Skip to content

gabrieldrn/Java-Maze-Solver

Repository files navigation

Java-Maze-Solver

This program is a console-based maze solving in Java with BFS, DFS, A*.

Maze structure

In this implementation, Mazes consists in a matrix of Squares.

Here is the orthogonal reprensentation of a Maze:

o---> X [Columns]
|
v
Y [Lines]

Solve mazes

Firstly, see how to load a maze from a .txt file or create one directly from code.

Next, refer to "Use a solver" to begin solving when your Maze is all set.

Like any solver, you can also set your own shift order. This program use cardinals as reference and the default one is North-East-South-West.

Load a maze

You can load a maze from a .txt as an argument in Maze class constructor, like this:

Maze myMaze = new Maze("./path/to/my/maze/myMaze.txt");

Example in Main.java line 25

The file must be written within this form :

-----------------e-
--xxxxxxxxxxxx-----
-------------x-----
----s--------x-----
-------------x-----
-------------------
-------------------

The number of characters in the first line will be the maze number of columns and the number of lines... the number of lines.

Every character is a square of the maze and is read like this:

  • s : maze starting point
  • e : maze ending point
  • x : wall
  • Other ones (whatever they are) are empty squares.

Create a maze

It's also possible to create a maze directly from code by following those steps, but this is the most laborious method.

1. In Main.java, create a starting and a ending square:

Square start = new Square(1, 0, "S");
Square end = new Square(3, 4, "E");

Square constructor arguments: Line (int), Column (int), Identifier (String)

S stands for starting point and E for ending point.

2. Create a new Maze:

Maze myMaze = new Maze(6, 5, start, end);

Maze constructor arguments : Number of lines (int), Number of columns (int), Starting point (Square), Ending point (Square)

3. Create walls:

(If you don't want walls, you can skip this step)

Use the setMazeWall() method:

myMaze.setMazeWall(2, 0);
myMaze.setMazeWall(3, 0);
myMaze.setMazeWall(1, 2);
myMaze.setMazeWall(2, 2);
...

Arguments: Wall line pos (int), Wall column pos (int)

Use a solver

You have 3 solvers available, each one corresponding to one algorithm:

  • AStarSolver
  • BFS_Solver
  • DFS_Solver

Simply use

BFS_Solver bfsSolver = new BFS_Solver(myMaze);
DFS_Solver dfsSolver = new DFS_Solver(myMaze);
AStarSolver aStarSolver = new AStarSolver(myMaze, true);

System.out.println(bfsSolver.solve());
System.out.println(dfsSolver.solve());
System.out.println(aStarSolver.solve());

Note that AStarSolver takes one other boolean argument to specifies if you want your maze to be solved by A* with Manhattan or Euclidean heuristic.

Set to true to use Manhattan heuristic. Euclidean heuristic will be used otherwhise.

Changing the shift order

Shift order is defined with a char vertor of size 4, each letter corresponding to a cardinal

Here's an example to set the order West-East-North-South

char[] order = {'W', 'E', 'N', 'S'};
myMaze.setOrder(order);

Example of resolution : A* + Euclidean heuristic

Path: [2, 12](23.0)(End) <- [1, 12](23.0) <- [1, 11](22.414213562373096) <- [1, 10](22.23606797749979) <- [2, 10](21.0) <- [2, 9](21.0) <- [2, 8](21.0) <- [1, 8](20.12310562561766) <- [0, 8](19.47213595499958) <- [0, 7](19.385164807134505) <- [0, 6](19.32455532033676) <- [1, 6](18.08276253029822) <- [2, 6](17.0) <- [3, 6](16.08276253029822) <- [3, 5](16.071067811865476) <- [3, 4](16.06225774829855) <- [4, 4](15.246211251235321) <- [5, 4](14.54400374531753) <- [6, 4](13.94427190999916) <- [6, 3](13.848857801796104) <- [6, 2](13.770329614269007) <- [6, 1](13.704699910719626) <- [6, 0](13.649110640673518) <- [7, 0](13.0)(Start) 
Path length: 23
Number of nodes created: 112
Execution time: 0.001 seconds
     0   1   2   3   4   5   6   7   8   9   10  11  12
   ╔═══╤═══╤═══╤═══╤═══╤═══╤═══════════╤═══╤═══╤═══╤═══╗
0  ║   │   │   │   │   │   │ *   *   * │   │   │   │   ║
   ╟───┴───┼───┤   └───┴───┘   ┌───┐   ├───┼───┴───┴───╢
1  ║       │   │             * │   │ * │   │ *   *   * ║
   ╟───┐   ├───┼───┬───┬───┐   ├───┤   └───┘   ┌───┐   ║
2  ║   │   │   │   │   │   │ * │   │ *   *   * │   │ o ║
   ╟───┤   └───┼───┼───┴───┘   ├───┤   ┌───┬───┼───┼───╢
3  ║   │       │   │ *   *   * │   │   │   │   │   │   ║
   ╟───┼───┐   └───┘   ┌───┐   ├───┤   └───┴───┴───┼───╢
4  ║   │   │         * │   │   │   │               │   ║
   ╟───┼───┼───┬───┐   └───┘   ├───┼───┐   ┌───┬───┼───╢
5  ║   │   │   │   │ *         │   │   │   │   │   │   ║
   ╟───┴───┴───┴───┘   ┌───┬───┼───┼───┘   └───┴───┼───╢
6  ║ *   *   *   *   * │   │   │   │               │   ║
   ║   ┌───┐   ┌───┐   └───┴───┼───┤   ┌───┐       ├───╢
7  ║ S │   │   │   │           │   │   │   │       │   ║
   ╟───┼───┤   ├───┤   ┌───┐   └───┘   └───┘       └───╢
8  ║   │   │   │   │   │   │                           ║
   ╟───┴───┘   └───┘   └───┘   ┌───┬───┐   ┌───┬───┬───╢
9  ║                           │   │   │   │   │   │   ║
   ║   ┌───┬───┐   ┌───┬───┐   └───┴───┘   └───┴───┴───╢
10 ║   │   │   │   │   │   │                           ║
   ║   └───┼───┼───┼───┼───┤   ┌───┬───┬───┬───┬───┬───╢
11 ║       │   │   │   │   │   │   │   │   │   │   │   ║
   ╚═══════╧═══╧═══╧═══╧═══╧═══╧═══╧═══╧═══╧═══╧═══╧═══╝

Class diagram

alt image

About the author

My name is Gabriel, I'm a french student in Video Games Development in Canada