Rework design of corners #4
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Here are some of the cases that need to be considered:
single route turning: constant area of arc kite (r ~ sqrt(tan(θ/2)))two (or more) parallel routes turning along the same path: innermost radius as (1), others have same center pointsingle route turning across a non-turning route: ignore the non-turning routetwo or more parallel routes turning along the same path, but shifting relative to each other: both routes treated as innermosttwo routes turning in opposite directions, away from each other: arcs start at same perpendicular of their shared linecases (3) and (4) are annoying:

Top row is case (2) as a control; middle row is case (3); bottom row is case (4). First column calculates the radius for the red route as if it were the outer of two concentric routes; second column calculates it as if it were the inner of two concentric routes. (In case (4), the blue route has the same radius as the red route)
People seem to prefer the first column for case (3) but the second column for case (4). I'm not sure how to reconcile these programmatically.
I think the answer is that the second column is good enough for the first case.
I was going to write out an outline of an idea for the algorithm, but ran into trouble with step 3.
Δrr=R * sqrt(tan(θ/2))whereRis the global inner radius value, andθis the angle the arc sweeps.r + Δrnew idea:
r = R * sqrt(tan(θ/2))and arc starting point