slam_toolbox

slam_toolbox is a package we use to implement the SLAM algorithm in roboboat :)

bryan's documentation

what is slam?

slam stands for Simultaneous Localization and Mapping… but what does that mean?! * localization: positioning yourself relative to a map without global positioning (like gps) * mapping: exploring surroundings and using the data to form a map * slam = localizing and mapping at the same time!

how to use slam_toolbox

• slam_toolbox subscribers to data from /tf and /scan
• /odom and /laser topics need to be linked in the /tf tree

installation

1. update your system sudo apt update && sudo apt upgrade
2. install slam_toolbox sudo apt install ros-humble-slam-toolbox
3. build your packages colcon build

now you can run slam_toolbox with the code below! ros2 launch slam_toolbox online_async_launch.py use_sim_time:=True * this command runs slam_toolbox in online asynchronous mode

configuration

slam_toolbox comes with template default config files you can copy and modify this code below will copy one of the template files to your current working directory cp /opt/ros/humble/share/slam_toolbox/config/mapper_params_online_async.yaml .

rviz and slam_toolbox

view slam_toolbox data

1. open rviz ros2 launch vrx_gazebo rviz.launch.py
2. add a laserscan node
3. profit?

slam_toolbox rviz plugin

1. open rviz ros2 launch vrx_gazebo rviz.launch.py
2. Menu bar (top left corner) -> Panels -> Add New Panel
3. You should now see the SlamToolboxPlugin panel in the list of panels!