摘要：Path planning for 2D grid maps has been studied for a long time, however the substantial leaps have just been achieved with recent series of Theta* variations, which allow the found path more efficient and realistic. Since the search is on grid, optimality is not guaranteed by incremental searching methods including Dijkstra, A*, Theta* and their variations like A*-PS, Lazy-Theta*, etc. In this paper, we present several attempts of improving sub-optimal path planning techniques in 2D grid maps. The combination of these attempts has reached results that are asymptotic to the optimal solution in terms of path length. The runtime for this method is much lower compared to that of the brute-force A* on Visibility Graphs.
