Self Balancing Robot Simulation using ROS and Gazebo

Paper

DOI link

Video

The youtube link of videos are given in the notebook files. They can be accessed from PID, Fuzzy and LQR respectively.

In this project I have build a Self Balancing Robot Simulation in Gazebo. The files are structured as following

Notebook Folder

It has two jupyter notebook files . One contains the Description of Robot Model development named Self Balancing Robot.ipynb . The Controllers.ipynb describes the testing of the controllers, with Videos.

Video

The video folder inside the notebook folder contains the videos of the project

Src Folder

The $Src$ folder contains the source codes for the project. The file SelfBalance.py shows PID, SelfBalance_withFuzzy.py shows the Fuzzy P controller implementation and SelfBalance_withLQR.py shows the LQR implementation.

Launch Folder

The launch folder contains the launch files First download the github repository in your catkin workspace, and compile the project by writing catkin_make in the terminal. Then cd to the src folder of repository and write in the terminal chmod +x *.py" Then for PID

roslaunch self_balancing_robot main.launch

For Fuzzy Controller

roslaunch self_balancing_robot main_Fuzzy.launch

For LQR Controller

roslaunch self_balancing_robot main_LQR.launch

For any feedbacks and or suggestions , please contact sezan92@gmail.com

Update 29 July 2018

DQN

For Deep Q Network control roslaunch self_balancing_robot main_no_control.launch For running the code of DQN, go to Notebook/SelfBalanceDeepRLQ.ipynb and run

Paper link

Paper link

Video

Video link