Learning Agile Quadrotor Flight in the Real World

2026-02-10

AI Summary

This work addresses the limited generalization of existing learning-based quadrotor agile flight controllers, which rely heavily on high-fidelity simulation and system identification and often sacrifice agility to satisfy safety constraints, particularly in out-of-distribution real-world scenarios such as external disturbances or hardware degradation. To overcome these limitations, the authors propose an adaptive control framework that requires neither precise system identification nor offline simulation. The approach combines online residual learning to construct a hybrid dynamics model with a novel adaptive time scaling (ATS) mechanism that actively probes the physical limits of the platform. Integrated with a short-horizon, real-world-anchored backpropagation-through-time strategy (RASH-BPTT), the system rapidly enhances a base policy’s peak velocity from 1.9 m/s to 7.3 m/s within approximately 100 seconds, enabling reliable execution of high-agility maneuvers near actuator saturation.

Technology Category

• Multiagent Systems: Multiagent Learning
• Intelligent Robots: Behavior Learning & Control
• Search and Optimization: Learning to Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Learning-based controllers have achieved impressive performance in agile quadrotor flight but typically rely on massive training in simulation, necessitating accurate system identification for effective Sim2Real transfer. However, even with precise modeling, fixed policies remain susceptible to out-of-distribution scenarios, ranging from external aerodynamic disturbances to internal hardware degradation. To ensure safety under these evolving uncertainties, such controllers are forced to operate with conservative safety margins, inherently constraining their agility outside of controlled settings. While online adaptation offers a potential remedy, safely exploring physical limits remains a critical bottleneck due to data scarcity and safety risks. To bridge this gap, we propose a self-adaptive framework that eliminates the need for precise system identification or offline Sim2Real transfer. We introduce Adaptive Temporal Scaling (ATS) to actively explore platform physical limits, and employ online residual learning to augment a simple nominal model. {Based on the learned hybrid model, we further propose Real-world Anchored Short-horizon Backpropagation Through Time (RASH-BPTT) to achieve efficient and robust in-flight policy updates. Extensive experiments demonstrate that our quadrotor reliably executes agile maneuvers near actuator saturation limits. The system evolves a conservative base policy with a peak speed of 1.9 m/s to 7.3 m/s within approximately 100 seconds of flight time. These findings underscore that real-world adaptation serves not merely to compensate for modeling errors, but as a practical mechanism for sustained performance improvement in aggressive flight regimes.

Problem

• agile quadrotor flight
• out-of-distribution uncertainty
• real-world adaptation
• physical limits exploration
• Sim2Real transfer

Innovation

• Adaptive Temporal Scaling
• Online Residual Learning
• RASH-BPTT
• Sim2Real-free Adaptation
• Agile Quadrotor Flight