A LUDO board-game simulator in Python3 with Q-learning players.
Game rules
- Always four players.
- A player must roll a 6 to enter the board.
- Rolling a 6 does not grant a new dice roll.
- Globe positions are safe positions.
- The start position outside each home is considered a globe position
- A player token landing on a single opponent token sends the opponent token home if it is not on a globe position. If the opponent token is on a globe position the player token itself is sent home.
- A player token landing on two or more opponent tokens sends the player token itself home.
- If a player token lands on one or more opponent tokens when entering the board, all opponent tokens are sent home.
- A player landing on a star is moved to the next star or directly to goal if landing on the last star.
- A player in goal cannot be moved.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
Installation of the project:
git clone https://github.com/martinandrovich/pyludo.git
cd pyludo
pip3 install -e .You can visualize a an example match by running:
python3 pyludo/examples/game_rng_visualize.pyThe state is a numpy array of shape (4 players, 4 tokens)
The game state is represented as integers in a 4x4 numpy state array (4 players each with 4 tokens). Such that state[i] contains the i'th player's tokens, and state[i][k] will be the state of the k'th token of player i.
Home state is -1, and goal is 99 for all players. The common area range from 0-51 relative to player 0. The end lane is 52-56 for player 0, 57-61 for player 1 and so forth. A LudoPlayer is always fed a relative state in the .play() method, where the player itself is player 0. Note that when you move from home into the common area, you enter at position 1, not 0.
Text.
Text.
No license has been decided yet.
- Rasmus Haugaard - original developer(s) of the pyludo project.
- Haukur Kristinsson - original developer(s) of the pyludo project.