0131] Specifically, based on the curvature data of each discrete monitoring point within the spinal support region, and combined with the arc length parameter of the spinal support region (which is determined by the spacing between each discrete monitoring point within the spinal support region), the pitch angle and yaw angle changes of each spinal support region are calculated. These changes directly reflect the bending angle shift trend of the spine morphology in each spinal support region, providing a precise attitude constraint basis for the subsequent three-dimensional coordinate transformation of the discrete monitoring points. Using the pitch angle and yaw angle changes of each spinal support region as constraints, and combining them with the two-dimensional coordinates of each discrete monitoring point through the aforementioned coordinate integral transformation formula, the three-dimensional coordinates of each discrete monitoring point in the three-dimensional spatial coordinate system can be obtained.

[0132] It is worth noting that, because the FBG grating points in the flexible fiber Bragg grating sensor array are arranged in an array (e.g., row i × column j), each FBG grating point (i.e., discrete monitoring point) has unique two-dimensional coordinates. .

[0133] In this embodiment, the change in the shear angle of the spinal support area directly reflects the trend of the bending angle deviation of the spinal morphology in each spinal support area. It is used as an attitude constraint to avoid positional deviation when mapping the two-dimensional coordinates of each discrete monitoring point to the three-dimensional coordinate system. Since discrete monitoring points have problems such as data discreteness, discontinuity, and missing data, cubic spline interpolation is used to smoothly fit the three-dimensional coordinates of each discrete monitoring point to generate a continuous and differentiable spatial curve, providing a reliable data basis for subsequent assessment of occupant spinal injury.

[0134] In a preferred implementation, in step S24 above, during the process of correcting the normal vector offset of the continuous differentiable spatial curve based on the occupant's vital signs information to obtain the occupant's unique three-dimensional spinal curvature curve, the surface pressure changes of the occupant are collected in real time by a distributed capacitive micro-motion sensor array to estimate the thickness of clothing and subcutaneous fat layer, and the normal vector offset of the continuous differentiable spatial curve is corrected based on the estimation results to eliminate the curve curvature error caused by the thick coat.

[0135] Specifically, in the process of estimating clothing thickness, the equivalent thickness of the clothing is estimated based on the correspondence between the occupant's surface pressure and contact deformation characteristics. Under the same or similar pressure conditions, the thicker the clothing layer, the more significant its buffering effect on pressure, manifested as lower local pressure values ​​detected by the distributed capacitive micro-motion sensor array and a slower rate of pressure change. During estimation, real-time pressure changes are detected by multiple distributed capacitive micro-motion sensors within the spinal support area. Combined with a pre-established empirical model of clothing compression, the estimated equivalent thickness of the clothing under the occupant's current wearing condition can be obtained.

[0136] In estimating subcutaneous fat layer thickness, occupants with thicker subcutaneous fat layers typically exhibit a larger contact area, lower peak pressure, and a gentler pressure gradient when in contact with the seat back. Conversely, occupants with thinner subcutaneous fat layers show concentrated pressure and higher peak pressure. The estimation process involves collecting characteristic parameters such as the peak pressure, effective contact area, and pressure change trend of the area where the occupant's back contacts the seat back. By combining these parameters with a feature information database, the equivalent thickness of the occupant's subcutaneous fat layer can be estimated.

[0137] The estimated equivalent thickness of clothing and the estimated equivalent thickness of subcutaneous fat layer are superimposed to obtain the total equivalent thickness value for depth compensation. Then, based on the principal normal vector calculated from each discrete monitoring point in the obtained continuous differentiable space curve, the continuous differentiable space curve is offset and corrected along the direction of the principal normal vector. The offset amount is the total equivalent thickness value for depth compensation, so that the corrected space curve returns from the position of the outer surface of clothing to a position that is closer to the actual surface contour of the occupant's spine.

[0138] In this embodiment, even when the occupant is wearing a thick coat or multiple layers of clothing, the curvature reduction caused by clothing cushioning and soft tissue deformation can be effectively corrected, thereby eliminating the spinal curvature error caused by such factors and obtaining a more realistic, stable, and individualized three-dimensional spinal curvature curve.

[0139] Optional, please refer to Figure 5As shown, in step S30, based on the occupant's three-dimensional spinal curvature curve and the real-time acquired vehicle acceleration, and using a vibration-attitude coupled fatigue algorithm, the risk level of spinal injury to the occupant during vehicle operation is determined. This also includes steps S31-S34, where:

[0140] S31, the median filtering method is used to remove the impulse noise caused by instantaneous body shaking on the three-dimensional curvature curve of the occupant's spine, resulting in a smooth three-dimensional curvature curve of the spine, and a low-pass filter is used to remove the high-frequency mechanical noise of the vehicle acceleration.

[0141] S32 performs frequency-weighted processing on the vehicle acceleration after noise filtering to obtain frequency-weighted acceleration;

[0142] S33, based on the three-dimensional curvature curve of the occupant's spine and the frequency-weighted acceleration after noise filtering, calculate the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment.

[0143] S34, based on the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment, determines the risk level of spinal injury to the occupant during vehicle operation.

[0144] In this embodiment, to eliminate the interference of environmental noise on the calculation accuracy, the median filtering method is used to remove the impulse noise caused by the instantaneous shaking / swaying of the body on the generated three-dimensional curvature curve of the spine. At the same time, a low-pass filter is used to remove the high-frequency mechanical noise of the vehicle acceleration to retain the low-frequency vibration component (usually <20Hz) that affects the human skeleton. Since the human spine is most sensitive to vibrations (resonance effect) at specific frequencies, the vehicle acceleration is converted into frequency-weighted acceleration that is close to the actual human perception. This avoids directly using vehicle acceleration to assess the risk of spinal injury, which ignores the physiological response characteristics of the human body and results that do not match the actual spinal injury situation.

[0145] Optional, please refer to Figure 6 As shown, in step S32, the vehicle acceleration after noise filtering is subjected to frequency-weighted processing to obtain frequency-weighted acceleration. The process also includes steps S321-S323, wherein:

[0146] S321 converts the time-domain data of vehicle acceleration after noise filtering into frequency-domain data through short-time Fourier transform.

[0147] S322, based on the ISO 2631 standard for evaluating whole-body vibration, generates a weighting function, and the weighting function takes the maximum value when the vibration frequency is between 4Hz and 8Hz (vertical resonance zone);

[0148] S323, based on the frequency domain data of vehicle acceleration and the weighting function, a weighted calculation is performed to obtain the frequency-weighted acceleration. The specific calculation process is as follows:

[0149] ;

[0150] In the formula, For weighted functions, This is the frequency domain data for vehicle acceleration.

[0151] Among them, the weighting function for ;

[0152] The Short-Time Fourier Transform (SFT) is a time-domain analysis method used to analyze the local spectral characteristics of non-stationary signals (signals whose frequency changes with time). Its core principle is to use a sliding window function to extract the time-domain signal and perform a Fourier transform on the time-domain signal within each window. This achieves time-domain localized analysis of non-stationary signals (such as acceleration signals that change with road conditions during vehicle movement), obtaining frequency-domain data showing the frequency variation over time. The SFT formula is as follows:

[0153] ;

[0154] In the formula, For the time-domain data of vehicle acceleration after noise filtering, For window functions (Hanning window or Gaussian window). The time variable represents the time center of the time-domain spectrum. For frequency variables, it represents the frequency components at that moment.

[0155] The aforementioned human whole-body vibration evaluation standard is based on the ISO 2631 standard, which is an authoritative medical standard published by the International Organization for Standardization (ISO) for assessing the impact of vibration environments such as vehicles and ships on human health.

[0156] In this embodiment, the time-domain data of vehicle acceleration after noise filtering is subjected to a short-time Fourier transform to obtain the frequency-domain data of vehicle acceleration. Then, according to the ISO 2631 standard and combined with the characteristic that the vehicle vibration meter is mainly vertical, a weighting function in the vertical direction is generated. The weighting function takes the maximum value when the vibration frequency is between 4Hz and 8Hz (vertical resonance zone). In other frequency ranges, the value is taken according to a preset piecewise function to ensure that the weighting logic is accurately matched with the physiological response characteristics of human vibration. Then, the frequency-domain data of vehicle acceleration and the weighting function are substituted into the above weighting calculation formula to obtain the frequency-weighted acceleration. This allows the vehicle acceleration feedback to characterize the effective biomechanical vibration intensity, which is more in line with the actual influencing factors of spinal injury.

[0157] Optional, please refer to Figure 7 As shown, in step S33, the calculation of the instantaneous comprehensive pressure on the occupant's spine and the cumulative fatigue index of the occupant's spine at the current moment, based on the three-dimensional curvature curve of the occupant's spine and the frequency-weighted acceleration after noise filtering, also includes steps S331-S333, wherein:

[0158] S331, based on the three-dimensional curvature curve of the occupant's spine after noise filtering, the angle deviation from the ideal spinal reference angle is calculated to obtain the angle attenuation coefficient. The specific calculation process is as follows:

[0159] ;

[0160] In the formula, The angular attenuation coefficient, This is the attitude sensitivity coefficient. It is a non-linear amplification index. This is the angular deviation value;

[0161] Among them, the angle deviation value for ;

[0162] In the formula, This includes parameters representing the real-time curvature changes of each spinal segment within the three-dimensional curvature of the spine. The ideal spinal reference angle;

[0163] S332, based on the frequency-weighted acceleration and the angle attenuation coefficient, calculate the instantaneous comprehensive pressure on the occupant's spine at the current moment. The specific calculation process is as follows:

[0164] ;

[0165] In the formula, For instantaneous comprehensive pressure, Static load weight, Dynamic shearing weights;

[0166] S333, based on the instantaneous comprehensive pressure on the occupant's spine at the current moment, combined with biological metabolic recovery factors, calculates the cumulative fatigue index of the occupant's spine at the current moment. The specific calculation process is as follows:

[0167] ;

[0168] In the formula, The cumulative spinal fatigue index of the occupants at the current moment. For attenuation term, This represents the muscle recovery time constant.

[0169] The angle attenuation coefficient is a dynamic weighting factor used to quantify the amplification or attenuation of the vibration damage effect caused by poor spinal posture. Its physical meaning is that the greater the deviation of the actual curvature of the spine from the ideal physiological curvature, the more nonlinearly the effective damage intensity of the vibration load on the spine increases. The bio-metabolic recovery factor in this embodiment is the muscle recovery time constant, a parameter used to characterize the fatigue recovery rate of human spinal muscle tissue after load application.

[0170] In this embodiment, the influence of static posture deviation on spinal stress is quantified by the angle attenuation coefficient. Combined with the frequency-weighted acceleration of the vehicle, the instantaneous comprehensive pressure on the occupant's spine at the current moment is calculated, which can accurately reflect the actual stress on the spine under the superposition of the vibration intensity of the vehicle and the occupant's sitting posture. Furthermore, considering the natural recovery ability of human spinal muscle tissue, the muscle recovery time constant is introduced to correct the assessment bias caused by neglecting the recovery effect in the fatigue accumulation model, making the obtained spinal cumulative fatigue index more consistent with human physiological characteristics and making the risk level assessment result of occupant spinal injury more accurate.

[0171] Optional, please refer to Figure 8 As shown, in step S34, the risk level of spinal injury to the occupant is determined based on the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment. This also includes steps S341-S344, where:

[0172] S341, by dividing the fatigue index into preset thresholds, the continuously changing cumulative spinal fatigue index of the occupant at the current moment is mapped into discrete risk levels, which are divided into extremely high risk level, medium risk level and safe level.

[0173] S342, If the instantaneous comprehensive pressure on the occupant's spine at the current moment is greater than or equal to the instantaneous danger threshold, the risk level of injury to the occupant's spine at this moment is determined to be an extremely high risk level.

[0174] S343, if the instantaneous comprehensive pressure on the occupant's spine at the current moment is less than or equal to the instantaneous danger threshold, but greater than or equal to the instantaneous warning threshold, the risk level of injury to the occupant's spine at this moment is determined to be a moderate risk level.

[0175] S344, if the instantaneous comprehensive pressure on the occupant's spine at the current moment is less than the instantaneous warning threshold, the risk level of injury to the occupant's spine at this moment is determined to be a safe level.

[0176] Specifically, the aforementioned preset fatigue index threshold has two thresholds to classify discrete risk levels into extremely high risk, medium risk, and safe levels. The two fatigue index thresholds are the high-risk fatigue index threshold and the medium-risk fatigue index threshold, respectively. The specific values ​​of the two fatigue index thresholds can be determined based on human biomechanical experiments and clinical data calibration.

[0177] It is worth noting that the aforementioned instantaneous danger threshold and instantaneous warning threshold can be obtained through experiments simulating long-term vehicle driving and encountering different road conditions.

[0178] In this embodiment, the cumulative spinal fatigue index of the occupant reflects the chronic spinal injury caused by long-term load superposition. It is used to identify the hidden fatigue risk under the long-term effects of low-intensity vibration and poor sitting posture. Combined with the instantaneous comprehensive pressure on the occupant's spine, it judges the risk level of spinal injury. This is to capture the sudden risks during rapid acceleration, sudden braking, and severe road bumps. By combining the two, the limitations of single-dimensional assessment are overcome, avoiding the difficulty in identifying long-term chronic injury risks, and also not ignoring the threat of spinal injury caused by sudden risks. This makes the risk assessment results of spinal injury more in line with the actual situation.

[0179] Optional, please refer to Figure 9 As shown, in step S40, triggering an intervention strategy corresponding to the risk level of spinal injury to the occupant further includes steps S41-S42, wherein:

[0180] S41, when the risk level of spinal injury to the occupant is extremely high, the vehicle will be subjected to vibration reduction intervention and the support force of the spinal support area of ​​the seat will be increased.

[0181] S42: When the risk level of spinal injury to the occupant is at the moderate risk level, an early warning will be issued and the support force in the spinal support area of ​​the seat will be increased.

[0182] In this embodiment, the vibration reduction intervention includes vehicle chassis intervention and seat vibration reduction intervention. The vehicle chassis intervention refers to the vehicle controller actively switching the continuously variable damping mode to the comfort damping mode when the risk of spinal injury is determined to be extremely high, thereby filtering road impact energy to the greatest extent by softening the damping. The seat vibration reduction intervention refers to adjusting the current of the MR damper at the bottom of the seat and adjusting the damping coefficient of the seat when the risk of spinal injury is determined to be extremely high, thereby performing reverse phase compensation for specific harmful frequencies (such as the 4-8Hz resonant frequency).

[0183] Please refer to Figure 10As shown in the embodiments of this specification, an in-vehicle health cloud system 100 is also proposed. This in-vehicle health cloud system 100 is used to implement the aforementioned spinal curvature monitoring method, including...

[0184] The acquisition module 110 includes a flexible fiber optic grating sensor array disposed inside the seat back for real-time acquisition of spinal curvature data of the occupant sitting on the seat. The acquisition module 110 also includes an inertial measurement unit fixed to the seat frame for real-time acquisition of vehicle acceleration during vehicle movement.

[0185] The vehicle-mounted health brain module 120 includes a modeling unit, a spinal injury judgment unit, and an execution unit.

[0186] The modeling unit is used to demodulate and reconstruct the curvature of the spine collected in real time to obtain the three-dimensional curvature curve of the occupant's spine; the spinal injury judgment unit is used to determine the risk level of spinal injury of the occupant during vehicle operation based on the three-dimensional curvature curve of the occupant's spine and the real-time collected vehicle acceleration, using a vibration-attitude coupled fatigue algorithm; the execution unit is used to trigger an intervention strategy corresponding to the risk level of spinal injury of the occupant.

[0187] In this embodiment, a three-dimensional curvature curve of the spine is constructed based on the occupant's spinal curvature data to obtain the actual state of the occupant's spine in the sagittal and coronal planes, accurately identifying the occupant's sitting posture. Combined with the vehicle acceleration during vehicle movement, the cumulative fatigue caused by the superposition of vehicle vibration and poor sitting posture is calculated in real time through a vibration-posture coupled fatigue algorithm. This effectively identifies the risk of chronic injury caused by the long-term superposition of vehicle vibration and poor sitting posture, achieving accurate predictive risk assessment before pain occurs.

[0188] In a preferred embodiment, the in-vehicle health cloud system 100 further includes an in-vehicle network communication module 130 and a self-learning module 140. The in-vehicle network communication module 130 is communicatively connected to the in-vehicle health brain module 120. The in-vehicle network communication module 130 is used to upload information from the in-vehicle health brain module 120, such as the three-dimensional curvature curve of the spine, the instantaneous comprehensive pressure on the occupant's spine at the current moment, the cumulative fatigue index of the occupant's spine at the current moment, intervention strategies, and real-time collected occupant data and vehicle data, to the cloud-based spinal health digital platform. The self-learning module 140 is used to perform self-learning based on the three-dimensional curvature curve of the spine, the instantaneous comprehensive pressure on the occupant's spine at the current moment, and the cumulative fatigue index of the occupant's spine at the current moment, to iteratively update the above thresholds and intervention strategies, thereby realizing the construction of a personalized adaptive monitoring system for different occupants, making the seat spinal health monitoring and assessment more consistent with the actual state of the occupant's spine.

[0189] It is worth noting that the aforementioned cloud-based spinal health digital platform is a remote spinal health monitoring and management system built on technologies such as cloud computing, big data analysis, and human biomechanics. This cloud-based spinal health digital platform also supports remote rehabilitation experts to view the real-time and historical spinal data of authorized users through video terminals when authorized by the users, and to provide the authorized users with professional assessments, guidance, rehabilitation suggestions, or intervention plans.

[0190] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0191] Those skilled in the art should understand that the above description is one embodiment provided in conjunction with specific content, and does not imply that the specific implementation of the present invention is limited to these descriptions. Furthermore, due to differences in industry naming conventions, the invention is not limited to the above names or English names. Any methods or structures similar to or identical to those of the present invention, or any technical deductions or substitutions made based on the concept of the present invention, should be considered within the scope of protection of the present invention.

Claims

1. A method for monitoring spinal curvature in a smart chair, characterized in that, include: Real-time data collection of spinal curvature and vital signs of occupants while seated, as well as vehicle acceleration during vehicle movement; The real-time acquired spinal curvature data is demodulated and reconstructed to obtain the three-dimensional curvature curve of the occupant's spine. The median filtering method was used to remove the impulse noise caused by instantaneous body shaking from the three-dimensional curvature curve of the occupant's spine, resulting in a smooth three-dimensional curvature curve of the spine. A low-pass filter was used to remove the high-frequency mechanical noise of the vehicle from the vehicle acceleration. The vehicle acceleration after noise filtering is frequency-weighted to obtain frequency-weighted acceleration; Based on the three-dimensional curvature curve of the occupant's spine after noise filtering, the angle deviation from the ideal spinal reference angle is calculated to obtain the angle attenuation coefficient. The specific calculation process is as follows: ； In the formula, The angular attenuation coefficient, This is the attitude sensitivity coefficient. It is a non-linear amplification index. This is the angular deviation value; Among them, the angle deviation value for ; In the formula, The time series of each spinal segment containing the three-dimensional curvature of the spine; The ideal spinal reference angle; Based on the frequency-weighted acceleration and the angle attenuation coefficient, the instantaneous comprehensive pressure on the occupant's spine at the current moment is calculated. The specific calculation process is as follows: ； In the formula, For instantaneous comprehensive pressure, Static load weight, Dynamic shearing weights; Based on the instantaneous comprehensive pressure on the occupant's spine at the current moment, and combined with biological metabolic recovery factors, the cumulative fatigue index of the occupant's spine at the current moment is calculated. The specific calculation process is as follows: ； In the formula, The cumulative spinal fatigue index of the occupants at the current moment. For attenuation term, This represents the muscle recovery time constant. Based on the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment, the risk level of spinal injury to the occupant during vehicle operation is determined. Based on the risk level of spinal injury to the occupant, an intervention strategy corresponding to that risk level is triggered.

2. The method for monitoring spinal curvature of an intelligent chair according to claim 1, characterized in that, The process of demodulating and reconstructing the real-time acquired spinal curvature data to obtain the occupant's three-dimensional spinal curvature curve includes: The real-time acquired spinal curvature data is demodulated using a demodulator to calculate the physical strain value of each discrete monitoring point. Based on the physical strain values ​​at each discrete monitoring point, the curvature data of the spine support region of the seat is derived using the Euler-Bernoulli beam theory. Based on the curvature data of the spine support area of ​​the seat, spatial curve integration and fitting are performed to obtain a continuous and differentiable spatial curve. Based on the occupant's vital signs, the normal vector of the continuous differentiable spatial curve is offset to obtain the occupant's unique three-dimensional spinal curvature curve.

3. The method for monitoring spinal curvature of an intelligent chair according to claim 2, characterized in that, The curvature data of the seat spinal support region is derived based on the physical strain values ​​of each discrete monitoring point and the Euler-Bernoulli beam theory, including: The curvature of each discrete monitoring point is calculated using the following formula: ； In the formula, Let be the curvature of the i-th discrete monitoring point. Let h be the strain value at the i-th discrete monitoring point, and h be the vertical distance between the fiber core and the bending neutral layer. Based on the curvature of each discrete monitoring point, the tangent vector and normal vector of each discrete monitoring point in three-dimensional space are recursively calculated using the Flyner-Serrey formula. The specific calculation formula is as follows: ； In the formula, Let be the principal normal vector of the i-th discrete monitoring point. Let be the tangent vector of the i-th discrete monitoring point. Let be the binormal vector of the i-th discrete monitoring point. Let be the torque of the i-th discrete monitoring point.

4. The method for monitoring spinal curvature of an intelligent chair according to claim 2, characterized in that, The process of integrating and fitting spatial curves based on the curvature data of the spinal support area of ​​the seat to obtain a continuously differentiable spatial curve includes: Based on the curvature data, the change in tangential angle of each spinal support area of ​​the seat is calculated. The specific calculation formula is as follows: ； In the formula, This represents the change in pitch angle and tangent angle. The rate of change of yaw angle and tangent angle. Let S be the curvature of the spinal support area S in the seat; The tangent of each spinal support region is transformed by coordinate integration with the two-dimensional coordinates of each discrete monitoring point within the corresponding region to obtain the three-dimensional coordinates of each discrete monitoring point in the three-dimensional coordinate system. The specific integration transformation process is as follows: ； In the formula, The pitch angle corresponds to the sagittal curvature of the spine. This is the yaw angle, corresponding to the coronal scoliosis of the spine; By using cubic spline interpolation, the three-dimensional coordinates of each discrete monitoring point are smoothly fitted to generate a continuous and differentiable spatial curve.

5. The method for monitoring spinal curvature of an intelligent chair according to claim 1, characterized in that, The step of performing frequency-weighted processing on the vehicle acceleration after noise filtering to obtain frequency-weighted acceleration includes: The time-domain data of vehicle acceleration after noise filtering is converted into frequency-domain data by short-time Fourier transform. Based on the ISO 2631 standard for evaluating whole-body vibration, a weighting function is generated, and the maximum value of the weighting function is taken when the vibration frequency is between 4Hz and 8Hz. The frequency-weighted acceleration is obtained by weighting the vehicle's acceleration based on the frequency domain data and a weighting function. The specific calculation process is as follows: ； In the formula, For weighted functions, Frequency domain data for vehicle acceleration.

6. The method for monitoring spinal curvature of an intelligent chair according to claim 1, characterized in that, The method of determining the risk level of spinal injury to the occupant during vehicle operation based on the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment includes: By using a preset fatigue index threshold, the continuously changing cumulative spinal fatigue index of the occupant at the current moment is mapped to a discrete risk level, which is divided into an extremely high risk level, a moderate risk level, and a safe level. If the instantaneous combined pressure on the occupant's spine at the current moment is greater than or equal to the instantaneous danger threshold, the risk level of injury to the occupant's spine at this moment is determined to be extremely high risk. If the instantaneous comprehensive pressure on the occupant's spine at the current moment is less than or equal to the instantaneous danger threshold, but greater than or equal to the instantaneous warning threshold, the risk level of spinal injury to the occupant at this moment is determined to be moderate risk level. If the instantaneous comprehensive pressure on the occupant's spine at the current moment is less than the instantaneous warning threshold, the risk level of spinal injury to the occupant at this moment is determined to be a safe level.

7. The method for monitoring spinal curvature of an intelligent chair according to claim 6, characterized in that, The method of triggering intervention strategies corresponding to the risk level of spinal injury of the occupant includes: When the risk level of spinal injury to the occupant is extremely high, vibration reduction intervention is performed on the vehicle and the support force is increased in the spinal support area of ​​the seat. When the risk level of spinal injury to the occupant is at the moderate risk level, an early warning will be issued, and the support force in the spinal support area of ​​the seat will be increased.

8. A vehicle-mounted health cloud system, wherein the vehicle-mounted health cloud system is used to implement the spinal curvature monitoring method as described in any one of claims 1 to 7, characterized in that, The in-vehicle health cloud system includes: The acquisition module includes a flexible fiber optic grating sensor array disposed inside the seat back for real-time acquisition of spinal curvature data of the occupant sitting on the seat. The acquisition module also includes an inertial measurement unit fixed to the seat frame for real-time acquisition of vehicle acceleration during vehicle movement. The vehicle-mounted health brain module includes a modeling unit, a spinal injury judgment unit, and an execution unit. The modeling unit is used to demodulate and reconstruct the curvature of the spine in real time to obtain the three-dimensional curvature curve of the occupant's spine. The spinal injury assessment unit is used to determine the risk level of spinal injury to the occupant during vehicle operation based on the three-dimensional curvature curve of the occupant's spine and the real-time collected vehicle acceleration, using a vibration-attitude coupled fatigue algorithm. The execution unit is used to trigger an intervention strategy corresponding to the risk level of spinal injury suffered by the occupant. The spinal injury assessment unit is also used to remove impulse noise caused by instantaneous body shaking from the occupant's three-dimensional spinal curvature curve using a median filter, resulting in a smooth three-dimensional spinal curvature curve. It also uses a low-pass filter to remove high-frequency mechanical noise from the vehicle's acceleration. The noise-filtered vehicle acceleration is then subjected to frequency-weighted processing to obtain frequency-weighted acceleration. Based on the noise-filtered three-dimensional spinal curvature curve of the occupant, the angle deviation from the ideal spinal reference angle is calculated to obtain the angle attenuation coefficient. The specific calculation process is as follows: ； In the formula, The angular attenuation coefficient, This is the attitude sensitivity coefficient. It is a non-linear amplification index. Here, the angle deviation value is... for In the formula, The time series of each spinal segment containing the three-dimensional curvature of the spine; The ideal spinal reference angle is used; based on the frequency-weighted acceleration and angle attenuation coefficient, the instantaneous comprehensive pressure on the occupant's spine at the current moment is calculated. The specific calculation process is as follows: ； In the formula, For instantaneous comprehensive pressure, Static load weight, The dynamic shear weight is used; based on the instantaneous comprehensive pressure on the occupant's spine at the current moment, combined with biological metabolic recovery factors, the cumulative fatigue index of the occupant's spine at the current moment is calculated. The specific calculation process is as follows: ； In the formula, The cumulative spinal fatigue index of the occupants at the current moment. For attenuation term, The muscle recovery time constant is used to determine the risk level of spinal injury to the occupant during vehicle operation, based on the instantaneous comprehensive pressure on the occupant's spine at the current moment and the cumulative fatigue index of the occupant's spine at the current moment.

Images