Pacman AI – Search Algorithms

This project was developed for the Artificial Intelligence course at Harokopio University of Athens – Dept. of Informatics and Telematics. It is based on the UC Berkeley CS188 Pacman AI framework, implementing and testing search algorithms and multi-agent search strategies for Pacman.

Description

Pacman is controlled by an AI agent that makes decisions based on:

• The distance to the nearest ghost
• The distance to the nearest food
• The current score (number of food pellets eaten)
• The remaining food

The main custom code is implemented in search.py, searchAgents.py and multiAgents.py.

---

Installation & Setup

1. Clone the repository

git clone https://github.com/AthosExarchou/pacman-ai.git
cd pacman-ai

2. Create and activate a Conda environment

conda create -n pacman-env python=3.6
conda activate pacman-env

3. Run the game

Example run using AlphaBetaAgent with custom evaluation:

python pacman.py -p AlphaBetaAgent -a evalFn=better,depth=2 -l smallClassic -k 2

pacman-ai/
│── pacman.py               # Main entry point
│── multiAgents.py          # Contains custom evaluation function
│── search.py               # Search algorithms
│── searchAgents.py         # Search agent definitions
│── util.py                 # Helper functions
│── game.py, ghostAgents.py # Core game logic
│── graphicsDisplay.py      # Graphics rendering
│── graphicsUtils.py
│── layouts/                # Maze layout files
│── README.md               # This file
│── .gitignore              # Ignored files

Testing & Project Commands

Run Pacman in a medium maze with BFS search:

python pacman.py -l mediumMaze -p SearchAgent -a fn=bfs

Run Pacman in a medium maze with A* search:

python pacman.py -l mediumMaze -p SearchAgent -a fn=astar,heuristic=manhattanHeuristic

Run minimax:

python pacman.py -p MinimaxAgent -a evalFn=better,depth=2 -l smallClassic -k 2

Run alpha-beta pruning:

python pacman.py -p AlphaBetaAgent -a evalFn=better,depth=2 -l smallClassic -k 2

If you want to run the UC Berkeley autograder tests:

python autograder.py --no-graphics

Or test a single question (e.g. question-6):

python autograder.py -q q6

---

Author

• Name: Exarchou Athos
• Student ID: it2022134
• Email: [email protected] or [email protected]

License

This project is based on UC Berkeley CS188 Pacman framework. Original materials available here.