I am attempting to come up with a way to detect when one ring object collides with another.
Basically, I have N number of rings all centered in the exact same position. I then have an extremely thin ring that grows and shrinks during runtime (the blue ring) and I want to receive collision detection when this ring “enters” and “exits” any of the larger rings.

What would be the best way to do something like this? I can usually figure problems like this out but I have to admit I am stumped on this one.
My current logic is just manually getting the min/max size of each ring that the blue ring will collide with and then check the blue rings radius (object scale actually) to see if it matches one of the predetermined min/maxes.
The problem is that I need to provide logic that will allow the collision detection of the blue ring w/ any other ring when the other rings are dynamically generated. Meaning, the above image might be one scenario of ring layout but the next time there might be 2 green rings fatter than the one above and only 1 thinner red ring.
I started down the path of creating a simple “ring” fbx object and then importing that into my scene as a prefab. My idea was that at runtime I would create the rings the size I needed at the time (by scaling them) and placing them all in the same position. I would then create the blue ring, place that in the center of all of them, and then scale it watching for OnTriggerEnter/OnTriggerExits from the collisions.
Where I am getting stuck is applying a collider to each of these rings. I cant use a primitive box or sphere collider due to the shape (i.e. they are hollow). I then try to add a mesh collider to each of them and a rigidbody to JUST the blue ring (with kinematic enabled so I can get one of the trigger responses from the collision matrix) but I am receiving the “Actor::updateMassFromShapes: Compute mesh inertia tensor failed for one of the actor’s mesh shapes! Please change mesh geometry or supply a tensor manually!” exception.
On top of all that, I am still not convinced this is the best approach to this problem. The ending result needs to run on mobile devices and I am worried that using mesh colliders in this fashion will be a problem for those platforms.
Any suggestions and help would be greatly appreciated!
PAR
