Skip to content

Latest commit

 

History

History
58 lines (32 loc) · 2.51 KB

README.md

File metadata and controls

58 lines (32 loc) · 2.51 KB

Grid-Search-on-a-Costmap

Given a grid, where certain cells are marked as obstacles, the objective is to find a path from an initial to a goal cell that avoids collisions with the obstacles. More specifically, each grid cell will have a numeric value in the range [0,1]. A value of 0 indicates that the cell is an obstacle, which must always be avoided. A value of 1 indicates that the cell is free.

grayscale image note

In this way, obstacle cells have a cost of ∞, while free cells have a cost of 1. Note that the cost increases as the value of the cell gets closer to 0. The method name is as follows: DFS for depth-first search, BFS for breadth-first search, AStarZero for A*-search with the zero heuristic, AStarEuclidean for A*-search with the Euclidean heuristic, AStarManhattan for A*-search with the Manhattan heuristic. The program should read the file grid.txt and search for a solution from (rstart,cstart) to (rend,cend). It should then write the solution to the file path.txt. The moves are LEFT, RIGHT, UP, DOWN (no diagonal moves allowed). Note that DFS and BFS treat the cells with value (> 0) as free cells since these methods just focus on computing a path. Essentially, these methods ignore the cost. A* should take the cell-traversal costs into account so that it can find the shortest path to the goal. The cost of passing through a cell (i, j) should be computed as cost(i, j) = 1/val(i, j).

Result

BFS METHOD

breadth first search

DFS METHOD

Depth first search

A* WITH ZERO HEURISTIC

A STAR

A* WITH MANHATTAN HEURISTIC

A STAR

A* WITH EUCLIDEAN HEURISTIC

A STAR

FRINGE SEARCH WITH MANHATTAN HEURISTIC

FRINGE SEARCH