Skip to content

gargivaidya/turtlebot_rl_gazebo

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

49 Commits
 
 
 
 
 
 

Repository files navigation

Reinforcement Learning with Turtlebot in Gazebo

TurtleBot3 with Intel RealSense D435

- [ROBOTIS e-Manual for TurtleBot3](http://turtlebot3.robotis.com/) - To install the Realsense Plugin from source follow steps from https://github.com/intel/gazebo-realsense, or as given below

Reinforcement Learning with Stable Baselines3

mkdir -p catkin_ws/src
cd catkin_ws/src
git clone https://github.com/gargivaidya/turtlebot_rl_gazebo.git
git clone https://github.com/ROBOTIS-GIT/turtlebot3.git
cd ..
catkin_make
. devel/setup.bash

Setup Virtual Environment for RL

python3 -m venv gymenv
source gymenv/bin/activate
pip3 install --upgrade pip
pip3 install pyyaml rospkg numpy gym matplotlib tensorboard scikit-build cmake stable-baselines3 scipy 
pip3 install <compatible version of torch https://pytorch.org/>

Start training RL model

roslaunch turtlebot3_gazebo turtlebot3_empty_world.launch

source gymenv/bin/activate
rosrun turtlebot3_gazebo sbtraining.py

Reinforcement Learning with Custom RL Algorithm

mkdir -p catkin_ws/src
cd catkin_ws/src
git clone https://github.com/gargivaidya/turtlebot_rl_gazebo.git
git clone https://github.com/ROBOTIS-GIT/turtlebot3.git
cd ..
catkin_make
. devel/setup.bash

Dependencies

turtlebot3
turtlebot3_msgs
turtlebot3_descriptions

Setup Virtual Environment for RL

python3 -m venv gymenv
source gymenv/bin/activate
pip3 install --upgrade pip
pip3 install pyyaml rospkg numpy gym matplotlib tensorboard scikit-build cmake stable-baselines3 scipy
pip3 install <compatible version of torch https://pytorch.org/>

Start training RL model

roslaunch turtlebot3_gazebo turtlebot3_empty_world.launch

source gymenv/bin/activate
rosrun turtlebot3_gazebo training.py

RealSense Camera Gazebo Plugin

  • This Gazebo plugin simulates a RealSense camera by publishing the 4 main RealSense streams: Depth, Infrared, Infrared2 and Color. It is associated to a RealSense model that is providade in ./models

Build

  1. Create a build folder under /src and make using CMAKE as follows:

    mkdir build
    cd build
    cmake ..
    make
    

Install

The plugin binaries will be installed so that Gazebo finds them. Also the needed models will be copied to the default gazebo models folder.

sudo make install

Launch Simulation with D435

roslaunch turtlebot3_gazebo turtlebot3_house.launch

Wiki for turtlebot3_simulations Packages

Open Source related to TurtleBot3

Documents and Videos related to TurtleBot3

About

Reinforcement Learning with Turtlebot in Gazebo

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published