/VacuumAgentJava

Primary LanguageHTML

Vacuum Cleaner Agent

This project implements an intelligent agent for a vacuum cleaner simulation in Java. The agent’s task is to clean a room while avoiding obstacles and minimizing movement autonomously. The agent operates in a grid environment where it can perceive dirt, obstacles, and bumps, and performs actions such as moving forward, turning, and cleaning.

Intro

Agent Features:

  • Perception: The agent can sense dirt, obstacles in front, and if it has bumped into something.
  • Actions: The agent can move forward, turn left, turn right, suck dirt, and shut off once the room is clean.
  • Efficiency: The agent is designed to minimize unnecessary movements and only clean when necessary.

Features

Intro

  1. Grid Representation: The environment is represented as a grid, where:

    • X represents walls or obstacles,
    • * represents dirt,
    • The agent (A, >, <, V) represents the vacuum cleaner, with the symbol indicating the agent’s direction.

  2. User Interface:

    • A Graphical User Interface (GUI) built with Java Swing to display the grid and agent's movement in real time.
    • A "Next" Button allows users to progress the simulation step by step.
    • The GUI shows the Percept and Action taken by the agent at each step.
    • The final score and performance evaluation are displayed after the agent shuts off.

  3. Simulation:

    • The agent starts at position (1,1) facing north and cleans the room efficiently.
    • It cleans in a zig-zag traversal manner to ensure it covers the whole floor.
    • After cleaning, the agent returns to the start and shuts off.
    • The performance of the agent is evaluated based on moves, turns, bumps, and dirt cleaned.

Setup and Installation

  1. Clone the repository:

    git clone git@github.com:melaniefayne/VacuumAgentJava.git
cd VacuumAgentJava

  2. Prerequisites:

    • Java 8+ installed.
    • Java IDE (IntelliJ recommended)

  3. Running the Project:

    • Compile the Java files: 
       javac -d out src/vacworld/agent/*.java src/vacworld/actions/*.java src/vacworld/*.java
    • Run the simulation: 
      java -cp out vacworld.actions.VacuumWorld

How the Agent Works

  • Starting Position: The agent begins in the upper-left corner (1,1), always facing north.
  • Perception: At each step, the agent perceives its environment, including whether it’s on dirt, an obstacle in front, and if it bumped into an obstacle on the last move.
  • Action Selection: The agent uses the percepts to decide its next move, aiming to clean all dirt and avoid obstacles.
  • Completion: Once the room is clean, the agent returns to its starting position and shuts off.

Performance Metrics:

The agent is scored based on:

  • Moves: Minimizing unnecessary movement,
  • Turns: Reducing excessive turning,
  • Bumps: Avoiding bumping into obstacles,
  • Dirt left: Ensuring all dirt is cleaned before shutting off.

Final Score