A* grid pathfinding: Examples & Code

Algorithm Visualizations November 16, 2026 javascript

Published by Dimitris Kougioumtzis · Updated

Find a path from top-left to bottom-right using four-direction moves. A* uses distance + admissible Manhattan heuristic and finds a shortest path.

Order open cells by accumulated distance plus Manhattan distance, which is admissible for four unit-cost moves.

Open the 3D visualization

A* grid pathfinding example state preview
Preview of the example's final state. Complete values are available in the worked example and interactive timeline.

When to use A* grid pathfinding

Shortest routes around grid obstacles.

Choosing an alternative

Greedy best-first ignores cost already paid; BFS is a heuristic-free baseline.

Worked example: input and state changes

Example input

0,0,0,0;1,1,0,1;0,0,0,0;0,1,1,0

These checkpoints follow the actual example timeline. State details show the stored values and decisions; use the visualization to inspect every intervening step.

  1. Step 0: Scores are Manhattan estimates plus distance traveled.

    State at this checkpoint
    {
      "values": [
        6,
        "∞",
        "∞",
        "∞",
        "×",
        "×",
        "∞",
        "×",
        "∞",
        "∞",
        "∞",
        "∞",
        "∞",
        "×",
        "×",
        "∞"
      ],
      "active": [],
      "kind": "matrix",
      "columns": 4,
      "blocked": [
        4,
        5,
        7,
        13,
        14
      ],
      "selected": []
    }
  2. Step 4: Improve cell 2: distance 2, estimate 4.

    State at this checkpoint
    {
      "values": [
        6,
        6,
        6,
        "∞",
        "×",
        "×",
        "∞",
        "×",
        "∞",
        "∞",
        "∞",
        "∞",
        "∞",
        "×",
        "×",
        "∞"
      ],
      "active": [
        2
      ],
      "kind": "matrix",
      "columns": 4,
      "blocked": [
        4,
        5,
        7,
        13,
        14
      ],
      "selected": [
        0,
        1
      ]
    }
  3. Step 9: Expand cell [1,2], distance 3.

    State at this checkpoint
    {
      "values": [
        6,
        6,
        6,
        6,
        "×",
        "×",
        6,
        "×",
        "∞",
        "∞",
        "∞",
        "∞",
        "∞",
        "×",
        "×",
        "∞"
      ],
      "active": [
        6
      ],
      "kind": "matrix",
      "columns": 4,
      "blocked": [
        4,
        5,
        7,
        13,
        14
      ],
      "selected": [
        0,
        1,
        2,
        3,
        6
      ]
    }
  4. Step 13: Improve cell 11: distance 5, estimate 1.

    State at this checkpoint
    {
      "values": [
        6,
        6,
        6,
        6,
        "×",
        "×",
        6,
        "×",
        "∞",
        8,
        6,
        6,
        "∞",
        "×",
        "×",
        "∞"
      ],
      "active": [
        11
      ],
      "kind": "matrix",
      "columns": 4,
      "blocked": [
        4,
        5,
        7,
        13,
        14
      ],
      "selected": [
        0,
        1,
        2,
        3,
        6,
        10
      ]
    }
  5. Step 17: Path length = 6.

    State at this checkpoint
    {
      "values": [
        6,
        6,
        6,
        6,
        "×",
        "×",
        6,
        "×",
        "∞",
        8,
        6,
        6,
        "∞",
        "×",
        "×",
        6
      ],
      "active": [
        0,
        1,
        2,
        6,
        10,
        11,
        15
      ],
      "kind": "matrix",
      "columns": 4,
      "blocked": [
        4,
        5,
        7,
        13,
        14
      ],
      "selected": [
        0,
        1,
        2,
        6,
        10,
        11,
        15
      ],
      "path": [
        0,
        1,
        2,
        6,
        10,
        11,
        15
      ],
      "result": 6
    }

Expected final state

Path length = 6.

Complexity and input bounds

Family: Graphs. Time complexity: O(V²).

Space complexity: O(V) distance, parent, and frontier state.

Working-space bounds describe algorithm state and any explicitly noted output, excluding the visualization's saved timeline. Recording frames can use more memory than running the function alone. Here n is sequence length, V and E are vertex and edge counts, and other parameters refer to the named input dimensions.

Rectangular grid up to 8 × 8: 0 = free, 1 = obstacle. Start and goal must be free.

The playground validates these limits before starting. Downloaded functions are callable implementations, with parameter and output-buffer contracts in their comments; they do not all repeat the playground's input validation.

Common mistakes and limitations

Changing movement or edge costs can invalidate the heuristic's admissibility argument.

Further reading

For background on algorithm analysis and the JavaScript function interface:

Editorial and testing approach · Browse all algorithm guides

Complete JavaScript implementation

javascript
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
// Grid dimensions 1–8; row-major blocked values 0/1, free start and goal. Writes the path indices including endpoints; returns moves or -1. A* uses the admissible Manhattan heuristic and returns a shortest path. Reference snippets assume validated playground inputs. Fixed scratch buffers have 256 slots. Sorting variants may use a different traversal strategy than the animation.
function algorithm(blocked, rows, columns, output) {
    let n = rows * columns; let dist = Array(256).fill(0); let parent = Array(256).fill(0); let closed = Array(256).fill(0);
    for (let i = 0; i < n; i++) { dist[i] = 1000000; parent[i] = -1; }
    dist[0] = 0;
    for (let step = 0; step < n; step++) {
        let u = -1; let best = 1000000;
        for (let v = 0; v < n; v++) {
            let estimate = rows - 1 - Math.floor(v / columns) + columns - 1 - v % columns;
            let score = dist[v] + estimate;
            if (closed[v] == 0 && dist[v] < 1000000 && score < best) { best = score; u = v; }
        }
        if (u < 0) { break; }
        closed[u] = 1; if (u == n - 1) { break; }
        for (let direction = 0; direction < 4; direction++) {
            let v = -1;
            if (direction == 0 && u >= columns) { v = u - columns; }
            if (direction == 1 && u + columns < n) { v = u + columns; }
            if (direction == 2 && u % columns > 0) { v = u - 1; }
            if (direction == 3 && u % columns < columns - 1) { v = u + 1; }
            if (v >= 0 && blocked[v] == 0 && closed[v] == 0 && dist[u] + 1 < dist[v]) { dist[v] = dist[u] + 1; parent[v] = u; }
        }
    }
    if (closed[n - 1] == 0) { return -1; }
    let path = Array(256).fill(0); let count = 0;
    for (let v = n - 1; v >= 0; v = parent[v]) { path[count++] = v; }
    for (let i = 0; i < count; i++) { output[i] = path[count - 1 - i]; }
    return count - 1;
}

Source

Primary research source for the algorithm or its underlying method. The playground is a bounded teaching implementation, not the publication's original program.