Introducing 'Exoskeleton' - A wireframe thickening tool

A collaborative effort by David Stasiuk and Daniel Piker, Exoskeleton brings simple wireframe thickening to Grasshopper. You input a network of lines, and it turns them into a solid

(without the heavy calculation of a Boolean intersection of many pipes and spheres).

The input line networks can have any topology, and need not form closed polygons or volumes, so could come from algorithms such as DLA, leaf venation, or Woolly threads.

The resulting meshes are ideal for 3d printing and further processing, such as subdivision with WeaverBird and relaxation with Kangaroo.

There are settings for the thickness of the struts, node sizes, and whether to leave openings at nodes with only one connected line.

The approach we used is loosely based on the one described in the paper Solidifying Wireframes by Ergun Akleman et al.

Thanks to Giulio Piacentino for helpful discussion in the development of this idea, for WeaverBird, and the GHA wizard, to Mateusz Zwierzycki for convex-hull ideas, and Kristoffer Josefsson for helpful discussion.

Exoskeleton001.gha

(component will appear under the Mesh>Triangulation Tab)

Comment

You need to be a member of Grasshopper to add comments!

Comment by David Stasiuk on May 14, 2013 at 3:25am: Hi Vicente-

I am working on just such an update this moment...In the short term I am looking at making it possible to locally alter both strut thickness and node offset distances, and hopefully will have something in the next several days. Additionally, a key component to this update is that node thicknesses will be better adjusted...currently, the node size is determined by either the node depth assigned by the user, or by the minimum offset required by the narrowest angle between struts in a node as a function of the strut thickness. The update will allow for the node offsets to shrink locally for each strut. In the longer term I am also interested in looking into mesh self-intersection for the purposes of making more organic meshes...but that's not for right now. So you can expect an update soon...

Unfortunately, we won't be actively supporting V4...we haven't been able to identify what the bug is, and we are working to support 0.9.0050+.

I also hope during this update to address the issue of the lines running to the origin. The reason for this error is pretty simple: the nodes are generated using a convex hull, and when these hulls overlap...which is to say, when any hull engulfs one of its struts entirely, the mesh breaks down. One way to help prevent this from happening is to use Daniel's topologizer, but that won't always work. I am hoping to implement some changes that will either generate "super nodes" that combine the input struts between highly proximate nodes, or at the very least generate a detailed error message that illustrates where the input is causing the breakage.

Comment by Vicente Soler on May 13, 2013 at 4:04pm: Are there plans in the near future for an update that allows different thickness per line? I'm using offset mesh right to change the thickness but the results aren't very good.

Nevertheless, great plugin.

Comment by Gieljan Vantyghem on April 19, 2013 at 1:30pm: Very nice tool! thanks alot!!

Comment by jose salinas on April 18, 2013 at 2:21am: GREAT JOB!!! The combination with Topologizer .... great!

Comment by martyn hogg on April 2, 2013 at 4:28pm: Hi,

Is Exoskeleton ever likely to work with Rhino4? It looks really useful and something I'd really like to experiment with, I just can't afford to upgrade to Rhino5 at the moment :(

I get the same problem as described in this post re: flattened meshes http://www.grasshopper3d.com/forum/topics/exoskeleton

Keep up the good work!

Comment by Isak Bergwall on March 15, 2013 at 9:47am: I tried using this on edges from a diagrid created with lunchbox, but when I add the curves (that comes from a dupborder on the panels) to Exoskeleton I get the following error. Error: Solution exception: Arithmetic operation resulted in an overflow." I tried with closed curves, exploded curves, fewer curves.. Any idea what could be wrong?

Comment by Mario A. Benavides on January 17, 2013 at 2:50pm: http://www.grasshopper3d.com/forum/topics/exoskeleton-lines-to-origin

Comment by Mario A. Benavides on January 17, 2013 at 9:02am: Hello

Is there a solution to patric guenther's problem?, I am running into the same problem repeatedly.

Thanks

Comment by djordje on January 10, 2013 at 10:43am: Understood. Sorry for disturbance.

Comment by David Stasiuk on January 10, 2013 at 9:11am: Hi djordje...haven't yet taken a close look at why it doesn't work in R4...I'll try to get to it soon if possible!