This repository contains an implementation of a Markov Decision Process (MDP) for the classic Wumpus World game, utilizing Policy Iteration (PI) and Value Iteration (VI) to compute optimal policies and value functions. The project includes Python code, graphics, and tools to visualize or save value function heatmaps.
- Implementation of PI and VI algorithms for MDP optimization.
- Includes graphical interface using Pygame.
- Generates or saves value function heatmaps as images.
- Clone the repository:
git clone https://github.com/yourusername/Wumpus_MDP.git - Install dependencies:
pip install numpy matplotlib seaborn pygame - Run the main script:
python main.py
wumpus_mdp.py: Core MDP logic and value function plotting.wumpus_gui.py: Pygame-based graphical interface.main.py: Entry point to run the simulation.graphics/: Directory for game assets.