By: N-Studios
A free-to-use open source project. Built using the A* pathfinding algorithm, on C#, it makes it incredibly easy to set up grid-based pathfinding in a Unity 2D workspace. You can have a look at or build on top of the scripts inside the /Scripts folder. A setup tutorial is listed below and the whole Examples.cs is described with comments on almost every line. Feel free to use in any kind of projects. Credit is not required but is appreciated.
At the very top of your code make sure to reference the AStar namespace by doing:
using Ultimate.Algorithms.AStar;Add some variables to make the tweaking of the behavior of the pathfinding easy. Here's an example:
public int startNodeIndex = 0; // The index of the start node in the grid.nodes list
public int endNodeIndex = 1; // The index of the end node in the grid.nodes list
public float nodeUnwalkablePercentange = 20f; // Chance of a node being unwalkable
public Vector2Int size; // Size of the grid - X means width and Y means heightThose are the values that can be changed from the user. However, we still need a few more, three to be precise. They are going to be the ones that are not accessible from out of the script and will store the pathfinder info:
Vector2Int originPosition; // This represents the bottom-left corner node of our grid
AStarGrid grid; // A local grid that is then being passed to the pathfinder
List<Vector3> path = new List<Vector3>(); // This empty Vector3 list is goint to store the path generated after the pathfinding is doneNow, time to generate the nodes:
public void Generate()
{
path = new List<Vector3>(); // We first define an empty list where we will store the path as a list of Vector3 points
originPosition = new Vector2Int(-size.x / 2, -size.y / 2); // Then we calculate the origin position so that the grid is always centered (you can change that)
List<Node> nodes = new List<Node>(); // Since the grid works with nodes we need an empty list for them as well
for (int y = originPosition.y; y < originPosition.y + size.y; y++)
{
for (int x = originPosition.x; x < originPosition.x + size.x; x++)
{
// For each point in the grid we create a new node with the X and Y coordinates
nodes.Add(new Node(1f, x, y)); // 1f means that it is walkable and 0f - that is is not walkable
}
}
}For now, we only have walkable points. What about the "obstacles"? Well, this code randomly generates unwalkable nodes (using the nodeUnwalkablePercentage):
public void Generate()
{
path = new List<Vector3>(); // We first define an empty list where we will store the path as a list of Vector3 points
originPosition = new Vector2Int(-size.x / 2, -size.y / 2); // Then we calculate the origin position (bottom-left corner) so that the grid is always centered (you can change that)
List<Node> nodes = new List<Node>(); // Since the grid works with nodes we need an empty list for them as well
for (int y = originPosition.y; y < originPosition.y + size.y; y++)
{
for (int x = originPosition.x; x < originPosition.x + size.x; x++)
{
// For each point in the grid we create a new node with the X and Y coordinates
nodes.Add(new Node(1f, x, y)); // 1f means that it is walkable and 0f - that is is not walkable
}
}
// For each of the already defined nodes...
for (int i; i < nodes; i++)
{
// If the current node is the starting nor the end node we return!
if (i == startNodeIndex || i == endNodeIndex) continue;
//...else based on the unwalkable chance we set its walkable state
bool walkable = Random.Range(0f, 100f) <= nodeUnwalkablePercentange;
if (unwalkable) node.SetPenalty(0f); // SetPenalty(1f) == walkable and SetPenalty(0f) means unwalkable
}
}Finally, create the grid:
grid = new AStarGrid(new Vector2Int(-size.x / 2, -size.y / 2), size, nodes);Now that we have the full grid, it is time to find the shortest path, if there is one, right? Here's how you do it:
public void FindPath()
{
// Simply instantiate a new AStarPathfinding class...
AStarPathfinding pathfinding = new AStarPathfinding();
//... and call the Pathding() method by specifying your grid start and end node
path = pathfinding.Pathfind(grid, startNodeIndex, endNodeIndex);
}Congratulations, you've now got a pathfinding code!
Just one last thing - we all love data, don't we? Why not visualize all of this? No worries, we got it covered:
// This method is just to visualize the nodes
private void OnDrawGizmosSelected()
{
// If the grid is null or has no nodes we return as there won't be anything to visualize
if (grid == null || grid.nodes.Count == 0) return;
// Otherwise we loop through all the nodes...
for (int i = 0; i < grid.nodes.Count; i++)
{
//...if it is the start node - paint it in red
if (i == startNodeIndex)
{
Gizmos.color = Color.red;
Gizmos.DrawCube(new Vector3(grid.nodes[i].X + 0.5f, grid.nodes[i].Y + 0.5f, 0f), new Vector3(1f, 1f, 0f));
continue;
}
//... if it is the end node - paint it in blue
else if (i == endNodeIndex)
{
Gizmos.color = Color.blue;
Gizmos.DrawCube(new Vector3(grid.nodes[i].X + 0.5f, grid.nodes[i].Y + 0.5f, 0f), new Vector3(1f, 1f, 0f));
continue;
}
//... if it is an ordinary node that is walkable - paint it in green
if (grid.nodes[i].Walkable)
{
Gizmos.color = Color.green;
Gizmos.DrawCube(new Vector3(grid.nodes[i].X + 0.5f, grid.nodes[i].Y + 0.5f, 0f), new Vector3(0.9f, 0.9f, 0f));
continue;
}
//...if it is an ordinary node paint and is not walkable - paint it in black
else
{
Gizmos.color = Color.black;
Gizmos.DrawCube(new Vector3(grid.nodes[i].X + 0.5f, grid.nodes[i].Y + 0.5f, 0f), new Vector3(0.9f, 0.9f, 0f));
continue;
}
}
if (path != null && path.Count > 0)
{
Gizmos.color = Color.red;
Gizmos.DrawLine(new Vector3(grid.nodes[startNodeIndex].X, grid.nodes[startNodeIndex].Y, 0), path[0]);
for (int i = 0; i < path.Count - 1; i++)
{
var p1 = new Vector3(path[i].x + 0.5f, path[i].y + 0.5f, 0);
var p2 = new Vector3(path[i + 1].x + 0.5f, path[i + 1].y + 0.5f, 0);
Gizmos.DrawLine(p1, p2);
}
}
}So that's it, just keep in mind that the visualization part only works in "Editor" mode and NOT "Play" mode. You can still create a code that does that in "Play" mode - the process should be very similar.
Contributions, issues, and feature requests are welcome!
Give a ⭐️ if you like this project!
You can also check us out on the Unity Asset Store.