# Rationale

This project aims to speed up the ideation-development-simulation-deployment cycle of PX4/ArduPilot-based applications.

## The Many Facets of the Sim2real Gap

The *sim2real gap* is a euphemism for robotic projects that work well on a developer's laptop but not so much in the field.
Aerial sim2real research often focuses on modeling and simulation of complex aerodynamics effects.

Nonetheless, at deployment time, an equally important component of *sim2real gap* arises from [system design](https://arxiv.org/abs/2510.20808)—in particular, software tooling and engineering, where the dynamic range between ["average and the best is 50-to-1, maybe 100-to-1"](https://www.youtube.com/watch?v=wTgQ2PBiz-g&t=35s).

This is the challenge—and good sport—of **vertical/full-stack integration** among:

- the **many frameworks** that go into drone autonomy (a physics engine to simulate drone dynamics, a rendering engine to generate realistic imagery, a GPU-accelerated machine learning runtime for perception, one or more inter-process and inter-thread communication middleware, the interface to the microcontroller and autopilot software performing state-estimation and low-level control, the SDKs of the deployed embedded systems, etc.)
- emulated **inter-robot communication** (in aerial systems, this is heavily affected by the actual flight plans and available RF hardware)

## Design Manifesto

- **Simplicity** (["simple is better than complex"](https://peps.python.org/pep-0020/), ["worse is better"](https://www.dreamsongs.com/RiseOfWorseIsBetter.html), and ["no fat software"](https://people.inf.ethz.ch/wirth/Articles/LeanSoftware.pdf) are the 3 slogans of the AAS)
- [おまかせ](https://dhh.dk/2012/rails-is-omakase.html) **end-to-end**ness (from camera frames, to autopilot uORB/MAVLink commands)
- **Recentness** (break and fix, rather than carrying technical [debt](https://c2.com/doc/oopsla92.html))
- **Deployment** focus
    - Clear, Dockerized split between aircraft, ground, and simulation software
    - ROS2 intra-companion board messaging
    - XRCE-DDS (PX4), MAVROS (ArduPilot) autopilot-to-companion board ROS2 bridge
    - GStreamer camera-to-companion board acquisition pipelines
    - Zenoh inter-vehicle ROS2 bridge, with networking over LAN (HITL) or emulated by `docker network` (SITL)
    - Dual network—in both SITL and HITL—to separate synthetic sensor data from inter-vehicle communication

## Presentations and Papers

- Open Robotics Aerial Robotics Meeting: ["On the System Design Reality Gap"](https://docs.google.com/presentation/d/1Sz0d7WPWNwgCM3Q49Nnq1mvkSwPyK_CCQ7viHxO_Ucs) (January 22, 2026)
- Paper: ["aerial-autonomy-stack—a Faster-than-real-time, Autopilot-agnostic, ROS2 Framework to Simulate and Deploy Perception-based Drones"](https://arxiv.org/abs/2602.07264) (2026) International Conference on Unmanned Aircraft Systems (ICUAS), Corfu, Greece
- Slides: [`aerial-autonomy-stack`](https://www.dropbox.com/scl/fi/2gzki29sqrd67k3g0ctbz/icuas2026.pptx?rlkey=chflf9twnk1n6dptfx2huxa7o&st=kglsygkk&dl=0) in ICUAS 2026 regular session on "UAS Testbeds" (June 17, 2026)

## Related Work

A summary of existing multi-drone flight stacks can be found in [Table II of this paper](https://arxiv.org/pdf/2303.18237). Notable ones are:

- *Universidad Politécnica de Madrid (UPM)*'s [`aerostack2`](https://github.com/aerostack2/aerostack2) (multicopter-only)
- *Czech Technical University in Prague (CTU)*'s [`mrs_uav_system`](https://github.com/ctu-mrs/mrs_uav_system) (multicopter-only)
- *Norwegian University of Science and Technology (NTNU)*'s [`unified_autonomy_stack`](https://github.com/ntnu-arl/unified_autonomy_stack) (multi-frame, high-level control-only)
- *Technische Universität (TU) Berlin*'s [`crazyswarm2`](https://github.com/IMRCLab/crazyswarm2) (indoor, crazyflie-only)
- *Peking University*'s [`XTDrone`](https://github.com/robin-shaun/XTDrone) (PX4-only)
- *Carnegie Mellon University (CMU)*'s [`AirStack`](https://github.com/castacks/AirStack) (marked "only for internal usage")

A summary of aerial robotics simulators can be found in [Table IV of this paper](https://arxiv.org/pdf/2311.02296), these include:

- *Norwegian University of Science and Technology (NTNU)*'s [`aerial_gym_simulator`](https://github.com/ntnu-arl/aerial_gym_simulator) (high-performance simulator for RL)
- *University of Lisbon - Instituto Superior Técnico (IST)*'s [`PegasusSimulator`](https://github.com/PegasusSimulator/PegasusSimulator) (IsaacSim PX4/ArduPilot SITL, multicopter-only)
- *University of Toronto (UofT)*'s [`gym-pybullet-drones`](https://github.com/utiasDSL/gym-pybullet-drones) (simple simulator for education, control, and RL)
- *University of Pennsylvania (UPenn)*'s [`RotorPy`](https://github.com/spencerfolk/rotorpy) (high-fidelity simulator for control)
- *UZH*'s [`flightmare`](https://github.com/uzh-rpg/flightmare), *ETH*'s [`RotorS`](https://github.com/ethz-asl/rotors_simulator), *NYU*'s [`RotorTM`](https://github.com/arplaboratory/RotorTM), *Microsoft*'s [`AirSim`](https://github.com/microsoft/AirSim), etc.

For even more resources, check out [`aerial_robotic_landscape`](https://github.com/ROS-Aerial/aerial_robotic_landscape).