Crazyflow: An Accurate, GPU-Accelerated, Differentiable Drone Simulator in JAX
Crazyflow: An Accurate, GPU-Accelerated, Differentiable Drone Simulator in JAX
Crazyflow is a new simulator for aerial robotics that combines high-fidelity dynamics with JAX for speed and differentiability. It supports both gradient-based and sampling-based methods, and enables in-flight reinforcement learning by training a recovery policy in under a second without domain randomization.
Crazyflow introduces a differentiable drone simulator in JAX that combines high-fidelity dynamics with GPU acceleration, enabling fast and accurate policy learning without domain randomization. It supports both gradient-based and sampling-based methods, and demonstrates in-flight reinforcement learning by training a recovery policy in under a second.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S2
- Data origin
- Synthetic dataOpen source S2
Reported numbers
Reported duration
0.37 seconds
View original evidence
g; we demonstrate this by throwing a physical drone into the air and training a recovery policy from scratch in 0.37 seconds, catching it before it hits the ground. By pushing accuracy, speed, and differentiability simultaneously, Crazyflow serves as an open-source resource for synthetic data generation, with emerging capabilities for large-scale parallelization for online, in-execution learning and optimization, ope
Open source S5
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
arXiv:2606.01478v3 Announce Type: replace Abstract: High-quality, large-scale synthetic data from simulations is becoming a cornerstone for pushing robots' capabilities. While aerial robotics simulators have evolved to support specialized needs such as fidelity, differentiability, and swarms independently, a unified platform that synthesizes data across all these domains is missing. In this work, we propose Crazyflow
Open source S2
rones on single CPUs and GPUs, and beyond on clusters. Real-world experiments show Crazyflow supports both analytical-gradient-based policy learning, achieving high-accuracy tracking without domain randomization, and sampling-based obstacle avoidance at speeds exceeding half a billion steps per second. Breaking the traditional train-then-deploy paradigm, Crazyflow's performance enables in-flight reinforcement learnin
Open source S4
g; we demonstrate this by throwing a physical drone into the air and training a recovery policy from scratch in 0.37 seconds, catching it before it hits the ground. By pushing accuracy, speed, and differentiability simultaneously, Crazyflow serves as an open-source resource for synthetic data generation, with emerging capabilities for large-scale parallelization for online, in-execution learning and optimization, ope
Open source S5
, a simulator designed to push the frontier of aerial robotics algorithm development, supporting model-based and data-driven methods, gradient-based and sampling-based approaches, single- and multi-agent systems, and state- and vision-based policy learning. Crazyflow combines high-fidelity dynamics with JAX for speed and differentiability, scaling to millions of environments or thousands of swarms with thousands of d
Open source S3
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Ask which physical-world evaluation and real-data validation the buyer requires for simulated data.
- Compare the reported units and scope before using these quantities in a budget. Recording hours, sensor-hours and trajectories are different measures.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org