ROS2 Build - All Packages

Please make sure that the Core Packages build successfully beforehand.

YOLO_ROS Setup

CUDA must be installed on Ubuntu.
It is also possible to use the CPU without installing CUDA, but the recognition speed will be significantly slower.

The procedure is as follows:

1. If you want to use CUDA, install the NVIDIA driver and reboot.
   (For details, please adjust according to your PC, such as disabling Nouveau if necessary.)
2. Install yolo_ros. Please follow the official installation instructions as the primary reference. For reference, the verified steps are shown below.

   cd ~/ros2_ws/src
   git clone https://github.com/mgonzs13/yolo_ros.git
   cd yolo_ros
   git checkout ecdc718  # Tested commit
   pip3 install -r requirements.txt
   cd ~/ros2_ws
   rosdep install --from-paths src/yolo_ros --ignore-src -r -y
   colcon build --symlink-install --packages-skip sigverse_turtlebot3
   source ~/ros2_ws/install/setup.bash
   

3. If you want to use CUDA, modify config files.
   Please change the YOLO "device" parameter in the launch file from "cpu" to "cuda:0".
4. Download YOLO models

   mkdir ~/yolo_weights
   cd ~/yolo_weights
   wget https://github.com/ultralytics/assets/releases/download/v8.3.0/yolo11n.pt
   wget https://github.com/ultralytics/assets/releases/download/v8.3.0/yolo11n-pose.pt
   

Clone the Official TurtleBot3 Packages

cd ~/ros2_ws/src
git clone --filter=blob:none --sparse -b humble https://github.com/ROBOTIS-GIT/turtlebot3_manipulation.git
cd turtlebot3_manipulation
git checkout 91492d7  # Tested commit
git sparse-checkout set turtlebot3_manipulation_description

Build - All Packages

cd ~/ros2_ws
colcon build --symlink-install
source ~/ros2_ws/install/setup.bash