User`s guide

Using CAD Models with the Simulink 3D Animation Product
5-33
Level of Detail Considerations
CAD models are usually parametric models that use proprietary object rendering
methods for use in various contexts. During VRML model export, the internal parametric
model of the assembly is tessellated. In this process, the model surface is divided into
triangular meshes, represented in VRML by the IndexedFaceSet nodes. During
tesselation, it is important to set the granularity of the mesh so that it is suitable for
further use. Modifying the polygon count afterwards would not only be very difficult,
but also not practical, as the resolution independent information of the object shape and
structure is lost and cannot be reconstructed based on the tessellated model.
For the effective rendering of moving parts, VRML models should be as simple as
possible. However, usually little, if any, visible model degradation is desired. It is often
just an issue of finding the appropriate compromise between these two requirements.
As there are significant performance differences among various computers and graphic
accelerators, there is no firm recommendation for the number of polygons or triangles
suitable for use with the Simulink 3D Animation product. To assess the model's
complexity, you can display the resulting VRML file in the Simulink 3D Animation
viewer and observe the viewer response to navigation. If you can navigate the virtual
world without any significant delays, the model is usually suitable for further work.
If you connect the virtual world to a Simulink model, you have access to more precise
measures of suitability, such as the number of frames rendered per second during
simulation.
Units Used in Exported Files
VRML length units are meters. To scale exported parts correctly in the virtual world,
export the parts using meters. If the exported objects are very small or very large, you
may want to create your virtual world in some other scale. In this case, you should export
the objects using units other than meters.
VRML viewers are made to measure using dimensions that are comparable to the
dimensions of people, to achieve the immersion effect of virtual reality. Viewers assume
that the author prepared the scene so that it can be walked through or examined by a
virtual visitor to the scene (sometimes called the Avatar), whose physical dimensions are
used in calculations for purposes like collision detection, near-object clipping, or terrain
following. You can customize avatar dimensions (and also other navigation-specific
parameters such as default navigation speed) using the NavigationInfo VRML node.
The Simulink 3D Animation viewer enables effective navigation in the virtual world,
including scaled scenes (e.g., inspecting miniature objects or visualizing a large-scale