Quick Start (SITL)

This gets you from zero to a drone flying in simulation on a single machine. One simulated drone, one computer, no physical drone hardware required.

1. Clone

git clone https://github.com/ortegarod/drone-os.git ws_droneOS
cd ws_droneOS

2. Start PX4 Simulation

cd PX4-Autopilot
HEADLESS=1 make px4_sitl gz_x500

Wait for Ready for takeoff! in the terminal. This runs PX4 with Gazebo — a full drone simulator.

3. Start DroneOS

In a new terminal:

cd ws_droneOS
docker compose -f docker/dev/docker-compose.dev.yml up -d --build drone_core micro_agent

This starts:

• drone_core — the SDK node that exposes drone control services
• micro_agent — the DDS bridge between PX4 and ROS 2

4. Verify

docker logs -f drone_core_node

You should see services registering: /drone1/arm, /drone1/takeoff, /drone1/set_position, etc. If you see those, you're good.

5. Fly

python3 drone_control.py --drone drone1 --set-offboard
python3 drone_control.py --drone drone1 --arm
python3 drone_control.py --drone drone1 --set-position 0 0 -50

The drone is now 50 meters up. The coordinate system is NED (North-East-Down), so negative Z = altitude. -50 means 50 meters above the takeoff point.

python3 drone_control.py --drone drone1 --set-position 80 -40 -50

Now it's flying 80m north and 40m west at 50m altitude.

python3 drone_control.py --drone drone1 --land

That's it. You just flew a drone.

6. Check State

python3 drone_control.py --drone drone1 --get-state

Returns full JSON: position, arming state, battery, nav mode.

7. Cleanup

docker compose -f docker/dev/docker-compose.dev.yml down

Then CTRL+C in the PX4 terminal.

Adding a Second Drone

In a new terminal:

cd PX4-Autopilot
HEADLESS=1 PX4_SYS_AUTOSTART=4001 PX4_GZ_MODEL_POSE="0,1" PX4_SIM_MODEL=gz_x500 MAV_SYS_ID=2 ./build/px4_sitl_default/bin/px4 -i 1

Then start a second drone_core instance inside the container, or see Multi-Drone Setup for the full setup.

python3 drone_control.py --fleet-status

Both drones show up. Control them independently with --drone drone1 or --drone drone2.

Next Steps

• drone_control.py Reference — full API
• System Architecture — how it all connects
• Real Hardware — deploy on a physical drone