🎮🕹️👾 Created a Pacman simulation in Python as part of UC Berkeley’s Artificial Intelligence course. The project is divided into two parts: a basic agent using search algorithms, and a more advanced multiagent system.
In this part, the Pacman agent navigates mazes to reach specific locations and collect food efficiently. General search algorithms were implemented and applied to various Pacman scenarios.
Algorithms implemented:
- Depth First Search (DFS)
- Breadth First Search (BFS)
- Uniform Cost Search (UCS)
- A* Search
For detailed explanations, refer to the analytical instructions.
This part focuses on designing intelligent agents for the classic Pacman game, including ghosts.
Key algorithms implemented:
- Minimax
- Alpha-Beta Pruning
- Expectimax
Additionally, experimented with evaluation function design to enhance agent performance.
- Reflex Agent
- Minimax behavior and decision-making
- Alpha-Beta pruning optimizations
- Expectimax algorithm
- Evaluation function design and tuning
python autograder.py