Optimal Pathfinding on Weighted Grid Maps

In many computer games up to hundreds of agents navigate in real-time across a dynamically changing weighted grid map. Pathfinding these situations is challenging because the grids are large, traversal costs not uniform, and each shortest path has symmetric permutations, all which must be considered by an optimal online search. this work we introduce Weighted Jump Point Search (JPSW), new type pathfinding algorithm breaks symmetries introducing tiebreaking policy that allows us apply effective pruning rules regions. We show preserve at least one every cell their application can yield large performance improvements for pathfinding. give complete theoretical description algorithm, including pseudo-code. also conduct wide-ranging experimental evaluation, data from real games. Results indicate JPSW orders magnitude faster than nearest baseline, search using A*.