Up to date

This page is up to date for Godot 4.2. If you still find outdated information, please open an issue.

AStarGrid2D

Inherits: RefCounted < Object

An implementation of A* for finding the shortest path between two points on a partial 2D grid.

Description

AStarGrid2D is a variant of AStar2D that is specialized for partial 2D grids. It is simpler to use because it doesn't require you to manually create points and connect them together. This class also supports multiple types of heuristics, modes for diagonal movement, and a jumping mode to speed up calculations.

To use AStarGrid2D, you only need to set the region of the grid, optionally set the cell_size, and then call the update method:

var astar_grid = AStarGrid2D.new()
astar_grid.region = Rect2i(0, 0, 32, 32)
astar_grid.cell_size = Vector2(16, 16)
astar_grid.update()
print(astar_grid.get_id_path(Vector2i(0, 0), Vector2i(3, 4))) # prints (0, 0), (1, 1), (2, 2), (3, 3), (3, 4)
print(astar_grid.get_point_path(Vector2i(0, 0), Vector2i(3, 4))) # prints (0, 0), (16, 16), (32, 32), (48, 48), (48, 64)

To remove a point from the pathfinding grid, it must be set as "solid" with set_point_solid.

Properties

Vector2

cell_size

Vector2(1, 1)

Heuristic

default_compute_heuristic

0

Heuristic

default_estimate_heuristic

0

DiagonalMode

diagonal_mode

0

bool

jumping_enabled

false

Vector2

offset

Vector2(0, 0)

Rect2i

region

Rect2i(0, 0, 0, 0)

Vector2i

size

Vector2i(0, 0)

Methods

float

_compute_cost ( Vector2i from_id, Vector2i to_id ) virtual const

float

_estimate_cost ( Vector2i from_id, Vector2i to_id ) virtual const

void

clear ( )

void

fill_solid_region ( Rect2i region, bool solid=true )

void

fill_weight_scale_region ( Rect2i region, float weight_scale )

Vector2i[]

get_id_path ( Vector2i from_id, Vector2i to_id )

PackedVector2Array

get_point_path ( Vector2i from_id, Vector2i to_id )

Vector2

get_point_position ( Vector2i id ) const

float

get_point_weight_scale ( Vector2i id ) const

bool

is_dirty ( ) const

bool

is_in_bounds ( int x, int y ) const

bool

is_in_boundsv ( Vector2i id ) const

bool

is_point_solid ( Vector2i id ) const

void

set_point_solid ( Vector2i id, bool solid=true )

void

set_point_weight_scale ( Vector2i id, float weight_scale )

void

update ( )


Enumerations

enum Heuristic:

Heuristic HEURISTIC_EUCLIDEAN = 0

The Euclidean heuristic to be used for the pathfinding using the following formula:

dx = abs(to_id.x - from_id.x)
dy = abs(to_id.y - from_id.y)
result = sqrt(dx * dx + dy * dy)

Note: This is also the internal heuristic used in AStar3D and AStar2D by default (with the inclusion of possible z-axis coordinate).

Heuristic HEURISTIC_MANHATTAN = 1

The Manhattan heuristic to be used for the pathfinding using the following formula:

dx = abs(to_id.x - from_id.x)
dy = abs(to_id.y - from_id.y)
result = dx + dy

Note: This heuristic is intended to be used with 4-side orthogonal movements, provided by setting the diagonal_mode to DIAGONAL_MODE_NEVER.

Heuristic HEURISTIC_OCTILE = 2

The Octile heuristic to be used for the pathfinding using the following formula:

dx = abs(to_id.x - from_id.x)
dy = abs(to_id.y - from_id.y)
f = sqrt(2) - 1
result = (dx < dy) ? f * dx + dy : f * dy + dx;

Heuristic HEURISTIC_CHEBYSHEV = 3

The Chebyshev heuristic to be used for the pathfinding using the following formula:

dx = abs(to_id.x - from_id.x)
dy = abs(to_id.y - from_id.y)
result = max(dx, dy)

Heuristic HEURISTIC_MAX = 4

Represents the size of the Heuristic enum.


enum DiagonalMode:

DiagonalMode DIAGONAL_MODE_ALWAYS = 0

The pathfinding algorithm will ignore solid neighbors around the target cell and allow passing using diagonals.

DiagonalMode DIAGONAL_MODE_NEVER = 1

The pathfinding algorithm will ignore all diagonals and the way will be always orthogonal.

DiagonalMode DIAGONAL_MODE_AT_LEAST_ONE_WALKABLE = 2

The pathfinding algorithm will avoid using diagonals if at least two obstacles have been placed around the neighboring cells of the specific path segment.

DiagonalMode DIAGONAL_MODE_ONLY_IF_NO_OBSTACLES = 3

The pathfinding algorithm will avoid using diagonals if any obstacle has been placed around the neighboring cells of the specific path segment.

DiagonalMode