Skip to content

ROS2 Packages for training a team of mobile robot to explore TRUE Unknown Environments (TUE) using Active SLAM with Deep Reinforcement Learning

Notifications You must be signed in to change notification settings

i1Cps/multi-agent-reinforcement-learning-active-slam

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 

Repository files navigation

multi-agent-reinforcement-active-slam

Logo

-----------------------------------------------------

➤ Table of Contents CURRENTLY BROKEN

-----------------------------------------------------

➤ 📝 About The Project


This repository is the culmination of advanced research into multi-agent reinforcement learning (MARL) applied within the context of active simultaneous localisation and mapping (SLAM). Developed during my time at the University of Bath, this project utilises ROS2 to orchestrate a team of autonomous agents, tasked with exploring and mapping unknown environments in a decentralised manner.

At its core, the project demonstrates the integration of advanced reinforcement learning techniques with real-world robotics applications. It underscores my ability to devise sophisticated algorithms adept at maneuvering the uncertainties of dynamic environments and enhancing the decision-making dynamics among multiple robots. This research is geared towards advancing autonomous exploration, contributing to the development of efficient and cooperative robotic systems.

This repository not only highlights my technical expertise but also provides a basis for further research and development in robotics and artificial intelligence.

-----------------------------------------------------

➤ 💾 Key Project File Description

Coming SOON

-----------------------------------------------------

➤ 🚀 Dependencies

This project is developed using; ROS2 and Gazebo for simulation and coordination of multiple robotic agents and Pytorch for Reinforcement Learning

ROS BadgePython Badge PyTorch Badge NumPy BadgeGazebo Badge

-----------------------------------------------------

➤ 🔨 Usage

Coming SOON!!

-----------------------------------------------------

➤ ☕ Buy me a coffee

Whether you use this project, have learned something from it, or just like it, please consider supporting it by buying me a coffee, so I can dedicate more time on open-source projects like this (҂⌣̀_⌣́)

Buy Me A Coffee

-----------------------------------------------------

➤ 📜 Credits

Theo Moore-Calters

GitHub Badge LinkedIn Badge

-----------------------------------------------------

➤ License

Licensed under MIT.

About

ROS2 Packages for training a team of mobile robot to explore TRUE Unknown Environments (TUE) using Active SLAM with Deep Reinforcement Learning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published