4 ms·
You may want to look into improvements to A* for grids, like Rectangular Symmetry Reduction.
by tavianator 1y ago
You may want to look into improvements to A* for grids, like Rectangular Symmetry Reduction.
- taneq 1y agoAlso jump point search: https://zerowidth.com/2013/a-visual-explanation-of-jump-point-search/ https://zerowidth.com/2013/a-visual-explanation-of-jump-poin...
- dietr1ch 1y agoIf just use A*, but you rank open to loop for lowest (f, h) pairs, then the search frontier just dives despite having multiple optimal paths, as the new node tie-breaking ensures we prefer nodes that seem closest to the goal.
- porphyra 1y agoafaik jump point search would work for uniform cost grids but not if there's the exponential term that OP has