Grasshopper

algorithmic modeling for Rhino

Hi Daniel,

 

I have tried the Origami example file you shared.

I noticed, that the folding process is depending on the rest angle which is useful to avoid the penetration when the folding factor up to 1(while 0 develop into a flat paper...).

 

There are some challenge Origami which required 100% folding + collision in order to create certain force to 'push and roll' them together, for example a flasher Origami:

 

 

I have tried this like below:

The foldability is fine, but I can’t solve the Problem of penetration based on self-collision.

Do you have any hint for that?

 

Thanks in advance!

 

 

Views: 2200

Replies to This Discussion

Okay, here's something that hopefully will help.

I wrote a new variation of the hinge goal that prevents the mesh folding through itself.

(because it is a custom goal, you will need to right click and update the referenced assembly location to your own copy of KangarooSolver.dll)

It was a bit tricky because we can't simply clamp the angle to some range, since all fold angles from -pi to +pi are valid. So we need to determine whether any change in sign of the fold angle is the result of it passing through itself or just an allowable switch of direction.

I tackled this by keeping track of whether the fold is currently a mountain or a valley, and only allowing this to switch when the angle is <pi/2.

Now this will only prevent individual pairs of joined faces from folding through each other - some origami designs might require actual collision prevention between non-adjacent faces, but that will require some additional work...

Attachments:

Testing this out - I managed to make a flapping bird starting from a flat mesh just using the grab tool:

It was a bit fiddly, and when multiple layers stack up there are still some collision issues, but I think it is promising. Probably it would also help to sequence the activation of the creases, so you can turn them on only after the preceding folding step is complete...

Also most origami actually relies on some degree of non-rigidity during the process - here I kept the hinge strength low, and used a low number of iterations in the solver to allow this.

Hello Daniel, I was trying to get the sequencing of activations to work where I can trigger the folding simulation step by step. Could you suggest an approach to achieve that using the origami component? Thanks!

it is really awesome!

I am trying this CG u made and will give my Feedback soon!

Hi Daniel,

many thanks for your custom goal again. I can tell that it works really much better and could be controlled rational!

As you mentioned about, that in some origami case we need actual collision prevention between non-adjacent faces. I think it is the key matter! Principally we could image this like a rolling paper Strip.

Without the overall collision the crease could just easily passing through itself.

 I found 2 interesting example on your vimeo channel.

This is very helpful because there are friction and both-side collision in between which prevent the penetration, and it increases the forces during the process so that you can actually ‘’tied-up’’ them!

 

I have tried to achieve this base on Collider-Goal but it doesn’t work appropriate…

So I was thinking probably the ‘tied-up knot’ principle could also work for those folding/ collision problem, what do you think?

Attachments:

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service