I am building a model of NASA's Galileo spacecraft in POV. It has a
paraboloc antenna which was to open somewhat like an umbrella after the
spacecraft was launched. This did not happen because 3 of 18 ribs
remained stuck to the central mast. This seriously impacted (but did not
ruin) the mission.
I have already made a model of the antenna showing what it would look
like if it had fully opened (see http://www.mmedia.is/~bjj/gal_1.jpg).
Here the antenna dish is modeled as a paraboloid. Needless to say the 18
ribs are also parabolic in shape, created as a CSG difference using a
parabolod and planes.
The problem is modelling the antenna is its partially deployed
configuration.
I was able to easily model the ribs in their partially deployed positions
(see http://www.mmedia.is/~bjj/gal_2.jpg). Here 3 ribs are stuck to the
central mast and the other 15 are not fully deployed. The farther a rib
is from the 3 stuck ones the closer it is to full deployment. I modelled
this by translating/rotating the ribs differently depending on their
position (in effect rotating them around an axis near their base slightly
outside the cetral mast).
The big problem is modelling the golden antenna material itself between
the ribs. I don't think I can use 18 paraboloids, one between two every
ribs etc. since different rotations would be needed at the two ribs. A
triangle mesh is a possibility but I'm not a big fan of these when using
POV and I would need a lot of triangles (which I could generate by
writing a small program to do so). The third possibility I've thought of
is bicubic patches but in that case I'm (hopefully incorrectly) afraid I
would have trouble "aligning" the antenna material perfectly to the ribs.
I have not tried any of the above ideas yet - any suggestions ?
Bjorn Jonsson
bjj### [at] zzzmmedia is
http://www.mmedia.is/~bjj - Planetary maps and space renderings
Remove yyy and zzz to reply
Post a reply to this message
|
