multi-agentreinforcementlearning_turtlebots

https://github.com/gogiputtar/multi-agentreinforcementlearning_turtlebots

Science Score: 31.0%

This score indicates how likely this project is to be science-related based on various indicators:

✓
CITATION.cff file
Found CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
○
.zenodo.json file
○
DOI references
○
Academic publication links
○
Academic email domains
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (7.2%) to scientific vocabulary

Last synced: 10 months ago · JSON representation ·

Repository

Basic Info

Host: GitHub
Owner: GogiPuttar
License: apache-2.0
Language: Makefile
Default Branch: main
Size: 3.7 MB

Statistics

Stars: 0
Watchers: 1
Forks: 0
Open Issues: 0
Releases: 0

Created about 2 years ago · Last pushed about 2 years ago

Metadata Files

Readme License Citation

Multi-Agent Reinforcement Learning for Turtlebots

This repo contains ROS2 packages for Multi-Agent Reinforcement Learning (MARL) development on simulated and real Turtlebots. The main motivation behind this repo is to implement multi-agent exploration on turtlebots using RL. This includes a custom sim environment built from scratch and uses slam_toolbox for SLAM. Many of the packages here that handle the integration of simulation, sensing, controls etc. have been adapted from my other repo where I built all of this from scratch to implement my version of Extended Kalman-Filter feature-based SLAM on a real turtlebot: https://github.com/ME495-Navigation/slam-project-GogiPuttar

Package List

This repository consists of several ROS2 packages

General Purpose

nuturtle_description - 3D models of the nuturtle for simulation and visualization.
turtlelib - CMake Library with functions for control and SLAM of a differential drive robot

EKF SLAM specific

nusim - Loads the world consisting of the robot, arena and obstacles, and simulates the robot motion, control and sensors, akin to a real robot.
nuturtle_control - Handles the control and sensing related to the wheels of the robot
nuslam - Performs EKF SLAM using the turtlelbot

MARL specific

multisim - Generates and simulates exploration environments with multiple robots. (TODO)
multicontrol - Controls multiple robots. (TODO)
multislam - Uses slam_toolbox for multi-agent centralized SLAM. (TODO)
multiRL - Uses RL for multi-agent exploration. (TODO)

Owner

Name: Aditya Nair
Login: GogiPuttar
Kind: user
Location: Delhi, India

Repositories: 2
Profile: https://github.com/GogiPuttar

Why spend 10 minutes doing something you probably won't do again, when you can spend 10 hours automating it?

Citation (citations.txt)

# Citations

## Collaborators
- Abhishek Sankar
- Srikanth Schelbert
- Shail Dalal
- Ananya Agarwal

## External References
- Marno's repository: https://github.com/Marnonel6/EKF_SLAM_from_scratch
- ChatGPT Transcripts for:
    - SVG: https://chat.openai.com/share/98aea199-64d2-497b-acc7-c4eb0615ce84
    - Vector Normalization: https://chat.openai.com/share/a60709d3-73a2-4eea-8e58-33918814af88 

- Others: 
    - https://en.cppreference.com/book/intro/classes
    -

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Open Source Science