Aerial Autonomy Stack
- Software
- Drones and boats
- Framework
A batteries-included, flight-tested open framework for developing, simulating, and deploying multi-drone autonomy with PX4, ArduPilot, Betaflight, and ROS2, including perception-based simulation and…
Source on GitHub: github.com/JacopoPan/aerial-autonomy-stack (opens in a new tab). Facts read at commit 31cc705 (opens in a new tab).
Licenses
- Code
- MIT (opens in a new tab)
Read by Robotheca from LICENSE (opens in a new tab) (MIT) at commit 31cc705 (opens in a new tab), on 22 Oct 2026.
- Commercial use
- Yes
What it does
- Simulates multi-vehicle quadrotors, quadplane VTOLs and tailsitters under PX4, ArduPilot, and Betaflight, with faster-than-real-time and multi-instance runs.
- Offers ROS2 action-based autopilot control via XRCE-DDS, MAVROS, or MSP, plus offboard controllers and behavior-tree missions.
- Provides perception in simulation: YOLO detection with ONNX GPU runtimes and LiDAR odometry via KISS-ICP, with 3D simulation worlds.
- Includes a steppable Gymnasium environment for reinforcement learning experiments.
- Runs in Docker images for simulation and deployment, with hardware-in-the-loop testing on Jetson boards.
- Supports inter-vehicle communication over a Zenoh ROS2 bridge and separate simulated-sensor and air networks.
Written by Robotheca from the project's README (opens in a new tab) at commit 31cc705.
Getting started
- Difficulty
- Advanced
- Hardware it needs, as the project states it
- Developed on Ubuntu 24.04 with an NVIDIA open driver (nvidia-driver-615-open) on an i7-11 with an RTX 3060; deployment is tested on a Holybro Jetson Baseboard with Pixhawk 6X and NVIDIA Orin NX 16GB. Windows 11 is supported via WSL.
- How to start
- The project's website (opens in a new tab)
Maintenance
- Status
- Active: a commit in the last 90 days
- Newest commit
- 22 Oct 2026 on main
- Latest release
- Betaflight Release (18 Oct 2026)
Checked on GitHub on 22 Oct 2026.
Links
Robots that can run it
No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.
Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.