Sokoban AI Project

Overview

Sokoban is a classic puzzle game originating from Japan. The player (or in our case, a robot) needs to push boxes to specific locations in a warehouse. This project aims to implement Artificial Intelligence to solve the Sokoban puzzle by maneuvering the robot efficiently and moving all boxes to their target destinations within the simulated environment.

Features

• Various Search Algorithms:
  • Breadth-First Search
  • Depth-First Search
  • Uniform Cost Search
  • Beam Search
  • Hill Climbing
  • A* Search
• Agents: Different types of agents that can be used to solve the puzzle.
  • Robot: Uses a simple pathfinding algorithm to move to the closest box, then pushes it to the closest target.
  • Wall: Blocks the path of the robot.
  • Floor: The robot can move freely on this tile.
  • Box: The robot can push this tile.
• World Simulation: An environment where agents interact, loadable from text files.
• Visualization: UI to visualize the pathfinding and box moving strategies.
• Customization: Different heuristic and search algorithm choices for the agent.

Installation

Prerequisites

• Python 3.x
• Pip (Python package manager)

Steps

1. Clone the repository:

   git clone https://github.com/juanjosegongi/sokoban.git

2. Navigate to the project directory:

   cd sokoban

3. Install dependencies:

   pip install -r requirements.txt

4. Run the project:

   python main.py

Usage

1. Load the World: Choose a pre-defined world or create your own using the provided examples.
2. Choose an Algorithm: Select the search algorithm and heuristic to be used by the agent.
3. Simulate: Watch the agent navigate and solve the puzzle, with paths and decisions visualized.

Documentation

Detailed documentation on the usage, modules, and other aspects of the project can be found here.

Testing

To run the tests, navigate to the project directory and run:

python -m unittest discover tests

Contributing

Please read CONTRIBUTING.md for details on our code of conduct, and the process for submitting pull requests to us.

License

This project is licensed under the MIT License - see the LICENSE.md file for details.

Acknowledgments

• Reference to the original game or any inspired project if applicable.
• Acknowledge the authors, contributors, and users.