brandonwagstaff/rob301_simulation

A Gazebo + ROS simulation environment for the virtual labs with the Turtlebot3 platform.

ROB 1514 Simulation

ROS package for the ROB 301 project simulation environment. Provides a Gazebo world in which the labs can be completed.

Install

Clone the repository into your catkin workspace:

git clone https://github.com/brandonwagstaff/rob301_simulation

You need to tell Gazebo where to find any custom models. Add the following line to your .bashrc:

export GAZEBO_MODEL_PATH=<path_to_catkin_ws>/src/rob301_simulation/models:$GAZEBO_MODEL_PATH

where you've replaced <path_to_catkin_ws> with the actual path.

Build and source the workspace as per usual ROS procedure.

Run

Once the workspace has been built and sourced, just run

roslaunch rob301_simulation lab_x.launch

to launch the world.

You can, of course, view all ROS topics using rostopic list. Topics for the front-facing camera are prefixed with /camera, while the topics for the side-facing camera are prefixed with /head_camera.

TODO:

• [] Update model from Turtlebot2 to Turtlebot3 - ideally we can set up a downward facing camera, and the 2D lidar.
• [] Finalize a strategy for providing students access to the virtual labs.
  • Set up a server with ROS that students can access via SSH: we will be using the Myhal PCs for this
  • [] Testing gzweb for gazebo visualization via a web browser
  • [] Testing other graphical ssh options like X11 forwarding
• Figure out how to add custom 'tape' patterns on the ground plane for path-following purposes
• [] Create environments for each lab:

  • [] Lab 1 - Turtlebot3 Intro:

    • [] Gazebo world: Very simple world needed - add a few features to test out the robot in simulation
    • [] Lab report update to introduce the Gazebo simulation

  • [] Lab 2 - Pose-to-Pose Control:

    • [] Gazebo world: add markers on floor to indicate starting and final poses
    • [] Updates to the lab report

  • [] Lab 3 - Path-Following (PID Control):

    • [] Gazebo world: create a practice world with smaller tracks for students to implement their controllers
    • [] Gazebo world: create a race world with the simulated racecourse - will need to think how this can be done - either students can practice on this as much as they would like, or we can withhold the course and they can send us code to run
    • [] Update the existing line detection node to work in simulation
    • [] Updates to the lab report

  • [] Lab 4 - Kalman Filtering:

    • [] Gazebo world: Add a pillar (or possibly even model a CN tower), and any other desired aesthetics
    • [] Implement the planar lidar node
    • [] Updates to the lab report

  • [] Project

Notes

• Tested with Gazebo 7.0 with ROS Kinetic on Ubuntu 16.04.