mesh angle constraints

We are working to optimize the number of unique angles on a given mesh while at the same time trying to maintain the initial shape of the input mesh as much as possible. At the moment we are getting better results in the new kangaroo but are still unable to get the hinge force angles totally consistent. There is a new angle snap component used in the kangaroo example files but I can't get that to work in 3d and I believe it is only meant to work in 2d. If there are any ideas about how we might be able to solve for this it would be greatly appreciated. thanks, jason

Attachments:

mesh_angle_constraint.gh, 31 KB

Replies to This Discussion

Permalink Reply by Daniel Piker on June 19, 2015 at 9:13am

Hi Jason,

The angle snap component works between pairs of lines - Are you saying you want to minimize the number of unique dihedral angles? I think it should be possible to make a hinge-snap goal for this.

Permalink Reply by machinehistories on June 19, 2015 at 11:35am

Daniel,

Yes the hope is to minimize the number of unique dihedral angles. Basically there will be custom hardware made to keep the mesh faces at the proper angle and the hope is that a reduced number of hardware can be made. This would eliminate the current situation where every angle is unique and ensure that some angles are reused a given number of times. I've tried to generate the proper lines on the mesh and feed them into the current angle snap component but it didn't work. The hinge component is getting very close to working but I am just unable to get a true snap. You can see here the input goal angles compared to the resulting angles. I think a hinge snap goal sounds like what I need. Can you make any suggestions on how to do this or are you generously (too generously) saying this is something you are going to code. I do feel this might be a helpful function that I imagine others can use to help optimize the construction of paneled works.

Permalink Reply by Daniel Piker on June 19, 2015 at 12:10pm

I just made a quick test - grouping the angles, then setting the target angle of each hinge to the average of its group. It seems to work very well - as you can see in the before and after graphs of the angles, they cluster nicely into groups with angles the same to 5 decimal places.So I don't think a hinge snap is actually even needed here (though it shouldn't be too hard to add, as the code is already there for length and angle snapping, and I probably will do so for completeness).

Without seeing the rest of your setup I can't know for sure, but maybe the angles aren't snapping because you have some other conflicting goals. Note that in the definition attached the edge lengths are allowed to vary, and I think without this freedom it would be impossible.

Attachments:

dihedralcluster.gh, 19 KB

Permalink Reply by machinehistories on June 19, 2015 at 1:19pm

This is fantastic. I will try it out right away. I can't thank you enough.

Permalink Reply by Arthur Lara on August 18, 2017 at 7:30am

Hi Jason and Daniel

I tried to make a sphere cut by a plane without triangulations
I would like to measure the angles between the bars of a shell
I can understand what you're discussing, but it's complicated
to understand the logical of the angle mesures

Whay I had to anchor the mesh of my sphere to work well???
Why the results were the same before and after ????

some help???

thanx in advance

Art

Attachments: