CN116956379A - Method for constructing passenger parameterized human body contour prediction model - Google Patents
Method for constructing passenger parameterized human body contour prediction model Download PDFInfo
- Publication number
- CN116956379A CN116956379A CN202311217113.2A CN202311217113A CN116956379A CN 116956379 A CN116956379 A CN 116956379A CN 202311217113 A CN202311217113 A CN 202311217113A CN 116956379 A CN116956379 A CN 116956379A
- Authority
- CN
- China
- Prior art keywords
- human body
- prediction model
- boundary
- parameters
- body contour
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 230000007246 mechanism Effects 0.000 claims abstract description 25
- 238000000611 regression analysis Methods 0.000 claims abstract description 12
- 238000009826 distribution Methods 0.000 claims abstract description 8
- 210000003414 extremity Anatomy 0.000 claims description 31
- 238000007493 shaping process Methods 0.000 claims description 30
- 238000013461 design Methods 0.000 claims description 26
- 210000000689 upper leg Anatomy 0.000 claims description 21
- 210000002310 elbow joint Anatomy 0.000 claims description 15
- 210000001699 lower leg Anatomy 0.000 claims description 14
- 210000002683 foot Anatomy 0.000 claims description 13
- 210000000245 forearm Anatomy 0.000 claims description 11
- 210000000629 knee joint Anatomy 0.000 claims description 11
- 210000000323 shoulder joint Anatomy 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 11
- 210000003857 wrist joint Anatomy 0.000 claims description 11
- 210000000544 articulatio talocruralis Anatomy 0.000 claims description 10
- 238000002474 experimental method Methods 0.000 claims description 8
- 210000004394 hip joint Anatomy 0.000 claims description 7
- 210000001503 joint Anatomy 0.000 claims description 7
- 244000309466 calf Species 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 4
- 210000003739 neck Anatomy 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 9
- 230000008859 change Effects 0.000 abstract description 7
- 238000004364 calculation method Methods 0.000 description 14
- 230000036544 posture Effects 0.000 description 8
- 238000005457 optimization Methods 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 238000012942 design verification Methods 0.000 description 4
- 210000001624 hip Anatomy 0.000 description 4
- 238000012417 linear regression Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000002829 reductive effect Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 210000003127 knee Anatomy 0.000 description 2
- 210000002414 leg Anatomy 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 210000003423 ankle Anatomy 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Computer Hardware Design (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a method for constructing a passenger parameterized human body contour prediction model, which belongs to the technical field of automobile ergonomics and comprises the following steps: selecting control variables and user samples to obtain a series of occupant human body contour data sample sets under specific control variable distribution; simplifying and extracting human body profile parameters of a user sample; regression analysis of human body profile parameters; establishing a human body contour parameterized model; checking and correcting the prediction model. The model constructed by the method can meet the layout requirements of large-scale free adjustment and change of cabin mechanisms and passengers in the environment of automatic driving and intelligent cabins in the future, does not need layout parameters of a traditional cabin operating mechanism, can enrich user samples and is flexibly combined with the environment of the intelligent cabins; in addition, the method can also realize the prediction of the size, the posture and the human body contour of the human body which are continuously changed, and further assist the intelligent control to avoid collision or interference possibly generated in the dynamic adjustment process.
Description
Technical Field
The invention belongs to the technical field of automobile ergonomics, and particularly relates to a method for constructing a passenger parameterized human body contour prediction model.
Background
In car cabin design and arrangement, the layout of the passenger space is closely related to driving operation, comfort, health and even safety of the passengers. It therefore plays an important role in the ergonomic design of automobiles. In conventional occupant space arrangement designs, the occupant seating position and seating arrangement should meet the requirements for comfortable seating for the target occupant group; the necessary space (such as leg space and head space) is reserved for passengers in the cockpit, so that the driver can drive stably and accurately; it should also be ensured that the driver and the rear passengers can easily adjust the body position, ensuring the comfort and safety of driving and driving. The most important basis for the spatial arrangement of the occupants is the size of the human body. For driving space, due to the individual size differences of the target driving population, the cabin organization must have a certain adjustment range to meet the driving and riding requirements of most people in the target population.
With the continuous progress of the automobile industry, automobiles have undergone profound changes in driving and steering performance, appearance and modeling aesthetics, production and manufacturing processes and the like, and have rapidly progressed toward intelligentization. In recent years, the degree of intellectualization of automobiles is remarkably improved, and as an important component of automobile intellectualization, the improvement of driving experience by an intelligent cabin is the most direct and remarkable feeling. The future intelligent cabins are far from meeting the driving space as a common vehicle, and become a third largest living space for people outside the families and workplaces. At this time, the pursuit of people is no longer a simple driving interaction, and the comfort in driving is becoming one of the key points of intelligent cabin design.
In the process, the traditional driving operation is gradually reduced, and the traditional man-machine interaction modes such as a steering wheel, a gear shifting mechanism and pedal operation are replaced by the more intelligent intangible interaction mode, so that the postures of passengers in the cabin can be more freely changed. In order to pursue ride comfort, the cabin mechanism will also have a more free, wider range of adjustment forms, which will tend to contradict the limited space within the cabin, so that the mechanism in the cabin may collide or interfere during dynamic adjustment. On the premise that passengers have comfortable postures, the cabin mechanism can be controlled in a limited space, collision and interference are avoided, and the cabin mechanism is an important target for the ergonomic design of the automobile cabin.
In order to achieve the above object, it is also an important task to determine the possible comfort posture in which the occupant is normally seated, and the spatial arrangement boundaries at that time. Therefore, the space boundary in the dynamic adjustment process of the cabin mechanism can be further calculated, and the cabin mechanism is further controlled. Therefore, how to relatively accurately predict the postures of passengers in the cabin and obtain the general rules of the human body profile change of different passengers, and the method is suitable for dynamic regulation control of mechanisms in the cabin, and is an important problem to be solved.
Early cabin arrangement is designed by a large scale of hand drawing paper, a human body template with a specific size is used for representing the position of an occupant, and the passenger is placed in the arrangement diagram for design verification and iterative optimization. In the later stage, a sample with a certain proportion or equal proportion can be manufactured to simulate an actual product, and the design verification is performed by using the elastic dummy model with the joints. The method for carrying out design verification by the traditional arrangement dummy has low efficiency, relies on iterative design optimization, consumes a large amount of manpower and material resources, has limited accuracy of arrangement test, cannot verify large-scale dynamic adjustment, and almost cannot meet the arrangement requirement of the intelligent cabin in the future.
There are also considerable limitations to the use of existing computer-aided ergonomic software. On one hand, the existing computer-aided ergonomic software is developed based on the traditional cabin arrangement method, and still depends on the arrangement parameters of the traditional cabin, such as smaller cushion angle and backrest angle, fixed positions of heel points and pedals, and the like; on the other hand, while the models in many software contain many degrees of freedom capable of simulating various gestures, they remain in a stage of static or quasi-static gesture models, and only can adjust changes within a relatively small range, so that it is difficult to meet the arrangement requirements of large-range free adjustment changes of mechanisms and passengers in the intelligent cabins in the future.
The application range of the existing prediction model established based on the experimental statistical method is limited by the traditional arrangement experimental data, and the experimental sample is quite limited, so that a flexible model suitable for the future intelligent cabin environment cannot be provided. In addition, for researches on gesture prediction models and spatial arrangement boundaries of passengers in a cockpit, most of the researches only analyze trends of preference gestures or influence factors thereof after experimental simulation, and only a generalized conclusion aiming at human bodies and gestures is obtained, and the method is not combined with a specific cockpit environment, so that the accuracy is limited.
Furthermore, the above methods often utilize human body dimensions of a particular percentile (male 95 percentile, male and female 50 percentile, female 5 percentile) based on conventional design experience and methods within the industry. While this has enabled providing adequate tuning range and design verification for traditional non-intelligent cabins, in future intelligent cabin environments, intelligent control and tuning for each user should also be enabled. At this time, only discrete boundary human body sizes cannot be used any more, and a continuous human body size, posture and human body contour prediction model should be constructed, so that intelligent control is assisted after a sensing link.
The numerical simulation from the biomechanics and kinematics angles has the advantages that the result is more accurate, the human body data can be flexibly changed, and the like, but the simulation calculation and the inverse kinematics solution need to be input into the space position of the real-time limb tail end, a large number of calculation iterations need to be carried out to optimize the target, and the calculation cost is high. In addition, the method is limited by simulation calculation and inverse kinematics solution calculation principles, the obtained predicted gesture may be distorted when the variation range is large, and a large number of limiting targets are needed to control the comfort and even the rationality of the gesture. Therefore, the method is more suitable for predicting the gesture of the passengers and generally verifying the future intelligent cabin space, and is relatively not suitable for real-time anti-collision and anti-interference control of the intelligent cabin mechanism.
Therefore, it is highly desirable to construct a model for predicting the posture and contour of a passenger with high efficiency, low cost and high accuracy.
Disclosure of Invention
Aiming at the problems that how to ensure that under the premise that passengers have comfortable postures, cabin mechanisms can be controlled in a limited space and collision and interference are avoided in the prior art, the invention provides a method for constructing a passenger parameterized human body contour prediction model, the model constructed by the method can meet the arrangement requirements of the cabin mechanisms and passengers in the future automatic driving and intelligent cabin environment for large-scale free adjustment and change, the arrangement parameters of the traditional cabin operating mechanism are not needed, and user samples can be enriched and flexibly combined with the intelligent cabin environment; in addition, the method can also realize the prediction of the size, the posture and the human body contour of the human body which are continuously changed, and further assist the intelligent control to avoid collision or interference possibly generated in the dynamic adjustment process. In order to better assist intelligent control, calculation iteration and optimization target limitation are avoided in the calculation of the prediction model, so that calculation cost is reduced.
The invention is realized by the following technical scheme:
the method for constructing the passenger parameterized human body contour prediction model specifically comprises the following steps:
s1, selecting control variables and user samples to obtain a series of passenger human body contour data sample sets under specific control variable distribution;
s2, simplifying and extracting human body profile parameters of a user sample;
the human body profile parameters comprise a shaping size and a positioning size; the shaping dimension is used for determining the size of the outline of each limb segment; the positioning size is used for determining the position relation of each key joint point relative to the wire frame to which the key joint point belongs;
s3, carrying out regression analysis on human body contour parameters;
s31, regression of the shaping parameters of the human body contour;
taking the height, weight and sitting height of the passenger as control variable parameters to obtain a parameter equation for predicting the shaping size;
s32, regression of human body contour positioning parameters;
selecting proper control variables and a regression method to obtain a parameter equation of a predicted positioning size;
s4, establishing a human body contour parameterized model;
s5, checking and correcting the prediction model.
Further, in step S1, the control variables include a control variable related to a human body size and a control variable related to a cabin arrangement;
the user sample is derived from bench test, agent model and benchmarking data.
Further, the control variables related to the human body size include the height, weight and sitting height of the passenger;
the control variables related to cabin arrangement are selected according to design targets, so that passenger arrangement boundaries are ensured to be changed, and cabin mechanism adjustment and control parameters are key design indexes.
Further, in step S2, the simplifying and extracting parameters of human body contours of the user sample specifically includes the following contents:
firstly, marking key joint points of side projection of a human body contour of a sample, wherein the key joint points comprise 7 ankle joints, knee joints, hip joints, shoulder joints, elbow joints, wrist joints and neck joints;
then, marking each limb segment including trunk, thigh, shank, foot, big arm, small arm, hand and head by using rectangular wire frame;
finally, marking human body contour parameters;
wherein the shaping dimensions are the length and width of a rectangular wireframe of each limb segment, and the positioning dimensions include: 15 boundary positioning dimensions, 8 included angle dimensions, and 5 pitch dimensions for each critical node.
Further, the 15 boundary positioning dimensions include distances of the cervical joint to the lower head boundary and the rear head boundary; the distance from the shoulder joint to the upper torso boundary, the front torso boundary, and the rear arm boundary; the distance of the elbow joint to the posterior border of the forearm; the distance from the wrist joint to the back hand boundary and the upper hand boundary; the distance of the hip joint to the lower torso boundary and the front torso boundary; the distance of the knee joint to the anterior thigh boundary, the lower thigh boundary, and the posterior calf boundary; the distance from the ankle joint to the upper boundary of the foot and the rear boundary of the foot step; the included angle size comprises included angles between each limb segment and the horizontal plane; the spacing dimension of each key articulation point comprises the spacing of each key articulation point, including trunk length, thigh length, shank length, big arm length and small arm length.
Further, the rectangular wire frame should fully encompass each limb segment, which should be parallel to the skeleton line, i.e. the connection of the corresponding critical articulation points on the respective limb segment.
Further, the ankle joint and the knee joint are positioned on the symmetry axis of the rectangular shank wire frame; the shoulder joint and the elbow joint are positioned on the symmetry axis of the rectangular wire frame of the big arm; the elbow joint and the wrist joint are located on the symmetry axis of the rectangular wire frame of the forearm.
Further, the step S4 specifically includes the following:
firstly, setting control parameters including the height, weight, sitting height and cabin arrangement related control variables of an occupant, then taking the initial constraint shaping size and positioning size as controlled parameters, and associating the controlled parameters with the control parameters according to a parameter equation obtained by regression analysis in the step S3 to obtain a parameterized occupant human body contour prediction model.
Further, in step S5, the checking of the prediction model includes: comparing and checking the prediction result with the original sample data and comparing and checking the prediction result with the actual contour;
the comparison and check of the prediction result and the original sample data comprises the following steps: randomly selecting original sample data under different control variables, inputting the control variables into a prediction model, comparing the obtained prediction result with the original sample, and checking whether the relative error is in an acceptable degree;
the comparison and check of the prediction result and the actual contour comprises the following steps: selecting a control variable with a value different from a user sample to perform an experiment, comparing a predicted result under the same corresponding control parameter with an experimental result, and checking whether the relative error is in an acceptable degree;
if the checking result is not ideal, analyzing checking deviation or regression residual of the characteristic test points or the characteristic samples, adding correction terms into a parameter equation of the prediction model, and compensating and correcting the prediction model according to the checking deviation and the regression residual.
Compared with the prior art, the invention has the following advantages:
1. the method for constructing the passenger parameterized human body contour prediction model selects parameterized design to construct the passenger human body contour model, and has the characteristics of real-time variation of the model along with parameters of control variables, capability of covering target passenger groups with different human body sizes and continuous variation, capability of increasing and decreasing the control variables to obtain prediction models under the control of different variables, and the like. Therefore, the parameterized human body contour prediction model is not only suitable for obtaining the spatial arrangement boundary of each passenger with different stature sizes, but also can add design variables which can be changed in a large range, and the parameterized human body contour prediction model is not limited by the arrangement parameters of the traditional cabin control mechanism, so that a large amount of data can be integrated into abundant user samples, and different automatic driving and intelligent cabin environments can be flexibly combined.
2. And calculating the spatial arrangement boundary in the mechanism change process in real time, and further performing real-time collision and interference monitoring in the mechanism dynamic adjustment process of the intelligent cabin.
3. The method can realize high-efficiency, low-cost and high-precision arrangement boundary calculation through a computer algorithm, and control variable parameters to drive human body contour parameters by using parameters obtained by regression analysis, so that calculation iteration and optimization target limitation are not needed, the calculation cost is obviously reduced, and the method is more suitable for real-time collision and interference monitoring and intelligent control assistance.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. Like elements or portions are generally identified by like reference numerals throughout the several figures. In the drawings, elements or portions thereof are not necessarily drawn to scale.
FIG. 1 is a flow chart of a method of constructing an occupant parameterized human contour prediction model in accordance with the present invention;
FIG. 2 is a schematic diagram of human contour extraction;
FIG. 3 is a diagram of a human contour locating parameter;
in the figure: trunk 1, thigh 2, shank 3, foot 4, big arm 5, forearm 6, hand 7, head 8.
Detailed Description
For a clear and complete description of the technical scheme and the specific working process thereof, the following specific embodiments of the invention are provided with reference to the accompanying drawings in the specification:
example 1: referring to fig. 1, a flowchart of a method for constructing an occupant parameterized human body contour prediction model according to the present embodiment is shown, and the method specifically includes the following steps:
step 1: and selecting the control variables and the user samples to obtain a series of occupant human body contour data sample sets under the specific control variable distribution.
Wherein the control variables are divided into a control variable related to the size of the human body and a control variable related to the cabin arrangement, and the control variable related to the size of the human body is determined as the height, weight and sitting height of the passenger; the control variables related to the cabin layout need to be selected by themselves according to the design targets, which enables significant changes in the passenger layout boundaries, and cabin mechanism adjustment and control parameters as key design indicators, such as control parameters related to the spatial position and attitude adjustment of the seat, and the like.
The source of the user sample can be determined according to design requirements, such as bench test, agent model, benchmarking data and the like. The number of user samples is sufficient, the range of the target control variable can be covered within a certain confidence, and the selection of the samples can be simplified by an optimization method such as orthogonal design when necessary. The user samples are oriented to and cover the designed target groups, and the human factor factors such as gender, age, race and the like are considered, and proper sample control variable distribution is selected according to the design targets;
step 2: simplifying and extracting human body profile parameters of a user sample.
The key joint points for marking the side projection of the human body outline of the sample comprise 7 ankle joints, knee joints, hip joints, shoulder joints, elbow joints, wrist joints and neck joints, and the circle among the limb sections in the figure 2 is shown. The rectangular wire frame is used for completely containing all limb segments, including 8 parts of trunk, thigh, shank, foot, big arm, forearm, hand and head, and the rectangular wire frame in the figure 2 is used for completely containing all limb segments. The rectangular wire frame of each limb segment should be parallel to the skeleton line, i.e. the connection of the corresponding articulation point on the corresponding limb segment, e.g. the trunk wire frame should be parallel to the connection of the neck articulation point and the hip articulation point, the thigh wire frame should be parallel to the connection of the knee articulation point and the hip articulation point, etc.
The human body profile parameters to be extracted include a shaping size and a positioning size.
The shaping dimensions are the length and width of the rectangular wire frame of each limb segment. The total of 8 limb segments of the whole body is 16 shaped sizes. The positioning size is used for determining the position relation of the key joint point relative to the wire frame to which the key joint point belongs. According to anatomical features, the partial positional relationship can be simplified as follows: the ankle joint and the knee joint are positioned on the symmetry axis of the shank wire frame; the shoulder joint and the elbow joint are positioned on the symmetry axis of the large arm wire frame; the elbow and wrist joints lie on the symmetry axis of the forearm wireframe, see symmetry notation in fig. 2. After simplification, 15 positioning sizes are needed, see fig. 3. In addition, the included angle between each limb segment and the horizontal plane is 8, and the distance between each key articulation point is 5, including trunk length, thigh length, shank length, big arm length and small arm length. After the human body contour parameters are extracted, the wire frame of the human body contour can be uniquely determined.
Step 3: regression analysis of human body contour parameters.
The shaping parameters of the human body outline are the characteristic dimensions of the human body, are only related to human factors and are irrelevant to external factors, and the control variables are the height, weight and sitting height of the passengers. Through relevant researches and experiments, the 3 control variables of height, weight and sitting height can carry out stepping linear regression on 16 shaping sizes, and the explanation degree is high enough. Finally, by means of step linear regression, the height, weight and sitting height of the passengers are used as control variable parameters, and a parameter equation for predicting 16 shaping sizes can be obtained.
It should be noted that the shape-stabilized dimensions of the body contour are not related to external factors, and thus may be equally distributed in different designs. Therefore, a special human contour shaping size database can be established, user samples can be specially enriched for the database, a more accurate shaping size parameterized prediction model can be obtained, and the model can be commonly used among different products.
The positioning parameters of the human body outline are related to human factor and external factor, and all control variables are included in the parameter equation. And selecting proper control variables and regression methods according to mathematical relations among variables, and finally obtaining parameter equations for predicting 28 positioning sizes.
Step 4: and establishing a human body contour parameterized model.
Initializing the rectangular wire frame outlines of 8 limb segments and 7 key nodes, and carrying out initial size constraint on all the shaping sizes and positioning sizes according to actual physical meanings. Setting a series of control parameters such as the height, the weight, the sitting height and control variables related to cabin arrangement of an occupant, taking the shaping size and the positioning size of initial constraint as controlled parameters, and correlating the controlled parameters with the control parameters according to a parameter equation obtained by regression analysis to obtain a parameterized occupant human body contour prediction model, thereby realizing the real-time change of driving the human body contour prediction model by utilizing various different control parameters.
Step 5: checking and correcting the prediction model.
The prediction model can be regarded as a proxy model, so that the check of the prediction model mainly has two aspects, namely, the comparison check of the prediction result and the original sample data and the comparison check of the prediction result and the actual contour. The comparison and check of the prediction model result and the original sample data can be embodied on regression significance and regression residual, the original sample data can be randomly selected under different control variables, the control variables are input into the prediction model, the obtained prediction result and the original sample are compared, and whether the relative error is in an acceptable degree is checked. The comparison and check of the prediction result and the actual contour can design a verification experiment, a control variable far away from a user sample is selected for experiment, the prediction result under the same corresponding control parameter is compared with the experiment result, and whether the relative error is in an acceptable degree is checked.
If the checking result is not ideal, the checking deviation or the regression residual error of the characteristic test points or the characteristic samples can be analyzed, a correction term is added into a parameter equation of the prediction model, and compensation correction is carried out on the prediction model according to the checking deviation and the regression residual error.
Example 2: the existing intelligent automobile seat with a certain determined target user has the function of self-adaptively adjusting the cushion angle and the backrest angle, and the embodiment constructs an occupant parameterized human body contour prediction model for the intelligent automobile seat, and specifically comprises the following steps:
step 1: selecting control variables and user samples to obtain a series of occupant human body contour data sample sets under specific control variable distribution;
wherein the control variables are divided into a control variable related to the size of the human body and a control variable related to the cabin arrangement, and the control variable related to the size of the human body is determined as the height, weight and sitting height of the passenger; the control variables related to cabin layout are the cushion angle and the back angle; which significantly changes the occupant placement boundary and serves as a key design parameter for seat mechanism adjustment and control.
User samples were obtained from 96 test cabin environment simulation bench tests. In order to simplify the test, the control independent variables related to the cabin layout are selected from four values of the cushion angle 0 °, 5 °, 15 °, 20 ° and the back angle 0 °, 10 °, 20 °, 30 ° by orthogonal design. According to the distribution requirement of the target user group, the ratio of men to women in 96 tested is 2:1, the ages are distributed between 20 and 50 years, and the heights and weights are subject to normal distribution; and after the cabin environment simulation bench test is completed, acquiring a passenger human body contour data sample set.
Step 2: simplifying and extracting human body profile parameters of a user sample;
the key joints for marking the side projection of the human body outline of the sample comprise 7 ankle joints, knee joints, hip joints, shoulder joints, elbow joints, wrist joints and neck joints, and the circles among the limb sections in the figure 2 are shown. The rectangular wire frame is utilized to fully contain each limb segment, including 8 of trunk, thigh, shank, foot, big arm, forearm, hand, head, see rectangular wire frame in fig. 2. The rectangular wire frame of each limb segment should be parallel to the skeleton line, i.e. the connection of the corresponding articulation point on the corresponding limb segment, e.g. the trunk wire frame should be parallel to the connection of the neck articulation point and the hip articulation point, the thigh wire frame should be parallel to the connection of the knee articulation point and the hip articulation point, etc.
The human body profile parameters to be extracted include a shaping size and a positioning size.
The shaping dimensions are the length and width of the rectangular wire frame of each limb segment. The total of 8 limb segments of the whole body is 16 shaped sizes. The positioning size is used for determining the position relation of the key joint point relative to the wire frame to which the key joint point belongs. The following simplification is made according to anatomical features: the ankle joint and the knee joint are positioned on the symmetry axis of the shank wire frame; the shoulder joint and the elbow joint are positioned on the symmetry axis of the large arm wire frame; the elbow and wrist joints lie on the symmetry axis of the forearm wireframe, see symmetry notation in fig. 2. After simplification, 15 positioning sizes are needed, see fig. 3. In addition, the included angle between each limb segment and the horizontal plane is 8, and the distance between each key articulation point is 5, including trunk length, thigh length, shank length, big arm length and small arm length. After the human body contour parameters are extracted, the wire frame of the human body contour is uniquely determined.
Taking a sample as an example, the shaping size is as follows: trunk frame length 738 width 244, thigh frame length 613 width 169, shank frame length 548 width 137, foot frame length 271 width 106, arm frame length 405 width 122, arm frame length 371 width 103, head frame length 335 width 233; the shaping size is as follows: distances 45 and 49 from the cervical joint to the lower and posterior head boundaries; distances 67, 71 and 38 of the shoulder joints to the upper torso boundary, the front torso boundary and the rear arm boundary; distance 32 of elbow joint to forearm posterior border; distances 12 and 25 from the wrist joint to the posterior hand boundary and the upper hand boundary; distances 64 and 50 of the hip joint to the lower torso boundary and the anterior torso boundary; distances 48, 43, and 46 from the knee joint to the anterior thigh boundary, the lower thigh boundary, and the posterior calf boundary; distances 32 and 38 from the ankle to the upper foot step boundary and the rear foot boundary; torso angle 24, thigh angle 16, shank angle 38, foot angle 33, thigh angle 24, forearm angle 34, hand angle 33, head 0; trunk length 573, thigh length 438, calf length 440, big arm length 298, and small arm length 276. (the unit of the above-mentioned un-injected length is mm and the unit of the angle is degree.)
Step 3: regression analysis of human body profile parameters;
the shaping parameters of the human body outline are the characteristic dimensions of the human body, are only related to human factors and are irrelevant to external factors, and the control variables are the height, weight and sitting height of the passengers. Through relevant researches and experiments, the 3 control variables of height, weight and sitting height can carry out stepping linear regression on 16 shaping sizes, and the explanation degree is high enough. Finally, by means of step linear regression, the height, weight and sitting height of the passengers are used as control variable parameters, and a parameter equation for predicting 16 shaping sizes is obtained.
Taking the length of the trunk wire frame as an example in the embodiment, the parameter equation is as follows: trunk frame length = -0.024 x height +0.917 x weight +0.697 x sitting height +23.943; taking the trunk wire frame width as an example in the embodiment, the parameter equation is as follows: trunk frame width = -0.070 x height +1.380 x weight +0.090 x sitting height +177.896. (the above parameters are all values taken when the length unit is mm and the mass unit is kg)
The positioning parameters of the human body outline are related to human factor and external factor, and the parameter equation comprises all control variables of height, weight, sitting height, cushion angle and backrest angle. And carrying out regression analysis according to the mathematical relationship among the variables to finally obtain parameter equations for predicting 28 positioning sizes.
Taking the trunk angle as an example in the embodiment, the parameter equation is as follows: torso angle = -0.036 x height +0.062 x weight +0.074 x sitting height +0.423 x cushion angle +0.516 x back angle +5.458; taking the leg angle as an example in the embodiment, the parameter equation is as follows: shank angle = 0.030 x height +0.106 x weight-0.095 x sitting height +0.373 x cushion angle +0.057 x back angle +57.280. (the parameters are all values obtained when the length unit is mm, the angle unit is DEG and the mass unit is kg)
Step 4: establishing a human body contour parameterized model;
initializing the rectangular wire frame outlines of 8 limb segments and 7 key nodes, and carrying out initial size constraint on all the shaping sizes and positioning sizes according to actual physical meanings. Setting 5 control parameters of the height, the weight, the sitting height, the cushion angle and the backrest angle of the passenger, taking the shaping size and the positioning size of the initial constraint as controlled parameters, and correlating the controlled parameters with the control parameters according to a parameter equation obtained by regression analysis to obtain a parameterized passenger human body contour prediction model, thereby realizing the real-time change of driving the human body contour prediction model by utilizing various different control parameters.
Step 5: checking and correcting a prediction model;
and comparing and checking the prediction model result with the original sample data, and checking through regression significance and regression residual errors. Taking the length of the trunk wire frame as an example in the embodiment, the significance of the three control variables is less than 0.005, the adjustable coefficient is 0.997, and the error is quite small. The comparison check of the prediction result and the actual contour designs 10 groups of verification experiments, all selects control variables different from the user samples, compares the prediction result and the experimental result under the same corresponding control parameters, and has the relative error of each group smaller than 0.06, so that the seat can meet the design requirement without further correcting the prediction model.
In summary, the method for constructing the passenger parameterized human body contour prediction model according to the above embodiment selects parameterized design to construct the passenger human body contour model, and has the characteristics of real-time variation of the model along with parameters of control variables, capability of covering target passenger groups with different human body sizes which continuously vary, capability of increasing and decreasing the control variables to obtain prediction models under the control of different variables, and the like. Therefore, the parameterized human body contour prediction model is not only suitable for obtaining the spatial arrangement boundary of each passenger with different stature sizes, but also can add design variables which can be changed in a large range, and the parameterized human body contour prediction model is not limited by the arrangement parameters of the traditional cabin control mechanism, so that a large amount of data can be integrated into abundant user samples, and different automatic driving and intelligent cabin environments can be flexibly combined. And calculating the spatial arrangement boundary in the mechanism change process in real time, and further performing real-time collision and interference monitoring in the mechanism dynamic adjustment process of the intelligent cabin. The method can realize high-efficiency, low-cost and high-precision arrangement boundary calculation through a computer algorithm, and control variable parameters to drive human body contour parameters by using parameters obtained by regression analysis, so that calculation iteration and optimization target limitation are not needed, the calculation cost is obviously reduced, and the method is more suitable for real-time collision and interference monitoring and intelligent control assistance.
In addition, the specific features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various possible combinations are not described further.
Moreover, any combination of the various embodiments of the invention can be made without departing from the spirit of the invention, which should also be considered as disclosed herein.
Claims (9)
1. The method for constructing the passenger parameterized human body contour prediction model is characterized by comprising the following steps of:
s1, selecting control variables and user samples to obtain a series of passenger human body contour data sample sets under specific control variable distribution;
s2, simplifying and extracting human body profile parameters of a user sample;
the human body profile parameters comprise a shaping size and a positioning size; the shaping dimension is used for determining the size of the outline of each limb segment; the positioning size is used for determining the position relation of each key joint point relative to the wire frame to which the key joint point belongs;
s3, carrying out regression analysis on human body contour parameters;
s31, regression of the shaping parameters of the human body contour;
taking the height, weight and sitting height of the passenger as control variable parameters to obtain a parameter equation for predicting the shaping size;
s32, regression of human body contour positioning parameters;
selecting proper control variables and a regression method to obtain a parameter equation of a predicted positioning size;
s4, establishing a human body contour parameterized model;
s5, checking and correcting the prediction model.
2. A method of constructing a parameterized occupant human body contour prediction model according to claim 1, wherein in step S1, the control variables include control variables related to human body size and control variables related to cabin layout;
the user sample is derived from bench test, agent model and benchmarking data.
3. A method of constructing an occupant parameterized body contour prediction model according to claim 2, wherein said body size related control variables include occupant height, weight, sitting height;
the control variables related to cabin arrangement are selected according to design targets, so that passenger arrangement boundaries are ensured to be changed, and cabin mechanism adjustment and control parameters are key design indexes.
4. The method for constructing a passenger parameterized body contour prediction model according to claim 1, wherein in step S2, the simplifying and extracting user sample body contour parameters specifically includes the following:
firstly, marking key joint points of side projection of a human body contour of a sample, wherein the key joint points comprise 7 ankle joints, knee joints, hip joints, shoulder joints, elbow joints, wrist joints and neck joints;
then, marking each limb segment including trunk, thigh, shank, foot, big arm, small arm, hand and head by using rectangular wire frame;
finally, marking human body contour parameters;
wherein the shaping dimensions are the length and width of a rectangular wireframe of each limb segment, and the positioning dimensions include: 15 boundary positioning dimensions, 8 included angle dimensions, and 5 pitch dimensions for each critical node.
5. The method of constructing an occupant parameterized human contour prediction model of claim 4, wherein said 15 boundary-positioning dimensions include distances of a neck joint to a lower head boundary and a rear head boundary; the distance from the shoulder joint to the upper torso boundary, the front torso boundary, and the rear arm boundary; the distance of the elbow joint to the posterior border of the forearm; the distance from the wrist joint to the back hand boundary and the upper hand boundary; the distance of the hip joint to the lower torso boundary and the front torso boundary; the distance of the knee joint to the anterior thigh boundary, the lower thigh boundary, and the posterior calf boundary; the distance from the ankle joint to the upper boundary of the foot and the rear boundary of the foot step; the included angle size comprises included angles between each limb segment and the horizontal plane; the spacing dimension of each key articulation point comprises the spacing of each key articulation point, including trunk length, thigh length, shank length, big arm length and small arm length.
6. A method of constructing a parameterized occupant human contour prediction model according to claim 4, wherein said rectangular wire frame shall fully encompass each limb segment, said rectangular wire frame of each limb segment shall be parallel to the skeleton line, i.e. the connection of the corresponding critical joints on the respective limb segment.
7. The method of constructing an occupant parameterized human contour prediction model of claim 6, wherein the ankle joint and the knee joint are located on the symmetry axis of the calf rectangular wireframe; the shoulder joint and the elbow joint are positioned on the symmetry axis of the rectangular wire frame of the big arm; the elbow joint and the wrist joint are located on the symmetry axis of the rectangular wire frame of the forearm.
8. The method of constructing an occupant parameterized human contour prediction model of claim 1, wherein step S4 specifically comprises the following:
firstly, setting control parameters including the height, weight, sitting height and cabin arrangement related control variables of an occupant, then taking the initial constraint shaping size and positioning size as controlled parameters, and associating the controlled parameters with the control parameters according to a parameter equation obtained by regression analysis in the step S3 to obtain a parameterized occupant human body contour prediction model.
9. The method of constructing a prediction model of occupant parameterized human contours of claim 1, wherein in step S5, the checking of the prediction model comprises: comparing and checking the prediction result with the original sample data and comparing and checking the prediction result with the actual contour;
the comparison and check of the prediction result and the original sample data comprises the following steps: randomly selecting original sample data under different control variables, inputting the control variables into a prediction model, comparing the obtained prediction result with the original sample, and checking whether the relative error is in an acceptable degree;
the comparison and check of the prediction result and the actual contour comprises the following steps: selecting a control variable with a value different from a user sample to perform an experiment, comparing a predicted result under the same corresponding control parameter with an experimental result, and checking whether the relative error is in an acceptable degree;
if the checking result is not ideal, analyzing checking deviation or regression residual of the characteristic test points or the characteristic samples, adding correction terms into a parameter equation of the prediction model, and compensating and correcting the prediction model according to the checking deviation and the regression residual.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311217113.2A CN116956379B (en) | 2023-09-20 | 2023-09-20 | Method for constructing passenger parameterized human body contour prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311217113.2A CN116956379B (en) | 2023-09-20 | 2023-09-20 | Method for constructing passenger parameterized human body contour prediction model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116956379A true CN116956379A (en) | 2023-10-27 |
| CN116956379B CN116956379B (en) | 2023-12-08 |
Family
ID=88447730
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311217113.2A Active CN116956379B (en) | 2023-09-20 | 2023-09-20 | Method for constructing passenger parameterized human body contour prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116956379B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103221257A (en) * | 2010-10-07 | 2013-07-24 | 佛吉亚汽车座椅有限责任公司 | System, methodologies, and components acquiring, analyzing, and using occupant body specifications for improved seating structures and environment configuration |
| CN106853784A (en) * | 2015-12-09 | 2017-06-16 | 李尔公司 | Pelvis and rumpbone capsule component |
| CN113942467A (en) * | 2020-07-15 | 2022-01-18 | 采埃孚汽车科技(上海)有限公司 | Occupant protection method, system and computer readable medium |
| CN114714995A (en) * | 2022-05-09 | 2022-07-08 | 上汽大众汽车有限公司 | Vehicle cabin matching and adjusting method based on human body parametric model |
| CN115099087A (en) * | 2022-06-16 | 2022-09-23 | 吉林大学 | Method for predicting neck comfort of driver and passenger through finite element method |
| CN115782709A (en) * | 2023-01-31 | 2023-03-14 | 张家港市宏博机械有限公司 | Automatic change seat adjustment system |
| CN115923607A (en) * | 2023-01-06 | 2023-04-07 | 东风柳州汽车有限公司 | Seat parameter adjusting method, device, storage medium and device |
-
2023
- 2023-09-20 CN CN202311217113.2A patent/CN116956379B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103221257A (en) * | 2010-10-07 | 2013-07-24 | 佛吉亚汽车座椅有限责任公司 | System, methodologies, and components acquiring, analyzing, and using occupant body specifications for improved seating structures and environment configuration |
| CN106853784A (en) * | 2015-12-09 | 2017-06-16 | 李尔公司 | Pelvis and rumpbone capsule component |
| CN113942467A (en) * | 2020-07-15 | 2022-01-18 | 采埃孚汽车科技(上海)有限公司 | Occupant protection method, system and computer readable medium |
| CN114714995A (en) * | 2022-05-09 | 2022-07-08 | 上汽大众汽车有限公司 | Vehicle cabin matching and adjusting method based on human body parametric model |
| CN115099087A (en) * | 2022-06-16 | 2022-09-23 | 吉林大学 | Method for predicting neck comfort of driver and passenger through finite element method |
| CN115923607A (en) * | 2023-01-06 | 2023-04-07 | 东风柳州汽车有限公司 | Seat parameter adjusting method, device, storage medium and device |
| CN115782709A (en) * | 2023-01-31 | 2023-03-14 | 张家港市宏博机械有限公司 | Automatic change seat adjustment system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116956379B (en) | 2023-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Reed et al. | A statistical method for predicting automobile driving posture | |
| Happee et al. | A mathematical human body model for frontal and rearward seated automotive impact loading | |
| Du et al. | Simulation of the interaction between driver and seat | |
| CN111666634A (en) | Method for establishing ellipse of driver's eye based on human motion simulation | |
| Ozsoy et al. | Simulated effect of driver and vehicle interaction on vehicle interior layout | |
| CN116956379B (en) | Method for constructing passenger parameterized human body contour prediction model | |
| CN106202783B (en) | Driving seat design method | |
| Reed et al. | Automobile occupant posture prediction for use with human models | |
| CN117171869A (en) | Ejection seat static comfort simulation layout calculation method | |
| CN114714995B (en) | Vehicle cabin matching and adjusting method based on human body parametric model | |
| Reed et al. | Comparison of methods for predicting automobile driver posture | |
| CN115099087A (en) | Method for predicting neck comfort of driver and passenger through finite element method | |
| CN111898205B (en) | RBF neural network-based human-machine performance perception evaluation prediction method and system | |
| Reed | Development of a new eyellipse and seating accommodation model for trucks and buses | |
| Summerskill et al. | Digital human modeling in the user-centered design process | |
| CN109871555B (en) | Simplified modeling method for automobile instrument board | |
| Chung et al. | Three‐dimensional analysis of a driver‐passenger vehicle interface | |
| Du et al. | Research and validation of intelligent control strategy for car seat comfort | |
| Seidl | 14.3 The RAMSIS and ANTHROPOS Human Simulation Tools | |
| Reed et al. | Methods for laboratory investigation of truck and bus driver postures | |
| Du Yihang et al. | Movement Description Method and Application for Dance of Chinese Traditional Drama | |
| Hanumantha et al. | Objective Seat Comfort Assessment using Finite Element Human Body Model Simulations | |
| Asadi et al. | Automation of ATD positioning for crash analysis simulation | |
| Reynolds et al. | Validation of virtual driver model for design of automotive seating packages | |
| Reed et al. | Improved ATD positioning procedures |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |