mmWave ROS SLAM

Auhor: Nicola Pace

Organization: Politecnico di Bari

Email: [email protected]

---

Introduction:

This repository contains code to implement SLAM on ROS using multiple mmWave sensors.

This project can be used in two ways:

1. Building the source using catkin (the source currently misses some CMake files)
2. Using the turtlebot3 and pointcloud_to_laserscan source from apt repositories

Installation:

To implement the following source on top of turtlebot3 source, install turtlebot3 ROS packages:

sudo apt-get-install ros-noetic-turtlebot3-slam
sudo apt-get-install ros-noetic-turtlebot3-navigation

Also install pointcloud to laserscan package:

sudo apt-get-install ros-noetic-pointcloud-to-laserscan

Then clone this repository and build ira_laser_tools and ti_mmwave_rospkg from the ros_pkgs folder into your catkin_ws/src folder using catkin_make.

Now, copy config and launch files for SLAM and navigation inside turtlebot3 packages for simplicity. The following commands assumes ROS folder is: /opt/ros/noetic/share/

Inside the repository:

sudo cp -p ros_pkgs/poincloud_to_laserscan/launch/  /opt/ros/noetic/share/poincloud_to_laserscan/launch/ 
sudo cp -p ros_pkgs/mmwave_slam/launch/  /opt/ros/noetic/share/turtlebot3_slam/launch/ 
sudo cp -p ros_pkgs/mmwave_slam/config/  /opt/ros/noetic/shareturtlebot3_slam/config/ 
sudo cp -p ros_pkgs/mmwave_navigation/launch/  /opt/ros/noetic/shareturtlebot3_navigation/launch/ 
sudo cp -p ros_pkgs/mmwave_navigation/config/  /opt/ros/noetic/share/turtlebot3_navigation/config/ 

Now the workplace is correctly setupped.

How to use:

Bringup:

First, launch sensors nodes using ti_mmwave_rospkg

roslaunch ti_mmwave_rospkg sensors_back.launch
roslaunch ti_mmwave_rospkg sensors_front.launch
roslaunch ti_mmwave_rospkg sensors_LX.launch
roslaunch ti_mmwave_rospkg sensors_RX.launch

Now launch your odometry node, the publishes /tf and /odom for example:

roslaunch ros_to_canfd ros_to_canfd_odom.launch

Now run Pointcloud to Laserscan node, for example:

roslaunch pointcloud_to_laserscan 3point2laser.launch

And the Laserscan data merger node:

roslaunch ira_laser_tools 3laser_merger.launch 

To run laserscan filter:

python3 scripts/scan_cluster.py 

SLAM:

To launch SLAM node on unfliterd data, run:

roslaunch turtlebot3_slam my_gmapping.launch

To launch SLAM node on filtered data, run:

roslaunch turtlebot3_slam my_gmapping_filtered.launch

Autonomous navigation:

To launch navigation node:

roslaunch turtlebot3_navigation my_move_base.launch 

Rviz visualization:

You can use a custom view and urdf of your robot, and place it in urdf_tutorial pkg and use:

 roslaunch urdf_tutorial my_robot.launch

or to just view data:

 rviz ros_pkgs/rviz/config_final.rviz

Sources:

Sources utilizied in the code are:

1. ti_mmwave_rospkg
2. ira_laser_tools
3. pointcloud_to_laserscan