TypeScript icon, indicating that this package has built-in type declarations

1.1.3 • Public • Published


An implementation of the A* Search Algorithm in JavaScript

See a demo at

Sample Usage

If you want just the A* search code (not the demo visualization), use code like this

<script type='text/javascript' src='astar.js'></script>
<script type='text/javascript'>
	var graph = new Graph([
	var start = graph.grid[0][0];
	var end = graph.grid[1][2];
	var result =, start, end);
	// result is an array containing the shortest path
	var graphDiagonal = new Graph([
	], { diagonal: true });
	var start = graphDiagonal.grid[0][0];
	var end = graphDiagonal.grid[1][2];
	var resultWithDiagonals =, start, end, { heuristic: astar.heuristics.diagonal });
	// Weight can easily be added by increasing the values within the graph, and where 0 is infinite (a wall)
	var graphWithWeight = new Graph([
	var startWithWeight = graphWithWeight.grid[0][0];
	var endWithWeight = graphWithWeight.grid[1][2];
	var resultWithWeight =, startWithWeight, endWithWeight);
	// resultWithWeight is an array containing the shortest path taking into account the weight of a node

A few notes about weight values:

  1. A weight of 0 denotes a wall.
  2. A weight cannot be negative.
  3. A weight cannot be between 0 and 1 (exclusive).
  4. A weight can contain decimal values (greater than 1).

Original (slower) implementation

The original version of the algorithm used a list, and was a bit clearer but much slower. It was based off the original blog post. The code is available at:

The newest version of the algorithm using a Binary Heap. It is quite faster than the original. Binary Heap taken from (license:

Running the test suite

Build Status

If you don't have grunt installed, follow the grunt getting started guide first.

Pull down the project, then run:

	npm install




Package Sidebar


npm i win111

Weekly Downloads






Unpacked Size

217 kB

Total Files


Last publish


  • win888