Planarization of Voronoi Patterns

Hello,

It had come to me that this question has been asked quite a bit as I googled for answers, but it seems like some of the posts were quite dated or complex which is a bit hard to understand for a student like me. Hence I thought trying the forum may be a good option!

I have a Voronoi pattern which had been created using kangaroo physics. My ultimate goal is to create cells with these voronoi patterns but first I think I must planarize them.

This is my semi dome pattern. I have a demonstration definition that uses kangaroo physics to planarize those voronoi patterns. However, it needs to have restraining curves 'CurvePull' to hold the overall form. The demonstration had curves that were spaced at the top and bottom of its polygon pattern, but for my semi-dome it does not have curves to restrain the overall form. I tried using the curves which I was finding the form with(in green on the picture), but it does not work, when i toggle the simulation, everything just disappears, i think the problem lies with its constraining curves.

Alternative on another script, I had a loft surface from the curves. Using pull point on the control points of the voronoi polylines and evaluate them to the lofted surface to create planarized surfaces. However, this scripting would create surfaces that do not have a uniform edge with adjacent cells, which would be problematic once i try to fabricate the cells.

I am really really hoping if kangaroo expert or planarization expert could give me some guidance in terms of planarizing the pattern I have here. Either by introducing me to a way of fixing the constraining curves or even new methods of planarization i am willing to learn. It does not need to be exact to the voronoi patterns as the kangaroo simulation I attempted would need to distort the lines anyways, but I was trying to retain its overall formal expression as a semi-dome.

Thanks very much in advance!

Cheers!

Attachments:

Semi Dome 2.3dm, 69 KB
Semi Dome 2.gh, 22 KB

Replies to This Discussion

Permalink Reply by Dimitar Baldzhiev on May 24, 2015 at 5:37am

I have successfully planarized voronoi domes with Kangaroo. I would suggest you use kangaroo 2 instead of kangaroo 0.99. Use the PlanarHex.gh example from here http://www.grasshopper3d.com/group/kangaroo/page/kangaroo2-addition...

Permalink Reply by Andy L. on May 24, 2015 at 6:28am

Hi Dimitar

thanks for the reply. I have tried the PlanarHex.gh example from the kangaroo page. Its turning out a bit funny for me. The surface is separate from the curves it's produced and the curves themselves are a bit out of shape. I tried to adjust the numbers for the limits in the ClampLength component it is in odd shapes.

Permalink Reply by Dimitar Baldzhiev on May 24, 2015 at 12:23pm

Graft the curves that come as output from the solver. This way you will force the planar surfaces to be created separately.

Permalink Reply by Andy L. on May 24, 2015 at 6:29am

I have used the Curves from the simulation I had shown in the post as the curves for input. Would that be the problem?

Permalink Reply by Dimitar Baldzhiev on May 24, 2015 at 12:26pm

The curves from your post look good to me. They shouldn't create any problems.

Permalink Reply by Andy L. on May 24, 2015 at 4:13pm

Hey Dimitar,

I got it to work, I used a another set of curves and the script had worked for me. Maybe the old curves had a little bit of discrete problems. But i will try grafting the curves output if i run into the problem again!

thanks a lot once again!

Permalink Reply by Dimitar Baldzhiev on May 24, 2015 at 4:45pm

The way I use the solver for planarization is with a datadam component and push the planar curves by clicking (or every 5 sec) because above certain number planarSrf component can be a bottleneck. Definitely graft your output if you dont want the regions to merge in single surface at xy plane.

RSS