My ROS2 Package

This ROS 2 package leverages third-party dependencies, which are managed via a thirdparty.repos file. Additionally, a Python virtual environment named py_deps is required for managing Python dependencies and allowing access to system packages.

Setup Instructions

Follow these steps to configure and build this package:

1. Clone and Import Third-Party Repositories

Ensure that the third-party dependencies are cloned into the thirdparty directory. Use the thirdparty.repos file to handle this:

mkdir -p my_workspace/src
cd my_workspace/src
git clone https://github.com/CoreSenseEU/cs4home_examples
vcs import < thirdparty.repos

2. Create and Configure the Python Virtual Environment

A virtual environment named py_deps is required. This environment will include system packages as well as specific third-party dependencies.

Steps:

1. Create the virtual environment with access to system packages:

   python3 -m venv --system-site-packages py_deps

2. Activate the virtual environment:

    source py_deps/bin/activate

3. Install the required packages in py_deps:

    pip install -r thirdparty/hri_face_detect/requirements.txt
    pip install -r thirdparty/yolov8_ros/requirements.txt

4. Add custom PYTHONPATH to the activate script:

   Edit py_deps/bin/activate to add the following line:

   export PYTHONPATH="${PYTHONPATH}:/path_to_your_venv_packages"

   Replace /path_to_your_venv_packages with the absolute path to your desired directory for custom Python packages.

3. Build the Package

With dependencies and environment set up, you can build your ROS 2 package:

colcon build --symlink-install

4. Running the face example

The face example node is a simple implementation in which you convert a hri_msgs/msg/IdList from hri_face_detect package to a knowledge graph representation.

After building, source the environment and activate the virtual environment you can ron the example as:

source install/setup.bash
source py_deps/bin/activate

run hri_face example, then execute the node

ros2 launch coresense cs4home_simple_project face_example.launch.py

finally if you want to visualize the output graph you can

rqt --force-discover 

then navigate to knowledge_graph plugin

if everything went ok you will see something like this: