Agile Fall Recovery for Quadrotors with Bidirectional Thrust via Reinforcement Learning
2026-06-15 • Robotics
Robotics
AI summaryⓘ
The authors developed a way for quadrotor drones to get back into stable flight after falling on the ground in any position. They used a type of machine learning called reinforcement learning combined with a special control method, training the drone in simulations that closely mimic real conditions. Their method works well even when the drone’s sensors are limited or sometimes give bad data. They tested this approach in real life and showed that drones can recover quickly without needing full knowledge of their exact position or state.
quadrotorfall recoveryreinforcement learningpartial observabilityactor-critic architectureIncremental Nonlinear Dynamic Inversion (INDI)sim-to-real transferoptical flowonboard sensingzero-shot transfer
Authors
Anke Zhao, Yuhang Zhong, Kenghou Hoi, Junyu Mou, Junjie Wang, Lijie Wang, Jialiang Hou, Fei Gao
Abstract
Autonomous fall recovery is a critical capability for quadrotors operating in real-world environments, where collisions or failures may leave the vehicle resting on the ground in an arbitrary attitude. This problem is challenging because recovery must be achieved under limited onboard sensing, in constrained free space, with ground contact, and in the presence of unknown disturbances. In this letter, we present an RL-based framework for autonomous fall recovery of a quadrotor from arbitrary ground attitudes to stable hover using only lightweight onboard sensors. To address severe partial observability and intermittent sensor invalidity, we train a recurrent policy within an asymmetric actor--critic architecture, leveraging an Incremental Nonlinear Dynamic Inversion (INDI) controller to track the policy output. Combined with high-fidelity simulations of motor response and optical flow, the overall training framework significantly reduces the sim-to-real gap. Simulation ablation studies validate the importance of the main design choices, while real-world experiments demonstrate zero-shot transfer and robust recovery under different initial attitudes, wind disturbances, and additional payloads. These results demonstrate that agile quadrotor fall recovery can be achieved without explicit state estimation using only limited and unreliable onboard sensing.