Lunar Lander AI using Deep Q-Learning & OpenAI's Gym

This repository contains the implementation of a deep Q-learning agent to solve the Lunar Lander environment from OpenAI Gym.

Table of Contents

• Introduction
• Getting Started
• Requirements
• Usage
• Results

Introduction

The Lunar Lander environment is a classic reinforcement learning problem where the goal is to safely land a spacecraft on the moon's surface. In this project, we use deep Q-learning to train an agent to learn optimal actions for successful landings.

Getting Started

To get started with this project, follow these steps:

1. Clone this repository:

   git clone https://github.com/Siddharth-2382/Lunar-Lander-AI.git
   cd Lunar-Lander-AI
   

2. Install the required dependencies. You can use a virtual environment if desired:

   pip3 install -r requirements.txt
   

Requirements

• Python 3.7+
• OpenAI Gym
• NumPy
• TensorFlow
• Other libraries

Usage

1. Open Lunar-Lander-Solved.ipynb using Jupyter.
2. Run the notebook cells to train the deep Q-learning agent and observe its performance.

Results

The trained agent demonstrates successful landings in the Lunar Lander environment. You can view a video of the agent's performance in the videos directory.

DEMO