Skip to content

Celestial-Descent is a reinforcement learning project designed to develop an intelligent agent capable of landing a spacecraft on the lunar surface. Leveraging the Gymnasium library and Stable Baselines3, this project provides an engaging platform for exploring the fundamentals of reinforcement learning (RL) in a simulated environment.

Notifications You must be signed in to change notification settings

ParagGhatage/Celestial-Descent

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Celestial-Descent

Lunar Navigator

Overview

Lunar Navigator is a reinforcement learning project that explores the dynamics of landing a spacecraft on the lunar surface. Using the Gymnasium library and Stable Baselines3, this project aims to develop an agent capable of safely navigating the challenges of lunar landing.

Objectives

  • Understanding Reinforcement Learning: This project serves as an introductory exploration into the principles of reinforcement learning (RL).
  • Agent Development: To create an intelligent agent that learns to land on the moon by iterating through its training environment.

Key Features

  • Environment Setup: Implementation of the lunar lander environment provided by Gymnasium.
  • Model Training: Utilization of Stable Baselines3 to train the agent over 10,000 iterations.
  • Exploration vs. Exploitation: Addressing the critical balance during training to optimize the agent’s landing performance.

Getting Started

Prerequisites

  • Python 3.x
  • Gymnasium
  • Stable Baselines3
  • NumPy

About

Celestial-Descent is a reinforcement learning project designed to develop an intelligent agent capable of landing a spacecraft on the lunar surface. Leveraging the Gymnasium library and Stable Baselines3, this project provides an engaging platform for exploring the fundamentals of reinforcement learning (RL) in a simulated environment.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published