How to Model Your Crazyflie Brushless

2026-03-06

AI Summary

This work addresses the lack of a high-fidelity dynamics model for the Crazyflie Brushless nano quadrotor platform by constructing and systematically identifying an accurate dynamics model, and for the first time releasing an open-source, high-fidelity simulation environment for this platform. Leveraging this model, the authors employ domain randomization combined with reinforcement learning to train both an end-to-end neural network position controller and an acrobatic controller capable of executing two full backflips within a 1.8-meter vertical space. The learned policies are directly transferred to physical hardware without fine-tuning. Experimental results demonstrate the effectiveness of the model in enabling successful sim-to-real transfer for complex, agile maneuvers, while also providing a comprehensive evaluation of how domain randomization influences transfer performance. This study establishes a reproducible benchmark for high-agility control of micro aerial vehicles.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Application Domains: Mobility, Driving & Flight

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

The Crazyflie quadcopter is widely recognized as a leading platform for nano-quadcopter research. In early 2025, the Crazyflie Brushless was introduced, featuring brushless motors that provide around 50% more thrust compared to the brushed motors of its predecessor, the Crazyflie 2.1. This advancement has opened new opportunities for research in agile nano-quadcopter control. To support researchers utilizing this new platform, this work presents a dynamics model of the Crazyflie Brushless and identifies its key parameters. Through simulations and hardware analyses, we assess the accuracy of our model. We furthermore demonstrate its suitability for reinforcement learning applications by training an end-to-end neural network position controller and learning a backflip controller capable of executing two complete rotations with a vertical movement of just 1.8 meters. This showcases the model's ability to facilitate the learning of controllers and acrobatic maneuvers that successfully transfer from simulation to hardware. Utilizing this application, we investigate the impact of domain randomization on control performance, offering valuable insights into bridging the sim-to-real gap with the presented model. We have open-sourced the entire project, enabling users of the Crazyflie Brushless to swiftly implement and test their own controllers on an accurate simulation platform.

Problem

• Crazyflie Brushless
• dynamics modeling
• sim-to-real transfer
• nano-quadcopter
• reinforcement learning

Innovation

• dynamics modeling
• reinforcement learning
• sim-to-real transfer
• domain randomization
• nano-quadcopter