ROS ~ Robotics course's second project

Politecnico di Milano - Accademic Year 2021-2022

Team

• Elia Maggioni
• Flavio Renzi
• Jaskaran Singh

The project was implemented utilizing ROS Noetic on a Linux Ubuntu 20.04 machine.
Tested also with ROS Melodic on a Linux Ubuntu 18.04 machine.

Goals

• Write a launch files to create the map
• Write a launch file to perform amcl based localization
• Write a service to save an image with the map and the trajectory of the robot
• A node is needed to publish the odometry as a tf
• A node is needed to merge the two lasers

To see the complete requirements -> Project presentation

Package content

• bags: directory containing the supplied bags.
• launch: directory containing the launch-file that sets the parameter for gmapping, amcl and the merging node
• src: contains the nodes, writtn in c++
• srv: contains the file to generate the service needed to save the map
• map: containing the final map generated by gmap
• rviz: directory containing the config file to see the maping or the localization.
• script: contains a python node using openCV to process the image, publish a processed version and a png map

Nodes

• OdomeryBroadcaster: this node reads the provided odometry and broadcasts the tf transformation.
• PathBroadcaster: the node reads the poses provided by amcl and stores them in a vector inside the nav_msgs::Path message pubblished on topic /path
• CustomMapSaver: this node is a modified version of the original one² of the ros map server. It has the callback from the service to save the map along with the path computed by amcl.
• occupancy_grid_processing.py: reads the topic /map pubblished from gmapping and pubblish the processed map on topic /my_map

Launch files

• link.launch: set up the basic transformations for the robot and start the merging node¹
• gmap.launch: set up the parameter and launch the gmapping node which will pubblish the map and the transformation between map and odom
• amcl.launch: set up and run amcl for the locaization pubblishing the transformation between map and odom

TF Tree

Our tf tree is composed of six transformation frames (map, odom, base_footprint, base_link, laser_front and laser_rear) There are static transformations linking the base_footprint to base_link and base_link to laser_front and laser_rear. The odom to base_footprint transformation is pubblished from the OdometryBroadcaster node from the data on the bag. The transformation between map and odom is broadcasted by the localization part of the algorithms running

How To Run

In order to run the gmapping algorithm and acml some packages are needed

For ROS Melodic on a Linux Ubuntu 18.04 machine

$ sudo apt install ros-melodic-robot-localization ros-melodic-imu-tools ros-melodic-rviz-imu-plugin ros-melodic-stage* ros-melodic-mapviz* ros-melodic-gmapping ros-melodic-amcl ros-melodic-navigation ros-melodic-ira-laser-tools

For ROS Noetic on a Linux Ubuntu 20.04 machine

$ sudo apt install ros-noetic-robot-localization ros-noetic-imu-tools ros-noetic-rviz-imu-plugin ros-noetic-stage* ros-noetic-mapviz* ros-noetic-gmapping ros-noetic-amcl ros-noetic-navigation ros-noetic-ira-laser-tools

The provided commands will run all the nodes specified in the launch file and a precofigured rviz to look at the mapping.

$ roslaunch RoboticsProject_2 gmap.launch

The map can be saved with this command

$ rosrun map_server map_saver -f map

The postprocessed map can be saved with this command

$ rosrun map_server map_saver -f map map:=/my_map

This command will run the localization with acml and open rviz preconfigured.

$ roslaunch RoboticsProject_2 acml.launch

To save the map with the computed path use the following command.
This will save the map in the package directory with the provided name

$ rosservice call save_map CustomMap

Results

We generate the map using the first bag provided.

We use the generated map to implement localization using amcl on the other 2 bags saving the path of the robot.

bag 2	bag 3

Reference

• Node for join different laser scan^[1]
• Original map saver^[2]