---
title: "Disturbance-Aware Hybrid Learning for Robust and Adaptive UAV Flight in Extreme Winds"
authors:
  - "Huidong Liu"
  - "Jiarui Dou"
  - "Jiangshan Ai"
  - "Enwen Hu"
  - "Xianlei Long"
  - "Mingyan Li"
  - "Chao Chen"
  - "Fuqiang Gu"
doi: "10.24963/ijcai.2026/844"
url: "https://ijcai.org/proceedings/2026/844"
pdf: "https://ijcai.org/proceedings/2026/0844.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "7592-7600"
track: "AI and Robotics"
keywords:
  - "AIR: Robot control, planning, and execution with guarantees"
  - "Robot control, planning, and execution with guarantees: Safe and robust control under uncertainty"
  - "Safety, trustworthiness, generalizability, and evaluation: Theoretical and algorithmic guarantees on safety, robustness, and out-of-distribution generalization"
---

# Disturbance-Aware Hybrid Learning for Robust and Adaptive UAV Flight in Extreme Winds

Huidong Liu, Jiarui Dou, Jiangshan Ai, Enwen Hu, Xianlei Long, Mingyan Li, Chao Chen, Fuqiang Gu

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 7592-7600. DOI: [10.24963/ijcai.2026/844](https://doi.org/10.24963/ijcai.2026/844)

## Abstract

Safe and precise maneuvering of quadrotor unmanned aerial vehicles (UAVs) in high-speed wind environments remains a critical challenge. Wind disturbances are nonlinear, time-varying, and difficult to model, causing traditional controllers to struggle with perception and compensation, especially under unseen wind distributions. To address these limitations, we introduce WA-TD3, a data-driven control framework that enables real-time wind disturbance perception and adaptive compensation without dedicated wind sensors. WA-TD3 employs a deep residual network to extract wind characteristics from temporal patterns in state deviations, forming a dynamics residual-driven perception mechanism that implicitly models and compensates for unknown winds. This residual is integrated into a perception-augmented reinforcement learning architecture, providing the policy with enhanced state information for proactive disturbance-aware control. Extensive experiments on complex trajectories under varying wind intensities demonstrate that WA-TD3 consistently outperforms state-of-the-art methods, achieving over 62% improvement in tracking accuracy under strong winds.