Add GUI node with shared memory communication system

[mmmarinho]

This PR ports the GUI and shared memory system from sas_force_sensor_bota into sas_force_sensor, and creates a new Python node that bridges sas::ForceSensorClient to the GUI via shared memory.

This PR was created by an AI agent (OpenHands) on behalf of the user.

Changes

• Shared memory module (sas_force_sensor/shared_memory/)

  • common.py – shared memory field index map
  • server.py – ForceSensorSharedMemoryServer for writing wrench data
  • client.py – ForceSensorSharedMemoryClient for reading wrench data
  • Ported from sas_force_sensor_bota with updated import paths.

• GUI (sas_force_sensor/gui.py)

  • Live plots for force (fx, fy, fz) and torque (τx, τy, τz)
  • Vertical sliders for force/torque norm
  • Reads from ForceSensorSharedMemoryClient via shared memory.

• Node (sas_force_sensor/gui_node.py)

  • Creates an sas::ForceSensorClient to subscribe to wrench data.
  • Spins the ROS 2 node and writes force/torque readings to shared memory.
  • Launches the GUI in a separate process via multiprocessing.

• Entry point (scripts/sas_force_sensor_gui.py)

  • Run via ros2 run sas_force_sensor sas_force_sensor_gui.

• Docker (docker/Dockerfile, docker/compose.yml)

  • Added pyqtgraph, PyQt6, qdarktheme pip packages to Dockerfile.
  • Added sas_force_sensor_gui service to compose.yml with X11 forwarding.
  • Added X11 environment variables to the existing sas_force_sensor service.

Usage

# Run the GUI node (requires an active force sensor publisher on the same ROS domain)
docker compose up sas_force_sensor_gui

[mmmarinho]

Added Xvfb headless display support to the sas_force_sensor_gui service in compose.yml. The service now:

1. Installs xvfb (sudo apt-get install -y xvfb)
2. Starts a virtual framebuffer on display :99 (sudo /usr/bin/Xvfb :99 -screen 0 1280x1024x24 &)
3. Exports DISPLAY=:99 so PyQt6 renders headlessly.

This allows the GUI service to run without a real display — e.g. on GitHub Actions CI.

This comment was added by an AI agent (OpenHands) on behalf of the user.

[mmmarinho]

Replaced the qdarktheme pip install with an explicit pyqtdarktheme==2.1.0 install using --ignore-requires-python to work around Python version compatibility issues:

# https://github.com/bhowiebkr/laser-level-webcam/issues/28
RUN python3 -m pip install pyqtdarktheme==2.1.0 --ignore-requires-python --break-system-packages

This follows the workaround from laser-level-webcam#28.

This comment was added by an AI agent (OpenHands) on behalf of the user.

[mmmarinho]

👍