How bulnabi-snu/rac2025 works
Project type and purpose
• This is a robotics/simulation full-stack application for the “Robot Aircraft Competition 2025.”
• It uses ROS2 (Humble) and PX4 SITL to simulate and control a VTOL aircraft in Gazebo, with onboard vision, gimbal control, mission planning, and custom ROS2 messages.
File structure
• Top-level:
• docker-related files and compose YAMLs for Intel/NVIDIA GPU setups
• assets/ and gazebo_assets/ for static images and SDF models
• src/ containing all ROS2 packages and custom scripts
• src/codes: one-off or historical/ad hoc scripts (gimbal, mission histories, USB monitoring, simple detection)
• src/custom_msgs: custom ROS2 message definitions (TargetLocation, VehicleState)
• src/vehicle_controller: ROS2 Python package for mission planning and low-level control
• src/vision_processing_nodes: ROS2 Python package for camera publishing and image-based detection
README for additional insights
• Environment specs (Ubuntu 22.04, ROS2 Humble, PX4 main).
• Mentions SITL and Gazebo integration (PX4_Gazebo Dockerfiles).
• Notes refactoring goals for vision/tracking code and parameter tweaks for flight.