Make sure to have successfully followed the instructions in BlueROV-Setup to install Linux Tegra with ConnectTech kernel modules and all of the required packages with CUDA support enabled.
- If you haven't already, create a ROS2 workspace folder (prefereably in the home directory).
- Clone this repository inside the ROS2 workspace.
cd ~/ros_ws
git clone --recurse-submodules git@github.com:KTH-DHSG/BlueROV-ROS-Modules.git- Check and install realsense-ros dependencies
sudo apt-get install python3-rosdep -y \
sudo rosdep init \
rosdep update \
rosdep install -i --from-path src --rosdistro $ROS_DISTRO --skip-keys=librealsense2 -y- Build the modules and source the environment
colcon build --cmake-args '-DBUILD_ACCELERATE_GPU_WITH_GLSL=ON'
source install/setup.bash- yolo_brov: Implementation of YOLO detection trained on BlueROV videos.
- bluerov2_control: A ROS2 package for running controllers with the PX4 BlueROV2 & Gazebo simulator.
There are two ways of adding a custom ROS2 package to this repository. The first is to crete a new branch and add the new code directly into the src/ directory and then create a pull request to have it merged into the main branch. The second (and recommended option) is to create a new branch and to add a git submodules using git submodule add where should be the url of the repository you want to add. IMPORTANT! If you want to add a repository that is not public, make sure to use the ssh option on GitHub; specifies the irectory where the submodules should be added - in this case we want the submodules to be located in src/ directory.
- realsense-ros: Submodule added from Intel RealSense2 GitHub repository.
- px4_msgs: PX4 message package for ROS2
- px4_ros_com: PX4 communications package for ROS2