The open stack for
smarter autonomous drones.
Astral builds the SDK, simulation path, mobile apps, and public datasets you need to train perception and evaluate autonomy honestly — including closed-loop benchmarks that keep VLMs honest about metric grounding.
What Astral ships today
Open AI drone software: runnable code, operator tools, public datasets, and simulation-first documentation — not vaporware hardware pages.
Open source stack
Python SDK, examples, and repositories you can run today — ArduPilot SITL for fast iteration, Isaac-class sim when you need full perception loops.
Mobile operator app
iOS and Android apps for real-world operation alongside the same autonomy stack we develop in the open.
Yonder dataset
A large drone-perspective indoor corpus on Hugging Face, built to train perception and to stress-test cross-simulator generalization.
Explore YonderSimulation guides
Concrete setup for SITL and Isaac-style workflows so your autonomy code is exercised the same way we evaluate research benchmarks.
Run in simulationResearch-grade autonomy
We publish on the metric gap in vision-language drone navigation, modular architectures that separate semantics from geometry, swarm sensing requirements at scale, and Yonder — a public dataset designed to expose when offline perception metrics mislead you in closed loop.
Read our research# Terminal 1 — ArduPilot SITL
sim_vehicle.py -v ArduCopter --console --map
# Terminal 2 — Astral SDK against SITL
pip install astral-sdk
export ASTRAL_SDK_SERIAL_PORT=tcp:127.0.0.1:5760
python -c "
import time, astral_sdk as drone
drone.takeoff(2.0)
time.sleep(5)
drone.land()
drone.disconnect()
"Built for developers and researchers
Install the SDK, point it at ArduPilot SITL, and iterate in minutes. When you are ready to stress perception and planning together, move to Isaac-class simulation — same API surface, harder realism.
Documentation
Guides and reference on web.astral.us
Simulation
SITL and Isaac-style workflows
GitHub
astral-sdk and related repos
Research
Papers, notes, and benchmarks
Mission Control in Your Pocket
Configure, deploy, monitor, and control real flights with the Astral mobile apps — alongside the same open SDK and simulation workflows we publish for developers.
Plan Your Mission
Design flight paths, set waypoints, and configure autonomous behaviors.
Simulate & Validate
Test your mission in our simulator before deploying to real hardware.
Deploy & Monitor
Launch your fleet and monitor real-time telemetry and video feeds.
From the blog
Practical guides and research-backed notes on AI drone autonomy, evaluation, and datasets.