Really? That seems like a super basic thing to do - the only thing that I’d be doing differently is using a hardcoded diameter
value. I’d like to explain my use case so you can understand why I’m having to jump through these hoops.
1.) I wanted to just use the Tilemap colliders
originally, however, picture a scenario where I have 2 tilemaps on top of each other. One is from the Ground
, other for any Vegetation
. What if I want a tree to be an obstacle, but don’t want it blocking the area behind the treetop? In this case because of the collider it still gets detected by an obstacle, as shown:

And besides, in this case just to demonstrate I used tilemap colliders
to detect where the obstacles are, and the diameter
option doesn’t seem to be changing the “size
” of the red squares around which Agents navigate, it just makes more of those small ones around the area. Maybe it does function as intended but the visualization is very misleading. (I want them to not cut corners).
2.) In case this really isn’t an option to me, and I have to resort to erosion, would you be able to explain a bit more on what erosion is and how it works? What sort of rule would I have to write and what would it do?
I just want to increase the little red box so people don’t look like they’re cutting into obstacles.