volumes (polygons) not touching the ground.

966 views
Skip to first unread message

tony_jay

unread,
Oct 18, 2005, 12:55:31 PM10/18/05
to KML Discussions
I am trying to make an extruded polygon that does not touch the ground.

This is to represent a block of airspace.

It can have a common height at the top and the bottom.

As far as I can see all my extruded object have to be extruded from the ground.

Is there any 3 dimensional positioning that I have missed ?

An easier way for me to express what I'm doing is 'how do I draw a cube 2000m above sea level, with dimensions 100m'

Tony

Brian_Flood

unread,
Oct 18, 2005, 1:51:30 PM10/18/05
to KML Discussions
The Extrude tag is not going to help in this case. However, you can build the 3D meshes yourself (e.g. create a 6 sided cube with your coordinates). Thankfully you can use polygons and GE will do the much harder tessellation calculations although it is certainly possible to submit triangles (your cube would have 12 of them).

so good and bad, you can accomplish what you want with a bit more work but the ultimate 3D Mesh to KML conversion is going to contain many, many extraneous vertices due to the lack of the 3D primitives like tri and fan strips.

tony_jay

unread,
Oct 18, 2005, 4:01:13 PM10/18/05
to KML Discussions
Thanks,

Knowing that it is not possible and requires me to build in from scratch is useful information. Fortunately I think airspace are simple convex shapes

Tony

Lrae

unread,
Oct 22, 2005, 3:42:33 AM10/22/05
to KML Discussions
Tony,
Welcome to the BBS and Google Earth.

You may also be trying to re-invent the wheel! Look at the attached placemark.

I am currently re-packaging the B and C airspace (only the first few B's are done)
166563-USAirspace.kmz

TJ1

unread,
Oct 22, 2005, 12:57:12 PM10/22/05
to KML Discussions
There's something to be aware of when creating large-area polygons, which caught me out recently.

I had several polygons at an altitude of around 1000 meters, each of which covered 1 degree-squared of latitude and longitude over an ocean, using the 4 corners as plot-points.

It took me a while to figure out that the round 'hole' in the middle of the square polygons was in fact the surface of the ocean poking through!

It then made me realise that these polygons are not useful for mapping airspace accurately over large areas unless you plot a point every minute of arc or so (I've not experimented to find out what the best gap would be).

It suggests that a useful feature to be added to GE would be the ability to specify that polygons and lines are plotted using straight-line or Great Circle algorithms.

This would be rather like the tessellate option for LineStrings I guess.

stillinorbit

unread,
Oct 29, 2005, 12:22:10 AM10/29/05
to KML Discussions
WOW!

Boy Howdy, this commercial pilot, flight instructor, and GIS business proprietor is impressed on all three counts. Thank you for your efforts!

If I may... how did you do it? Where did you get the digital x,y,z coordinates for the airspace volumes?

stillinorbit

unread,
Oct 29, 2005, 12:23:48 AM10/29/05
to KML Discussions
I forgot to add this.

Any chance we'll see any class D airspace?

How about TFR's??

Regards,

Stu

justtesting

unread,
Nov 14, 2005, 5:48:45 PM11/14/05
to KML Discussions
hi,

I have been working on a gospatial project which used DAFIF data. It holds information on all the airspaces, ATS routes, military training routes etc. Rather suprisingly, you can download the spec and the data here, DAFIF stuff

cheers,

Alex

Hanglider

unread,
Apr 21, 2006, 12:16:05 PM4/21/06
to KML Discussions
Here is a link to another post that has a Postmark of the Class C steps over Melbourne Victoria Australia.

http://bbs.keyhole.com/ubb/showthreaded.php/Cat/0/Number/394081/page/0/vc/1
Reply all
Reply to author
Forward
0 new messages