sKratchBOT, a mobile robot with a custom-made base using Kelo wheels and a Kinova 7-DOF Manipulator, developed for the b-it bots team for the RoboCup @Work League
- Minimun Ubuntu 22.04
- Ros humble (see installation guide)
sudo apt install -y ros-humble-gazebo-ros-pkgs gazebo libgazebo-dev python3-colcon-common-extensions ros-humble-xacro ros-humble-joint*mkdir -p ~/skratch_ws/src
cd ~/skratch_ws/srcgit clone -b dev_classic --single-branch https://github.com/b-it-bots/skratch_description.gitgit clone -b dev_classic --single-branch https://github.com/b-it-bots/skratch_simulation.gitNavigate to the workspace root and build:
cd ~/skratch_ws
colcon build --symlink-install --packages-select package_nameCurrently available packages:
colcon build --symlink-install --packages-select skratch_description skratch_gazeboSource the setup file and launch Gazebo:
source ~/skratch_ws/install/setup.bash
ros2 launch skratch_gazebo gazebo.launch.pyonce the robot is launched you can use teleop twist keyboard to control the roobt.
ros2 run teleop_twist_keyboard teleop_twist_keyboard If not already done, install following packages
sudo apt-get install ros-humble-slam-toolbox ros-humble-navigation2 ros-humble-nav2-bringup ros-humble-teleop-twist-keyboardRun mapping node, and move the robot around the
ros2 launch skratch_navigation online_async.launch.pycd ~/skratch_ws/src/skratch_simulation/skratch_navigation/maps/
ros2 run nav2_map_server map_saver_cli -f map_name-
Build docker image Choose the dockerfile based on your hardware spec, e.g.,
dockerfiles/Dockerfile.rocm(for AMD GPU),dockerfiles/Dockerfile.nvidia(for NVIDIA GPU), ordockerfiles/Dockerfile(for CPU only).docker build -t skratch_sim:latest -f dockerfiles/Dockerfile.rocm dockerfiles -
Run simulation in container
- Wayland display with AMD GPU
./sim-docker-amd-wayland.sh - For NVIDIA GPU:
# ToDo
- Wayland display with AMD GPU