A* Search
A special case of best-first graph search that uses heuristics to improve speed. Works with 2D grids and currently only supports walls as obstacles within the environment. Written in CoffeeScript.
Usage
Add a-star-search to your package.json and run $ npm install
Require aStar in your app and define the parameters of your 2D grid:
aStar = require 'a-star-search'
startLocation = {xAxis: 1, yAxis: 1}
destination = {xAxis: 4, yAxis: 1}
environment = {walls: [{xAxis: 2, yAxis: 1}, {xAxis: 2, yAxis: 2}, {xAxis: 2, yAxis: 3}], worldSize: {xAxis: 10, yAxis: 10}}
Call the run method passing in the parameters defined above:
aStar.run(startLocation, destination, environment)
This returns the following shortest path:
[
{xAxis: 1, yAxis: 1, gCost: 0, hCost: 30, fCost: 30, parent: {xAxis: 1, yAxis: 1}},
{xAxis: 2, yAxis: 0, gCost: 14, hCost: 30, fCost: 44, parent: {xAxis: 1, yAxis: 1}},
{xAxis: 3, yAxis: 1, gCost: 28, hCost: 10, fCost: 38, parent: {xAxis: 2, yAxis: 0}},
{xAxis: 4, yAxis: 1, gCost: 38, hCost: 0, fCost: 38, parent: {xAxis: 3, yAxis: 1}}
]
Visual representation:
y
10 . . . . . . . . . . .
9 . . . . . . . . . . .
8 . . . . . . . . . . .
7 . . . . . . . . . . .
6 . . . . . . . . . . .
5 . . . . . . . . . . .
4 . . . . . . . . . . .
3 . . w . . . . . . . .
2 . . w . . . . . . . .
1 . x w x X . . . . . .
0 . . x . . . . . . . .
0 1 2 3 4 5 6 7 8 9 10 x
Running the test suite
$ npm install
$ grunt && NODE_ENV=development jasmine-node spec/aStarSpec.js
References
http://www.policyalmanac.org/games/aStarTutorial.htm