A driver spine fine posture prediction method based on a cockpit layout parameter

Abstract

The application discloses a driver spine fine posture prediction method based on a cockpit layout parameter and belongs to the technical field of automobile man-machine engineering. The method comprises the following steps: collecting human feature data and cockpit layout parameters of a driver; fitting a back curve in a driver sagittal plane and extracting a curvature feature; calculating a polar angle and a polar radius of each spine segment based on a pre-established regression formula; and finally determining the spatial position of the mass center of each vertebra according to a skin surface spinous process marker point and polar coordinate parameters, and completing the spine fine posture prediction. The application overcomes the limitation that the spine is simplified into three segments in the prior art, realizes the fine prediction of all spine segments, establishes a direct quantitative relationship between the cockpit layout parameter and the spine posture, significantly reduces the implementation cost and complexity, and is suitable for the comfort evaluation and optimization in the forward design stage of the automobile cockpit.

Description

Technical Field

[0001] This invention belongs to the field of automotive ergonomics technology, specifically relating to a method for predicting the fine posture of the driver's spine based on cockpit layout parameters. Background Technology

[0002] In automotive ergonomics, the driver's spinal posture and seat parameters are closely related, directly affecting driving comfort, spinal health, and driving safety. With the development of intelligent vehicles, cars have gradually transformed from traditional modes of transportation into "mobile third spaces" centered on user experience. This shift has led users to place higher demands on the comfort and interactivity of vehicles, driving the in-depth application of human-centered design principles in automotive seat and cabin design.

[0003] Currently, as a key platform for human-vehicle interaction, the intelligent cockpit's seat design must meet higher standards in terms of safety, comfort, and intelligence. Existing technologies for seat development mainly include two approaches:

[0004] The first category is based on subjective comfort evaluation and body pressure distribution experiments. This method collects subjective comfort data through numerous experiments and uses a body pressure distribution system for seat design and optimization. However, this method has limitations such as large experimental sample requirements, significant subjective bias, and difficulty in uniformly measuring individual differences, resulting in low efficiency and insufficient universality. These limitations make this method difficult to apply effectively to the forward design process and more suitable for evaluating and validating existing designs.

[0005] The second category is methods that rely on driver posture estimation to guide seat design. These methods, based on predicting comfortable driving postures, offer significant advantages in terms of economy and operability, and can provide a basis for parameter optimization in the early stages of design. Currently, most methods relying on driver posture estimation to guide seat design employ computer-aided design models and predictive models built based on experimental statistical methods. However, due to the adjustable height of car seats and the diverse target user groups, accurately predicting driving postures, especially fine spinal postures, remains a significant challenge. A systematic and high-precision spinal segment posture estimation system has not yet been established.

[0006] Computer-aided design (CAD) using human body models is now widely employed in cockpit design, utilizing digital human body models to predict driving posture, including spinal posture. Well-known models include RAMSIS (a widely used ergonomics design and simulation software in the automotive industry) and Transom Jack (a digital human body modeling software for ergonomics simulation and analysis). These models can roughly simulate basic physiological structures, perform preliminary simulations of basic skeletal joint functions, conduct virtual experiments and analyses in some simple scenarios, and reduce reliance on real human experiments to some extent. They are widely used in ergonomics for determining human posture and position. These models can also provide linear relationship models of spinal segments to predict spinal posture during driving.

[0007] A predictive model based on experimental statistical methods is used in the field of automotive cockpit research. A driving simulation platform is established to study the impact of pedals, steering wheel, seat, and human body parameters within the passenger compartment on driving posture through simulation experiments. Furthermore, a quantitative relationship model between the seat, steering wheel, pedals, and driving posture within the driver's cockpit space is established from a statistical perspective. Existing models can calculate based on some design parameters of the passenger compartment and some human body parameters of the occupants to predict the driving posture and human body contour of a target occupant under specific layout parameters.

[0008] However, using existing computer-aided human body models for driving posture prediction has significant limitations. There are adaptation differences between these digital human body models and real-world seating systems, making it difficult to accurately translate realistic comfortable sitting postures into model postures. A more critical limitation lies in the insufficient model detail: these models typically oversimplify the spine into three rigid segments—cervical, thoracic, and lumbar—providing only angular parameters for each segment and failing to achieve detailed predictions at the vertebral segment level. Even those models with more detailed parametric designs fail to effectively correlate with automotive cabin layout parameters, resulting in poor prediction performance in real-world vehicle applications. Therefore, current digital human body models cannot meet the needs for detailed spinal posture prediction in driving environments.

[0009] Secondly, prediction methods based on experimental statistics also have limitations. These methods primarily focus on large joints or overall trunk posture, such as the neck, shoulder, elbow, wrist, torso, hip, knee, and ankle, lacking refined modeling of multiple spinal segments. Their models are severely limited by the number and diversity of experimental samples, resulting in insufficient generalization ability and difficulty adapting to the personalized needs of different body types and sitting postures. Most importantly, existing statistical models have failed to establish a direct and precise mapping relationship between spinal posture and cabin layout parameters, limiting their prediction accuracy and practicality.

[0010] Therefore, existing technologies cannot meet the urgent need for high-precision, high-efficiency, and low-cost attitude prediction of the driver's spine during the automotive cockpit design phase. Summary of the Invention

[0011] To address the problems in existing technologies that make it difficult to accurately assess driving posture due to reliance on complex human body parameters, oversimplification of models, and insufficient prediction accuracy, this invention provides a method for predicting the fine posture of the driver's spine based on cockpit layout parameters. This method establishes a quantitative mapping model between cockpit layout parameters and human body feature data that are easy to standardize and measure, and the posture of each segment of the spine, thereby achieving high-precision, high-efficiency, and low-cost prediction of spinal posture.

[0012] This invention is achieved through the following technical solution:

[0013] A method for predicting the fine posture of the driver's spine based on cockpit layout parameters includes the following steps:

[0014] S1. Collect the driver's anatomy data and cockpit layout parameters;

[0015] Human characteristic data includes the driver's height, weight, and the spatial coordinates of the head contact points and spinous process markers on the skin surface of key spinal segments when the driver is in a stable driving posture.

[0016] S2, Data Processing and Back Curve Fitting;

[0017] Body mass index (BMI) is calculated based on collected human feature data. The inflection point coordinates of the curve projected onto the driver's sagittal plane using the acquired spinous process markers on the skin surface of key spinal segments and the area between the C7 spinous process marker on the skin surface of the seventh cervical vertebra and the head contact point are then used to fit a back curve within the driver's sagittal plane. Based on the fitted curve, interpolation is performed with reference to pre-stored proportional relationships of spinal segment projection lengths to calculate a complete two-dimensional coordinate sequence from the head contact point to all spinous process markers of the lumbar vertebrae. The arc length distance Iflx from the head contact point to the first inflection point of the curve is calculated, and the curvature at the designated spinal segment marker is extracted.

[0018] S3. Spinal posture parameter prediction: Based on the cabin layout parameters collected in step S1 and the human feature parameters obtained in step S2, the polar angle and polar diameter of each spinal segment are calculated using a pre-established regression formula.

[0019] S4. Calculation of spinal spatial position: Using the spinous process markers on the skin surface of each spinal segment calculated in step S2 as the origin, and using the polar angle and polar diameter of the corresponding spinal segment calculated in step S3, the precise spatial position of the vertebral mass center of each spinal segment is determined in the sagittal coordinate system, thus completing the prediction of the fine posture of the spine.

[0020] Further, in step S1, the cockpit layout parameters include the accelerator pedal reference point PRP, the accelerator pedal heel point AHP, the midpoint of the line connecting the left and right eye points Eye, the occupant seating base point SgRP, the seat back angle A40, the vertical distance H30 from SgRP to AHP, the front-to-back horizontal distance L99 from SgRP to PRP, the front-to-back horizontal distance L6 from PRP to the steering wheel center, the front-to-back horizontal distance X_Eye from Eye to PRP, and the vertical distance Z_Eye from Eye to AHP.

[0021] Furthermore, step S1 specifically includes the following:

[0022] S11. Before the driver enters the cockpit, collect the driver's height Ht and weight Wt data; then, the driver sits in a normal pre-driving posture and adjusts to a comfortable position; after the sitting posture is stable, the driver holds the steering wheel with both hands, looks forward, and forms and maintains a stable driving posture; the visual sensor is used to accurately determine the spatial position of AHP and PRP in this coordinate system.

[0023] S12. Establish a two-dimensional coordinate system on the driver's sagittal plane: take the projection point of the accelerator pedal heel point AHP onto the horizontal plane passing through the accelerator pedal reference point PRP as the origin of the coordinate system, with the positive Z-axis pointing vertically upward and the positive X-axis pointing horizontally backward; where the eye-viewing direction is forward.

[0024] S13. Using a body pressure sensor and a vision sensor, collect spatial position data of the driver's head contact point, the spinous process marker C7 on the skin surface of the seventh cervical vertebra, the spinous process marker T7 on the skin surface of the seventh thoracic vertebra, and the spinous process marker L4 on the skin surface of the fourth lumbar vertebra.

[0025] Furthermore, step S13 specifically includes the following:

[0026] S131. Locate the position of the C7 marker point on the skin surface of the spinous process of the seventh cervical vertebra using a visual sensor;

[0027] S132. Determine the corresponding position of the inferior angles of the two scapulae on the backrest using the seat pressure distribution. Connect the inferior angles of the two scapulae with a line. The midpoint of the line is in the same coronal plane as the center of mass of the seventh thoracic vertebra T7_prof. Use the Z coordinate of the midpoint of the line to determine the coronal plane. The intersection of the coronal plane with the back skin on the driver's sagittal plane is the spinous process marker T7 on the skin surface of the seventh thoracic vertebra.

[0028] S133. Using a visual sensor, mark the highest points of the bilateral iliac crests and connect them. Locate the coronal plane position of the L4_prof segment of the fourth lumbar vertebra by the midpoint of the line. The intersection of this coronal plane and the back skin on the driver's sagittal plane is the skin surface spinous process marker L4 of the fourth lumbar vertebra. Then, use a visual sensor to measure the Z coordinate values ​​of the skin surface spinous process markers C7, T7, and L4 of the seventh cervical vertebra, the seventh thoracic vertebra, and the fourth lumbar vertebra, respectively.

[0029] S134. Based on the determined Z-coordinate values ​​of the spinous process markers on the skin surface, locate their respective horizontal planes, mark the corresponding anterior body markers, and calculate the X-coordinates of the corresponding spinous process markers by measuring the X-coordinates of the chair back surface measurement points that are the same as the Z-coordinates of the corresponding spinous process markers on the skin surface, the front-to-back horizontal distance from the anterior body marker to the corresponding seat surface measurement point, chest_SB_L, and the total trunk thickness TTD.

[0030] S135. Use a visual sensor to mark the head contact point and obtain its coordinates.

[0031] Furthermore, in step S2, cubic spline curves are used to fit the back shape;

[0032] The designated spinal segment markers include the spinous process marker C4 on the skin surface of the fourth cervical vertebra, the spinous process marker C5 on the skin surface of the fifth cervical vertebra, the spinous process marker C7 on the skin surface of the seventh cervical vertebra, the spinous process marker T3 on the skin surface of the third cervical vertebra, the spinous process marker T6 on the skin surface of the sixth thoracic vertebra, and the spinous process marker T9 on the skin surface of the ninth thoracic vertebra.

[0033] Further, in step S3, the regression formula includes a first set of formulas for calculating the polar angles of each spinal segment of the cervical spine Ci_prof, thoracic spine Ti_prof, and lumbar spine Li_prof, and a second set of formulas for calculating the polar diameters of each spinal segment of the cervical spine Ci_prof, thoracic spine Ti_prof, and lumbar spine Li_prof.

[0034] Furthermore, the input variables of the first set of formulas and the second set of formulas are independently selected from the following groups of variables: Wt, X_Eye, Z_Eye, L99, A40, H30, L6, Ht, BMI, Iflx, kC4, kC5, kC7, kT3, kT6, kT9, cosCi, cosTi, cosLi, and combinations thereof; where kC4, kC5, kC7, kT3, and kT6 represent the curvature of the fitted curves at the spinous process markers C4, C5, C7, T3, T6, and T9 on the skin surface of the spinal segment, respectively, and cosCi / cosTi / cosLi represent the cosine values ​​of the polar angles of the corresponding spinal segments.

[0035] Furthermore, in step S4, the polar angle and polar radius are used to represent the spatial position of the center of mass of each vertebral segment relative to the spinous process marker point on the skin surface in polar coordinates in the sagittal coordinate system.

[0036] Compared with the prior art, the advantages of the present invention are as follows:

[0037] 1. This invention overcomes the limitation of existing models that oversimplify the spine into three segments, and achieves precise prediction of the spatial position of all vertebral segments of the spine.

[0038] 2. A direct and quantitative mapping model between cockpit layout parameters and spinal segment posture was established, abandoning the complex modeling method that relies on human physiological parameters. Instead, existing, standardized, and measurable layout parameters in the car cockpit were used as model inputs, significantly reducing implementation costs and difficulties.

[0039] 3. By establishing a quantitative mapping relationship between easily measurable cockpit hardpoint parameters and the posture of each segment of the spine, accurate and efficient prediction of the human spine in driving posture can be achieved. Attached Figure Description

[0040] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0041] Figure 1 This is a flowchart illustrating a method for predicting the fine posture of the driver's spine based on cockpit layout parameters according to the present invention.

[0042] Figure 2 A schematic diagram of the cockpit layout parameters;

[0043] Figure 3 A schematic diagram of the human body posture and skeletal structure viewed from the front during driving;

[0044] Figure 4 This is a schematic diagram showing the relative position and parameters of the spine;

[0045] In the figure: 1. Marker point on the front of the torso; 2. Head contact point; 3. Measurement point on the seat surface; 4. C1_prof; 5. C7_prof; 6. Ci_prof (C1_prof~C7_prof); 7. T1_prof; 8. Ti_prof (T1_prof~T12_prof); 9. T12_prof; 10. L1_prof; 11. Li (L1_prof~L5_prof); 12. L5_prof; 13. Local magnification of the relative position of a certain spinal segment; 14. Back fitting curve; 15. Negative normal of the back fitting curve at the spinous process marker point on the spinal skin surface; 16. Polar diameter; 17. Spinous process marker point on the spinal skin surface; 18. Polar angle; 19. Vertebral mass center. Detailed Implementation

[0046] To clearly and completely describe the technical solution of the present invention, the specific embodiments of the present invention are as follows, in conjunction with the accompanying drawings:

[0047] Example 1

[0048] like Figure 1 As shown, this embodiment provides a method for predicting the fine posture of the driver's spine based on cockpit layout parameters, specifically including the following steps:

[0049] S1. Collect the driver's anthropometric data and cockpit layout parameters, specifically including the following:

[0050] S11. User human body feature collection;

[0051] Before the driver enters the cockpit, the driver's height (Ht) and weight (Wt) are collected. The driver then sits in a normal, ready-to-drive posture, adjusting the seat's fore-aft position, height, and backrest angle to a comfortable position according to personal preference. Once the seating position is stable, the left foot is placed naturally in its preferred position, the right heel firmly contacts the floor, and the ball of the right foot lightly rests on the accelerator pedal. The whole body is relaxed, both hands grip the steering wheel, and the eyes look forward, forming a stable driving posture, as detailed in Table 1.

[0052] Table 1. Human characteristic variables required for spinal posture prediction

[0053]

[0054]

[0055] S12. Establish a coordinate system;

[0056] Establish a two-dimensional coordinate system on the driver's sagittal plane: take the projection point of the accelerator pedal heel point AHP onto the horizontal plane passing through the accelerator pedal reference point PRP as the origin of the coordinate system, with the positive Z-axis pointing vertically upward and the positive X-axis pointing horizontally backward (the visual direction is forward).

[0057] S13. Collect the coordinates of key spinal markers;

[0058] Using specialized equipment such as body pressure sensors and vision sensors, spatial position data were collected for the driver's head contact point, the spinous process markers C7, T7, and L4 on the skin surface of the seventh cervical vertebra, and the seventh thoracic vertebra; specifically including:

[0059] S131. Accurately locate the spinous process marker C7 on the skin surface of the seventh cervical vertebra (its anatomical feature is the inflection point of the spinous process at the transition segment of the skin surface of the neck and chest) using a visual sensor.

[0060] S132. Determine the corresponding position of the inferior angles of the bilateral scapulae on the backrest using the seat pressure distribution. The corresponding position area is usually the pressure concentration area of ​​the thoracic vertebrae. Connect the two inferior angle points. The midpoint of the line is in the same coronal plane as the center of mass of the seventh thoracic vertebra T7_prof. Use the Z coordinate of the midpoint of the line to determine this coronal plane. The intersection of this coronal plane with the back skin on the driver's sagittal plane is the spinous process marker point C7 on the skin surface of the seventh cervical vertebra.

[0061] S133. Mark the highest points of the bilateral iliac crests using a visual sensor and connect them. Locate the coronal plane position of the L4_prof segment of the fourth lumbar vertebra by the midpoint of the line. The method for locating the skin surface markers corresponding to L4_prof is the same as that for T7.

[0062] S134. Measure the Z coordinate values ​​of the spinous process markers C7, T7 and L4 on each skin surface using a visual sensor;

[0063] S135. Based on the determined Z-coordinate values, locate the horizontal plane where the three spinous process markers are located, mark the corresponding anterior body markers, measure the chest_SB_L and X_SBi variable data, and obtain the total torso thickness TTD data of this horizontal plane through a pre-stored human body parameter database or direct measurement. Calculate the anterior-posterior horizontal distance Ci_SB_L / Ti_SB_L / Li_SB_L from the skin surface spinous process markers to the seat surface and related values ​​(Ci_SB_L / Ti_SB_L / Li_SB_L = chest_SB_L - TTD). Finally, combine the X_SBi variable data to calculate the X coordinate of each skin surface spinous process marker (X coordinate of each skin surface spinous process marker = X_SBi - Ci_SB_L / Ti_SB_L / Li_SB_L).

[0064] S136. Use a visual sensor to mark the head contact point and obtain its coordinates;

[0065] S137. Using the acquired C7 point and head contact point, locate the skin area between the two points, and use a visual sensor to obtain the inflection point coordinates of the curve projected onto the sagittal plane of the skin area.

[0066] S14. Collect cockpit layout parameters, such as Figure 2 As shown;

[0067] In the cockpit layout environment determined in step S12, the SgRP point (i.e., the H-point, hip point, referring to the hinge center between the human model or driver's torso and thigh) and the seat back angle A40 are accurately measured using a visual sensor or H-point measuring device. Simultaneously, combined with the acquired AHP and PRP spatial positions, five cockpit layout parameters—L6, H30, L99, X_Eye, and Z_Eye—are measured using visual sensors, displacement sensors, and other equipment. All ultimately recorded data are the projection distances of these parameters onto the driver's sagittal plane; see Table 2 for details.

[0068] Table 2. Cockpit layout parameters required for spinal posture prediction

[0069]

[0070] Step S2, Data Processing and Back Curve Fitting:

[0071] S21. Calculate BMI;

[0072] Calculate the user's BMI based on height (Ht) and weight (Wt).

[0073] S22, Back curve fitting;

[0074] Five coordinate points were obtained: the head contact point, the coordinates of the skin surface spinous process markers C7, T7, and L4, and the inflection point coordinates of the curve projected onto the sagittal plane from a specific skin region. A cubic spline curve was then used to fit the shape of the back, with the smoothing parameter set to [value missing]. The weight ratios of the five points are 1:32:16:1:1. After successful fitting, the spinous process markers on the skin surface of the spinous segments are interpolated on the fitted curve according to the order of their arrangement, referring to the proportional relationship of the projection length (Z-direction projection) in Table 3. Finally, the complete two-dimensional coordinate sequence from the head contact point to all spinous process points of the lumbar spine is calculated.

[0075] Table 3. Proportional Relationship of Spinal Segment Projection Length

[0076]

[0077] S23. Calculate the characteristic parameters;

[0078] Based on the fitted curve, the arc length from the head contact point to the first inflection point of the curve is calculated and denoted as Iflx. At the same time, the curvature of the fitted curve at the spinous process markers C4, C5, C7, T3, T6, and T9 on the skin surface of the spinous process segment is extracted and denoted as kC4, kC5, kC7, kT3, kT6, and kT9, respectively.

[0079] Step S3, Spinal Posture Parameter Prediction:

[0080] Based on the cabin layout parameters collected in step S1 and the human feature data obtained in step S2, the polar angles and polar diameters of each spinal segment (cervical Ci_prof, thoracic Ti_prof, and lumbar Li_prof) are calculated using a pre-established regression formula (with distance data in mm and angles in °). A schematic diagram of the parameters can be found in [reference needed]. Figure 4 .

[0081] The regression formula is constructed as follows:

[0082] The method for constructing the regression formula mainly includes four stages: data collection and sample construction, feature extraction and preprocessing, regression model training, and model validation and optimization.

[0083] Data Acquisition and Sample Construction. Drivers with different body types, including height, weight, and age, were selected as experimental samples. Under standard driving posture, anthropometric data (including height, weight, and seat pressure distribution) and cabin layout parameters (such as AHP, PRP, SgRP, A40, H30, L99, L6, X_Eye, and Z_Eye) were simultaneously collected for each driver. The actual spatial coordinates of the vertebral mass centers in the sagittal plane of each spinal segment were obtained using medical imaging equipment (such as X-rays or MRI) as ground truth data for subsequent model training and validation. Simultaneously, the coordinate data of spinous process markers on the skin surface of each spinal segment were extracted from the medical images.

[0084] Feature extraction and preprocessing. Based on the collected coordinates of marker points on the skin surface of the spine, a back curve was fitted, and key geometric feature parameters (kC4, kC5, kC7, kT3, kT6, kT9, Iflx) were extracted. These parameters, combined with cockpit layout parameters and driver anthropometric data, were used to construct a complete candidate set of independent variables. All data were standardized to eliminate the influence of dimensions and improve the stability and convergence efficiency of model training.

[0085] Regression model training. The polar angle and diameter of each spinal segment were used as the prediction targets (i.e., dependent variables), and a multiple linear regression method was employed for modeling. Feature selection was performed through stepwise regression analysis and ridge regression to filter out a subset of variables that significantly influenced the posture prediction of each spinal segment from a large pool of candidate independent variables, and the corresponding regression coefficients were obtained by fitting the model.

[0086] Model Validation and Optimization. Cross-validation was used to evaluate the performance of the trained regression model. The sample data was divided into training and validation sets, and the training and validation process was repeated multiple times to objectively evaluate the model's prediction accuracy and stability. Based on the validation results, the model was further optimized. Finally, the regression formulas for the polar angles and diameters of each spinal segment were obtained, as shown in Tables 4 and 5.

[0087] The final regression formulas are shown in Tables 4 and 5:

[0088] Table 4 Regression formulas for polar angles

[0089]

[0090]

[0091] Table 5 Regression formulas for the extreme diameter

[0092]

[0093] Step S4, Calculation of spinal spatial position:

[0094] Using the spinous process markers on the skin surface of each spinal segment calculated in step S2 as the origin, and the polar angles and polar diameters of the corresponding segments calculated in step S3, the precise spatial position of the vertebral mass center of each spinal segment is determined in the sagittal coordinate system. The calculated angle values ​​are first modulo 180°, and then the supplementary angle of the modulo is calculated. The smaller angle between the modulo angle and its supplementary angle is the polar angle; the polar diameter is directly calculated using a formula. Based on polar coordinate transformation, the two-dimensional coordinates of the vertebral mass center can be obtained, thus completing the prediction of the fine posture of the spine. Since the obtained polar angles have been folded to the 0°–90° range and no longer possess directionality, a unified definition of direction is now added: the polar angles of spinal segments C1_prof to C6_prof are based on the negative normal of the back fitting curve at the corresponding spinous process marker on the skin surface, with counterclockwise rotation being positive; the polar angles of other spinal segments are based on the negative normal under the same definition, with clockwise rotation being positive.

[0095] Example 2

[0096] This embodiment provides a method for predicting the fine posture of the driver's spine based on cockpit layout parameters, specifically including the following steps:

[0097] S1. Collect the driver's anthropometric data and cockpit layout parameters, specifically including the following:

[0098] S11. User human body feature collection;

[0099] Driver's anatomy data: Ht=1800mm, Wt=75Kg.

[0100] S12. Establish a coordinate system;

[0101] Establish a two-dimensional coordinate system on the driver's sagittal plane: take the projection point of the accelerator pedal heel point AHP onto the horizontal plane passing through the accelerator pedal reference point PRP as the origin of the coordinate system, with the positive Z-axis pointing vertically upward and the positive X-axis pointing horizontally backward (the visual direction is forward).

[0102] S13. Collect the coordinates of key spinal markers;

[0103] Using specialized equipment such as body pressure sensors and vision sensors, spatial location data were collected for the driver's head contact point, the spinous process markers C7, T7, and L4 on the skin surface of the seventh cervical vertebra, and the seventh thoracic vertebra. Head contact point: (966.5060, 1026.7089); C7: (962.4673, 882.2639); T7: (1014.5103, 688.0361); L4: (1018.5904, 475.3068). Based on the acquired C7 point and head contact point, the skin region between the two points was located. The inflection point coordinates of the curve projected onto the sagittal plane by the vision sensor were obtained: (940.0091, 920.3558).

[0104] S14. Collect cockpit layout parameters, such as Figure 2 As shown;

[0105] L6=496.80; H30=387.80; L99=813.44; X_Eye=943.22; Z_Eye=1018.21; All data are in mm.

[0106] Step S2, Data Processing and Back Curve Fitting:

[0107] S21. Calculate BMI;

[0108] BMI = 23.148;

[0109] S22, Back curve fitting;

[0110] By fitting the back curve and referring to Table 3, the proportional relationship of the projection length of the spinal segments, a complete two-dimensional coordinate sequence from the head contact point to all spinous processes of the lumbar vertebrae was obtained, as shown in Table 6.

[0111] Table 6. Complete two-dimensional coordinate sequence of skin surface spinous process markers

[0112]

[0113] S23. Calculate the characteristic parameters;

[0114] Based on the fitted curve, the characteristic parameters are calculated.

[0115] Iflx=30.1;kC4=0.0104;kC5=0.021;kC7=0.005;kT3=0.0019;kT6=0.003;kT9=0.0023.

[0116] Step S3, Spinal Posture Parameter Prediction:

[0117] Based on the cabin layout parameters collected in step S1 and the human feature data obtained in step S2, the polar angles and polar diameters of each spinal segment (cervical Ci_prof, thoracic Ti_prof, and lumbar Li_prof) are calculated using a pre-established regression formula (with distance data in mm and angles in °). The processed results are as follows:

[0118] Table 7 Calculation results of polar angle and polar radius

[0119]

[0120] Step S4, Calculation of spinal spatial position:

[0121] Using the spinous process markers on the skin surface of each spinal segment calculated in step S2 as the origin, and the polar angles and polar diameters of the corresponding segments calculated in step S3, the precise spatial position of the vertebral mass center of each spinal segment is determined in the sagittal coordinate system, thereby completing the prediction of the fine posture of the spine. The two-dimensional coordinates of the predicted vertebral mass center are shown in Table 8.

[0122] Table 8 Two-dimensional coordinates of each vertebral mass center

[0123]

[0124] Through specific implementation and calculation in Example 2, the prediction method of the present invention was verified. By comparing the predicted coordinates of the vertebral mass centers of each spinal segment with the actual coordinates of the corresponding segments acquired through medical imaging (such as X-rays or MRI), the statistical results showed that the average root mean square error of the predicted position of the entire spine was (12.0 ± 3.9) mm. This level of accuracy fully meets the refined evaluation requirements for positive design of automotive cabin ergonomics.

[0125] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various simple modifications can be made to the technical solution of the present invention, and these simple modifications all fall within the protection scope of the present invention.

[0126] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not describe the various possible combinations separately.

[0127] Furthermore, various different embodiments of the present invention can be combined in any way, as long as they do not violate the spirit of the present invention, they should also be regarded as the content disclosed by the present invention.

Claims

1. A driver spine fine posture prediction method based on cockpit layout parameters, characterized in that, Specifically comprising the following steps: S1, collecting the human body feature data of the driver and the cabin arrangement parameters; The human body feature data includes the height and weight of the driver, the spatial coordinates of the head contact point and the key spinal segment skin surface spinous process marker points of the driver in a stable driving posture; S2, data processing and back curve fitting; Based on the collected human body feature data, the body mass index is calculated; and the coordinates of the inflection points of the curve projected on the sagittal plane of the driver between the key spinal segment skin surface spinous process marker points and the skin surface spinous process marker point of the seventh cervical vertebra and the head contact point are obtained, the back curve fitting is performed in the sagittal plane of the driver, based on the obtained fitting curve, the pre-stored spinal segment projection length proportion relationship is referred to for interpolation, the complete two-dimensional coordinate sequence from the head contact point to all the spinous process marker points of the lumbar vertebrae is calculated; and the arc length distance Iflx from the head contact point to the first inflection point of the curve is calculated, and the curvature at the specified spinal segment marker point is extracted; S3, spinal posture parameter prediction: based on the cabin arrangement parameters collected in step S1 and the human body feature parameters obtained in step S2, the polar angle and polar radius of each spinal segment are calculated through a pre-established regression formula; S4, spinal space position calculation: taking each spinal segment skin surface spinous process marker point obtained in step S2 as the origin, the polar angle and polar radius of the corresponding spinal segment obtained in step S3 are used to determine the accurate spatial position of the mass center of each spinal segment vertebra in the sagittal plane coordinate system, and the prediction of the fine posture of the spine is completed; In step S1, the cabin arrangement parameters include the accelerator pedal reference point (PRP), the accelerator pedal heel point (AHP), the midpoint of the left and right eye points (Eye), the passenger seating basic point (SgRP), the seat back angle (A40), the vertical distance (H30) from the passenger seating basic point (SgRP) to the accelerator pedal heel point (AHP), the front-back horizontal distance L99 from the passenger seating basic point (SgRP) to the accelerator pedal reference point (PRP), the front-back horizontal distance L6 from the accelerator pedal reference point (PRP) to the steering wheel center, the front-back horizontal distance X_Eye from the midpoint of the left and right eye points (Eye) to the accelerator pedal reference point (PRP), and the vertical distance Z_Eye from the midpoint of the left and right eye points (Eye) to the accelerator pedal heel point (AHP); Step S1 specifically includes the following content: S11, before the driver enters the cabin, the height Ht and weight Wt data of the driver are collected; then, the driver sits in a normal ready-to-drive posture and adjusts to a comfortable state; after the sitting posture is stable, the driver holds the steering wheel with both hands, looks forward, forms and maintains a stable driving posture; The spatial positions of the accelerator pedal heel point (AHP) and the accelerator pedal reference point (PRP) in the coordinate system are accurately determined by using a visual sensor; S12, a two-dimensional coordinate system is established on the sagittal plane of the driver: taking the projection point of the accelerator pedal heel point (AHP) on the horizontal plane passing through the accelerator pedal reference point (PRP) as the coordinate origin, the positive direction of the Z-axis is vertically upward, and the positive direction of the X-axis is horizontally backward; wherein, the visual direction is forward; S13, the body pressure sensor and the vision sensor are used to collect the spatial position data of the head contact point, the skin surface spinous process marker point C7 of the seventh cervical vertebra, the skin surface spinous process marker point T7 of the seventh thoracic vertebra, and the skin surface spinous process marker point L4 of the fourth lumbar vertebra; Step S13 specifically includes the following contents: S131, the vision sensor is used to locate the position of the skin surface spinous process marker point C7 of the seventh cervical vertebra; S132, the corresponding positions of the bilateral lower corners of scapula on the backrest are determined by the seat pressure distribution, the bilateral lower corners of scapula are connected, the midpoint of the connecting line is on the same coronal plane as the vertebral barycenter of the seventh thoracic vertebra T7_prof, the Z coordinate of the midpoint of the connecting line is used to determine the coronal plane, and the intersection of the coronal plane and the back skin on the sagittal plane of the driver is the skin surface spinous process marker point T7 of the seventh thoracic vertebra; S133, the highest points of the bilateral iliac crests are marked by the vision sensor and connected, the coronal plane position where the vertebral barycenter of the fourth lumbar vertebra L4_prof segment is located is located by the midpoint of the connecting line, the intersection of the coronal plane and the back skin on the sagittal plane of the driver is the skin surface spinous process marker point L4 of the fourth lumbar vertebra, and the Z coordinate values of the skin surface spinous process marker points C7, T7, and L4 of the seventh cervical vertebra, the seventh thoracic vertebra, and the fourth lumbar vertebra are measured by the vision sensor; S134, based on the Z coordinate values of the skin surface spinous process marker points, the horizontal planes where the skin surface spinous process marker points are located are located, the corresponding anterior body marker points are marked, the X coordinate values X_SBi of the chair back surface measurement points corresponding to the Z coordinate values of the corresponding spinal skin surface spinous process marker points, the front-back horizontal distances chest_SB_L from the anterior body marker points to the corresponding chair surface measurement points, and the total thickness TTD of the torso are measured, and the X coordinates of the corresponding skin surface spinous process marker points are calculated; S135, the head contact point is marked by the vision sensor, and the coordinate of the head contact point is obtained.

2. The method of claim 1, wherein the method further comprises: In step S2, the back shape is fitted by using a cubic spline curve; The specified spinal segment marker points include a fourth cervical vertebra skin surface spinous process marker point C4, a fifth cervical vertebra skin surface spinous process marker point C5, a seventh cervical vertebra skin surface spinous process marker point C7, a third cervical vertebra skin surface spinous process marker point T3, a sixth thoracic vertebra skin surface spinous process marker point T6, and a ninth thoracic vertebra skin surface spinous process marker point T9.

3. The method of claim 1, wherein, In step S3, the regression formula includes a first group of formulas for calculating the polar angle of each spinal segment of the cervical vertebra Ci_prof, the thoracic vertebra Ti_prof, and the lumbar vertebra Li_prof, and a second group of formulas for calculating the polar radius of each spinal segment of the cervical vertebra Ci_prof, the thoracic vertebra Ti_prof, and the lumbar vertebra Li_prof.

4. The method of claim 3, wherein the method further comprises: The input variables of the first set of formulas and the second set of formulas are independently selected from the group consisting of Wt, X_Eye, Z_Eye, L99, A40, H30, L6, Ht, BMI, Iflx, kC4, kC5, kC7, kT3, kT6, kT9, cosCi, cosTi, cosLi, and combinations thereof; wherein kC4, kC5, kC7, kT3, kT6 represent the curvatures of the fitted curve at the spinous process marker points C4, C5, C7, T3, T6, T9 on the skin surface of the spinal segment, and cosCi / cosTi / cosLi represent the cosine values of the polar angle of the corresponding spinal segment.

5. The method of claim 1, wherein, In step S4, the polar angle and polar radius are used to represent the spatial position of the center of mass of each vertebra of the spinal segment relative to the spinous process marker points on the skin surface in polar coordinates in the sagittal plane coordinate system.

Images