DE102013001365A1 - Method for visual analysis of vehicle design for vehicle mirror and for radar system, involves mapping boundary contour lines of computer-aided design vehicle data record and of modified computer-aided design-vehicle data record - Google Patents

Abstract

The method involves combining a computer-aided design (CAD) vehicle data record and a record of a human model in a CAD system, where a radiated view is generated with the CAD system from the ocular points of the human model. The boundary contour lines of the CAD vehicle data record and boundary contour lines of a modified CAD-vehicle data record are mapped as visual reference lines on a screen by using the radiated view, where the visual reference line is provided to define a viewing surface. The viewing surface and visual reference lines are quantified.

Description

The driver's view from a vehicle represents an essential safety-relevant component in road traffic. For the first time, a three-dimensional CAD visual model is presented, which enables a quantification of the visual characteristics in the early design process of a vehicle. In the process, boundary contours are specifically defined with the aid of the 3D-CAD visual model and the effects on the field of view are visualized and the optimization potential for the view is determined if the limit contours of the components are changed by constructive optimizations.

State of the art:

In the vehicle construction is often the old SAE J941 from 1965 and the eye ellipse described therein is used to determine the eye points in the vehicle. This was determined in the United States on vehicles with a continuous front seat and without seat height and steering wheel adjustment. In order to adapt it to reality, the SAE J941 was completely redesigned and based on the seat reference point. But even this definition does not reflect the wide variation possibilities of the various seating positions, which allows a driver's seat today in combination with the steering wheel adjustment.

The invention is based on a method for visual analysis, as developed in the dissertation by Mr. Hudelmaier. Hudelmaier, Jörg: Visual analysis in the car: taking into account movement and individual body characteristics. Munich, Technical University, Department of Ergonomics, Dissertation, 2002. Hudelmaier has combined the CAD data sets of vehicles with the datasets of the human model RAMSIS.

The human model RAMSIS (Computer-aided Anthropological Mathematical System for Inmate Simulation) was launched by the FAT (Automotive Technology Research Group) with the involvement of representatives of the automobile companies AUDI, BMW, Ford, Mercedes-Benz, Opel, Porsche and Volkswagen as well as the seat manufacturer Keiper Recaro and Naue , The company tecmath in Kaiserslautern (today tecmath AG) took over the development. The necessary ergonomic data was created by the IfE (engineering office for ergonomics), which in turn is in close cooperation with the Department of Ergonomics (LfE) of the Technical University of Munich, and the Catholic University of Eichstätt.

Through the combination of CAD records of the vehicle and the records of RAMSIS in a CAD system, Hudelmaier was able to simulate the head movements of the human model and determine whether the statutory minimum visibility when looking into the rearview mirror of the vehicle can be maintained by the vehicle design.

This visual analysis tool has been further developed to product maturity and is operated under the name CAD tool RAMSIS by Human Solutions GmbH as Ergonomics Software. In the vehicle development according to the prior art CAD design data sets z. From CATIA V5 with three-dimensional virtual systems, e.g. B. Cave Automatic Virtual Environment and combined with the RAMSIS human model to perform visual analysis and verify whether the statutory provisions and the requirements of technical standards can be met with the modeled vehicle design.

However, the visibility can only be assessed subjectively by subjects. A comparison between the individual persons provable by reference values is also not possible here. Nor is it yet possible to quantify which optimization potential has a constructive change in the CAD model of the vehicle to improve visibility. So far, it has not been possible to quantify the combination of individual component changes and the possible individual optimization potentials and to supply them in their entirety or individually to a valuation. Quantified information which constructive change would bring which visual advantage was therefore not available in the early phase of a design and design process.

It is therefore an object of the invention, with a further development of the visual model and with a quantitative analysis, to make the optimization potential of design changes accessible to the CAD vehicle data record.

The solution succeeds mainly through the extension of the viewing model with visible surfaces and visual reference lines, which are generated on a projection surface by means of a 3D-CAD visual model. The 3D-CAD model generates visual beams. Also from the CAD model are the boundary contour lines derived from the vehicle data set that limits the field of view of the human model, which is also integrated in the CAD system. Eye position of the human model and boundary contour lines of the vehicle design determine the edge visual rays of the field of view, which limit the field of view on the projection screen. These boundary lines are referred to below in the description of the embodiments as visual reference lines.

This has the following advantages:
Depending on the CAD model of the vehicle, clear and quantifiable visible surfaces are generated, the area of which is determined, for example. Changes in the design datasets of the vehicle model, which also affect the boundary contour lines of the field of view, lead to changes in the visible surfaces. The changes can be z. B. be determined quantitatively by changing the surface area. By quantitatively comparing the visible surfaces of various constructive embodiments of the boundary contour lines in the vehicle data set, a quantitative measure can immediately be indicated which enlargement or reduction is achieved with a modified boundary contour line. Conversely, it can also be determined how components or installation spaces which bound the field of view have to be changed in their dimensions in order to achieve a specific enlargement of the visible surfaces.

• – Form oder Flächeninhalt der Sichtfläche
• – Ausprägung de Sichtfeldes nach oben und unten
• – Ausprägung des Sichtfeldes nach rechts und links
• – Homogenität des Sichtfeldrandes

In addition to the quantitative evaluation, the visible surfaces also allow a qualitative assessment of the future vehicle design. These are in particular the following properties of the field of view:

• - Shape or area of the visible surface
• - Expression of field of view up and down
• - Characteristic of the field of view to the right and left
• - Homogeneity of the field of view

In a further embodiment of the viewing model, the known eye ellipse from the SAE J941 is replaced by an improved eye ellipse. The improved eye ellipse is created by combining different RAMSIS human models with the seat adjustment field of the vehicle design, which is taken from the CAD vehicle data set. In the RAMSIS model, 3 people make up DIN 33402 the anthropometric body measurements of the 5% percentile of the woman, the 50% percentile, and the 95% percentile. These three person sizes are combined with 4 positions from the seat adjustment field of the vehicle to represent the possible extreme visibility conditions:
It is combined:
Rear seat back position, 95% percent for rearmost eye point
Sitting extreme front position with 5% percentile for the front eye point
Sitting position mid-top with 50% percentile for the highest eye point
Mid-bottom seating position with 50% percentile for the deepest eye point.

If necessary, an ellipse can be spanned by the 4 eye points, with the help of which the position further eye points for the determination of the visible surfaces with the visual model are determined.

This has the following advantages:
The eye points are the starting points for the visual beams of the visual model and, together with the boundary contour lines, determine the visual reference lines of the visible surfaces. This can also simulate the visibility that results for the different anthropometric body measurements in the vehicle. In particular, it can be determined to what extent the Sitzverstellfeld provides sufficient adjustment options to grant the selected percentiles a sufficient view.

Another advantage here is that changing the Sitzverstellfeldes results in a change in the eye points, which leads to a change in the field of view, which is again a quantitative and qualitative assessment accessible. It can thus be given a quantitative measure of what effects a change in Sitzverstellfeldes for the different percentiles in the field of view. In one possible embodiment of the visual model, the projection surface is modeled as a spherical surface.

In one possible embodiment, the projection surface is simulated as a virtual projection surface in the CAD model.

In another embodiment, the projection surface is a real projection surface onto which the visible surfaces are projected by means of a projector.

Further details of the invention are disclosed in the following description of the various embodiments.

Showing:

1 The generation of eye points from the real people groups and the seat adjustment field

2 The generation and mapping of the visual reference lines from the visual model

3 An overall presentation of the procedure

4 Examples of visual reference lines

5 An example of a visible surface

6 Averaged and visible on the screen visual rays

7 The naming of the most important visual reference lines

8th An application of the invention to the field of view in the rearview mirror

9 An application of the invention to a radar system z. Eg Distronic

In order to be able to influence the view in the early development process, a model is needed that can analyze and evaluate the visual conditions already in the early concept and design phase, and point out optimization potentials. It is important to ensure that the model allows an assessment of the visual situation for the largest possible inmate collective. In addition, the model is to enable comparison with other vehicles, such as predecessor models and competing vehicles, by generating standardized measures. Ideally, specifications for the construction of additional or new vehicles can be derived from such a model. The model should be kept as simple as possible in order to be able to use it quickly if necessary. It should also be able to apply not only to pure CAD data but also to scan data such as those available in the initial development phase of the design and concept process. A later comparison of the model data with hardware testing procedures should also be possible.

The invention provides a three-dimensional holistic design model for the 360 ° description of the field of vision of a vehicle driver or the other vehicle occupants. In order to represent, assess, analyze and optimize visibility, such a model is urgently needed in construction. The visual model was created using the CAD program CATIA V5 and RAMSIS, the use of other CAD software is just as possible.

1 : The position of the eye points for the determination of visibility depends essentially on the body dimensions and the respective seating position in the vehicle. Therefore, a completely different approach for determining the eye points is chosen for this model. This is largely based on the anthropometric body measurements, the respective Sitzverstellfeld and the seat position relevant components such as steering wheel, pedal system, shifter and vehicle interior dimensions. The ergonomic and human model RAMSIS offers the ideal environment to position people's groups ergonomically in the driver's cab and to derive the positions of the eye points required for the model. In order not to have to check all person sizes, the application of so-called percentiles in the ergonomic design is widespread. In order to make it possible to compare the view of individual persons, the anthropometric body measurements are reproduced in this model DIN 33402 used as a basis. These were published in 2005. It is based on the dimensions of the German population between 18 and 65 years. In order to represent 90% of the possible driver collective in this visual model, therefore, the 5% percentile of the woman (short: 5% woman), the 50% percentile (short: 50% man) and the 95% percentile (short: 95% man) of the man of the age group 18-65 for this purpose.

All body measurements after DIN 33402 will now be implemented in the RAMSIS program for the 5% woman, 50% man and 95% man, and positioned in the vehicle using the RAMSIS seat posture model depending on the surrounding components and seat adjustment field. The 50% man is placed in the highest and lowest possible ergonomic position. From the specific seat adjustment field of a Vehicle and the respectively placed persons results in a relation to the SAE J941 improved eye ellipse, which also includes the seat adjustment field from the CAD vehicle data record. This results in the advantages already mentioned at the outset.

• – 5% Frau den vordersten Augpunkt
• – 50% Mann hoch den höchsten Augpunkt
• – 50% Mann tief den tiefsten Augpunkt
• – 95% Mann den hintersten Augpunkt

The 4 people represent the respective extreme seating positions and their respective visibility:

• - 5% woman the foremost eye point
• - 50% man high the highest eye point
• - 50% man deep the deepest eye point
• - 95% man the rearmost point

In order not to have to perform the visual model for the right and left eye of the human, the convention is to use the middle eye point, an eye point on the half line connecting the right and left eye. The small distance of less than 30 mm of the center eye point to the right / left eye point leads in the relation of the distances to the components to a very small deviation from the reality. This is gladly accepted by halving the effort. In the software tool RAMSIS the middle eye point can be removed very easily.

2 : The respective boundary contour lines of the individual components must now be determined individually for the eye point of each percentile. Border contour lines are the extreme line course of a component with respect to the respective center eye point. It is important to note that the boundary contour line of one percentile usually does not agree with that of the other percentile. The determined boundary contour lines are then superimposed on the eye point. The result is the first visual beam up to the boundary contour line, which must be extended beyond this. This process must be carried out as often as possible until all components are obscured by the visible beams and thus all visual field boundaries are mapped.

After all the relevant visual rays have been generated, the visible component surfaces can be recognized by their different colors. As a view restricting components, in this model, all components are defined, which define the outer boundary of the field of view.

In his dissertation, Mr. Hudelmaier mentions a surface analysis with which visual rays can be blended with "arbitrary surfaces or cylinders". This was not specified by him. This approach of surface analysis is taken up in the 3D-CAD visual model. However, the visual beams are not projected onto a dimensionally undefined cylinder or surface which, in retrospect, does not allow for comparable statements between the percentiles or other vehicle models.

3 : For the surface analysis in the 3D-CAD visual model, a percentile-specific reference sphere is defined at a distance of 5 m by the respective mean eye point as the projection surface. A sphere was chosen as the reference surface, as it allows a distortion-free analysis and quantification of the view. Each viewing beam covers exactly the same path between the middle eye point and the reference sphere. This can not be guaranteed by any surface, a cylinder or even the road as a reference surface. The radius of the reference sphere can also be chosen individually. However, a visible surface comparison is only possible with the same radius.

In the next step, all relevant visual rays are blended with the reference sphere. This creates so-called visual reference lines on the sphere. The visual reference lines now allow a targeted analysis, evaluation and optimization of the visual components. The analysis, evaluation and optimization steps will be further explained in the following chapters.

In order to be able to analyze, evaluate and optimize a field of vision of a vehicle driver, criteria are needed by which the visual characteristics can be quantified. In order to be able to describe the properties of the field of view, the first question to ask is: what is view? How can vision be described?

The dictionary defines the word view as follows:
"Access the view has to more or less distant objects."

If one takes this definition and transmits it to the field of vision of a vehicle, then it quickly becomes clear that the mere viewing of the visible surface is not sufficient. The viewing surface alone does not say anything about whether it's narrow, wide, tall or long, but the surface can always be the same. For this reason the criteria view up and down as well as to the right or left are introduced. These allow a statement about the field of view height and width.

Looking at the visual reference lines in 3 and the RAMSIS visual analysis in the 4 Thus, one finds that these criteria are also not yet sufficient to fully reflect the view. Because in terms of vision, the number of components in the field of view and their design contour lines also play an important role. If a field of vision edge is very restless and wavy, the field of view is smaller, but if it is calm and straight, the field of view is larger and more homogeneous. Therefore, another criterion is introduced the homogeneity of the field of view edge.

• – Sichtfläche
• – Ausprägung des Sichtfeldes nach oben und unten
• – Ausprägung des Sichtfeldes nach rechts und links
• – Homogenität des Sichtfeldrandes

In summary, there are now 4 criteria for describing the property view in the 3D-CAD visual model:

• - visible surface
• - Expression of the field of view up and down
• - Characteristic of the field of view to the right and left
• - Homogeneity of the field of view

In contrast to the minimum requirements, the evaluations in the 3D-CAD visual model are not locally limited to sections, but include the entire inner visual reference lines.

The quantitative determination of the individual visual properties will now be described. Care must be taken that all these steps must be performed separately for all 4 percentiles in order to be able to make a statement about the quality of the field of view.

To date, there is no reference vehicle that defines the ideal visibility conditions. Thus, it is not possible to specify the exact visual optimization potential of a luxury sedan. As an orientation for the construction of new models, for example, the viewing characteristics of a predecessor or competing model can serve.

In this work, an existing upper class sedan will be analyzed and possible optimization potentials will be shown. Therefore, the optimization percentage for this work is defined as follows:

This measure is given in percent.

5 : Visible surfaces are the areas enclosed by the inner visual reference lines on the reference sphere. The loss of visible area through the design contour lines of the components is referred to as the visible area loss. Visible surface loss is always specified and measured with respect to the nearest visual reference line.

For example, the visible area loss of the black coloring is indicated with respect to the next visual reference line of the A-pillar trim, and the visible area loss of the use of the instrument panel with respect to the visual reference line of the hood.

The quantitative description of the visible surfaces is easy to perform in the CAD program. In the first step, the inner visual reference line is blended with the reference sphere to create the viewing surface. The same is done by blending the other visual reference lines with the reference sphere to determine the visible area losses. The areas, or the size of the areas, can be read in a CAD program by a measuring function.

Far more complex is the determination of the expression of the field of view up and down, right and left and the description of the homogeneity of the field of view. So far, the field of view width, bounded to the front by the A-pillar, is determined and evaluated in defined sections. This method allows only local limited to the respective cutting plane a conclusion on the field of view width and height and allows no statement about the entire visual situation. In order to avoid targeted design measures in the check-cut levels, which serve only for certification purposes, an assessment is needed that includes the entire inner visual reference line.

• – Ist ein Bauteil der horizontalen Ebene zuzuordnen, so wird dessen innere Sichtreferenzlinie als untere oder obere innere Sichtreferenzlinie bezeichnet.
• – Ist das Bauteil hingegen der vertikalen Ebene zuzuordnen, so wird dessen innere Sichtreferenzlinie als rechte oder linke innere Sichtreferenzlinie bezeichnet.

In order to distinguish the manifestation of the field of view up and down as well as right and left, a differentiation into upper / lower / right / left inner visual reference line is needed. The following convention is taken:

• - If a component is to be assigned to the horizontal plane, its inner visual reference line is referred to as the lower or upper inner visual reference line.
• If, on the other hand, the component is to be assigned to the vertical plane, its inner visual reference line is referred to as the right or left inner visual reference line.

6 . 7 In order to allow a comparable, quantifiable optimization in the test section planes in the future, in a next step the upper / lower / right / left inner visual reference line is approximated by a close-meshed number of points. The distance between two points should not be greater than 35 mm, because then already the first components and contours, which are smaller, can fall out of the rating.

n:

Anzahl der Punkte

x, y, z:

Koordinaten der Punkte

In a further step, the coordinates of each individual point on the respective inner visual reference line are read out and the upper / lower / right / left average is determined from all point coordinates.

n:

Number of points

x, y, z:

Coordinates of the points

If the four mean values are connected to the respective middle eye points of the percentiles by a line, the averaged visual ray is spoken up / down / right / left.

Of course, the coordinates of the middle eye points of the percentiles are needed for the following calculations of the values up and down, or right and left.

The four averages for each percentile are used in the 3D CAD visual model as a reference for quantitative assessment. It is important to ensure that all recalculated values are only used in a percentile specific way.

The certification requirements of the approval guidelines specify the minimum field of view downwards as an angle with respect to the horizontal. This approach should also be applied here in relation to the averaged visual beam down. The angle between the averaged visual ray and the horizontal through the percentile-specific mean eye point can be given mathematically by the following formula.

Another characterizing measure, which should however only be used for visualization purposes, is the average impact point (in short: mA) of the averaged visual beam on the road. For this, the distance in z to the road with respect to the coordinate system is required, which is to be determined.

The coordinates of this point can be determined mathematically as follows:

The distance of the mean impact point to the road can be specified both with respect to the origin of the coordinates, the bumper in the middle of the vehicle or with respect to the center eye point.

Tabelle: Formeln zur Ermittlung des Abstands zum mittleren Auftreffpunkt auf die Straße

Table: formulas for determining the distance to the mean impact point on the road

So far, only the characteristics were presented down. For the consideration of the characteristic upwards, the angle calculation is sufficient. The viewing angle upwards is determined with the following formula.

It should be noted that the angle downwards is positive and the angle upwards is negative.

Characteristic of the field of view to the right and left

Also for the expression of the field of view to the right and left, the description of the angle to the vertical through the middle eye point is chosen as a quantitative measure. The angle between the respective averaged visual ray to the right and left and the vertical through the percentile specific mean eye point can be given mathematically by the following formula.

The indication of the angle to the left (driver side) takes place with a positive sign and that of the angle to the right (passenger side) with negative.

Homogeneity of the field of view

Homogeneity is the fourth criterion for describing the visual properties. It describes the course and gives a measure of the straightness of the field of view.

Homogeneity is indicated in the 3D-CAD visual model by the number of homogeneity violations. The less homogeneity violations exist, the more straightforward is the field of view edge, there is no homogeneity violation, so the distances of the individual points are in the homogeneity range.

n:

Anzahl der Punkte

x, y, z:

Koordinaten der Punkte

The homogeneity range describes the maximum permissible percentage deviation to the average point distance between adjacent points of the point cloud.

n:

Number of points

x, y, z:

Coordinates of the points

A homogeneity violation occurs when the amount of the distance of the respective relevant coordinate to the relevant coordinate of its directly adjacent point is greater than the permissible homogeneity range.

• – oben und unten: z-Koordinate
• – rechts und links: y-Koordinate

Relevant coordinate for homogeneity:

• - top and bottom: z-coordinate
• - right and left: y-coordinate

• – oben und unten: 10% des gemittelten Punktabstandes
• – rechts und links: 40% des gemittelten Punktabstandes

nicht überschreiten sollte.First trade fair trials have shown that the homogeneity area

• - above and below: 10% of the average point distance
• - right and left: 40% of the average point distance

should not exceed.

For the lower and upper field of view edge a permissible percentage deviation of 10% was determined, since here many components and their design contour lines meet and first subjective analysis by means of RAMSIS visual analysis resulted in a developer acceptable deviation from 2 mm to 20 mm grid spacing. For the range of homogeneity to the right and left, these analyzes gave a permissible deviation of 40%, since the lateral components reflect an upwardly opening funnel due to their inclination angle in the vehicle on the reference sphere. As a result, the difference of the y-coordinates deviates more from each other than it is in the lower / upper range of the case.

In order to be able to define the percentage deviation more accurately, a subjective test subject survey and consideration of many different vehicles would be necessary.

Thus, various aspects of the field of view or the visible surface generated on the reference sphere of a quantitative detection are accessible by the invention. For the individual aspects, the optimization potential for the transition from one CAD vehicle design to the other CAD vehicle design can then also be specified as an optimization percentage.

• – Die Größe des Flächeninhaltes
• – Die Ausprägung des Sichtfeldes nach oben und unten
• – Die Ausprägung des Sichtfeldes nach links und rechts
• – Die Homogenität der Sichtfeldränder

The aspects of the field of view are in particular:

• - The size of the area
• - The expression of the field of view up and down
• - The expression of the field of view to the left and right
• - The homogeneity of the visual field edges

The calculations can be carried out with common EDP programs. In particular, the spreadsheet program Microsoft Excel can be used.

8th : This visual optimization using the 3D-CAD visual model can be extended to the entire vehicle in a next step in order to be able to make a holistic statement about the visibility in motor vehicles. The mirror surfaces in the 3D-CAD visual model are also taken into account. This can then be the first time a statement and assessment of the entire situation of the driver in a vehicle to be taken.

9 : Camera, radar and sensor beams can also be projected onto the reference sphere and evaluated together with the sight and mirror beams. This makes it possible to detect and display dead angles. Likewise, it is possible to determine which blind spots are covered by the assistance systems, or where the positions of the sensors and cameras must be changed so that all blind spots of the vehicle are covered. The transition from the real point of view to the digital perspective can be captured and demonstrated in three dimensions. Redundant areas are also recognizable by this method.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited non-patent literature

• SAE J941 [0002]
• DIN 33402 [0012]
• DIN 33402 [0031]
• DIN 33402 [0032]
• SAE J941 [0032]

Claims (9)

A method for visual analysis of a vehicle construction in which: a CAD vehicle data set and a human model dataset are combined in a CAD system, characterized in that the CAD system generates visual beams starting from the human model's eye points with those boundary contour lines of a first CAD vehicle dataset and Bound contour lines of a modified CAD vehicle data set are displayed as visual reference lines on a projection surface, this visual reference line define a viewing surface; Visible surface and visual reference lines are quantified; and for the visual reference lines or the visible surface of the modified CAD vehicle data set an optimization potential compared to the first CAD vehicle data set is determined. A method according to claim 1, characterized in that the quantification or the optimization potential for the following properties of the visible surface are formed: - Form or area of the visible surface - Expression of the field of view down and up - Expression of the field of view to the right and left - Homogeneity of the field of view Method according to claim 1 or 2, characterized in that the coordinates of the eye points in the CAD vehicle data set are generated from the human model and the seat adjustment field of the seat adjustment. A method according to claim 3, characterized in that from the human model three person sizes, a 5% percentile, a 50% percentile and a 95% percentile are selected and these three person sizes are combined with the Sitzverstellfeld as follows: - sitting position at the rear, below with 95% Percentile for the rearmost point of the eye, - sitting position in front, top with 5% percentile for the foremost ocular point, - sitting position in the middle, top with 50% percentile for the highest eye point, - sitting position in the middle, bottom with 50% percentile for the deepest ocular point. Method according to one of claims 1 to 4, characterized in that the projection surface is a spherical surface. Method according to one of claims 1 to 5, characterized in that the projection surface is a virtual projection surface. Method according to one of claims 1 to 5, characterized in that the projection surface is a real projection surface. Method according to one of Claims 1 to 7 for vehicle mirrors: Method according to one of claims 1 to 7 for radar systems.

Images