News for hackers. Stuff that matters.
Re:How to smooth and spread A* paths for an RTS by jcarrano on Tuesday January 31, 2023 @12:44PM (#34598559)
I wonder why the Eikonal equation (and variations) and level-set methods are so rarely discussed. They yield any-angle paths and can also take into account the kinematics of the vehicle (e.g. minimum turn radius). I used it to solve a pathfinding problem in robotics and it could perform parking maneuvers all by its own.

The resulting algorithm bears similarities to Dijkstra's or A* (i.e. it is a sort of value iteration, see the "fast marching method") but yields a scalar field, whose gradient will give the direction of the optimal path at any point.

Re:How to smooth and spread A* paths for an RTS by BlueTemplar on Wednesday February 01, 2023 @09:59AM (#34611337)
I guess because while minimum turn radius is unavoidable in real life, most games prefer to minimize it and its consequences as much as possible, due to the explosion in complexity they create ?
Re:How to smooth and spread A* paths for an RTS by graffix on Thursday February 02, 2023 @12:27PM (#34629950)
In game AI, these are discussed often under various guises like "influence maps" and "Dijkstra maps".