measuring and visualizing performance of different pathfinding alogirthms across an aggregated, randomised maze generation algorithm
Some results:
both of these use Manhattan Distance for end heuristic and steps taken so far for start heuristic. I plan to mess around with the start a little bit, put less weight on the steps, see if that makes life interesting
and with these, im pushing this project to semi-done, pending bugs. the interactive interface is very elementary. it mostly just takes in coordinates. I'll probably work on it a little when im free. also update djikstra(basically just A* with no g heuristic) and breath for search (Don't see the point. weighted heuristic almost always better) soon.
Questions, comments and concerns welcome at [email protected]