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

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

doi:10.24963/ijcai.2026/844

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

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. AI and Robotics. Pages 7592-7600.

https://doi.org/10.24963/ijcai.2026/844

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

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.

Huidong Liu, Jiarui Dou, Jiangshan Ai, Enwen Hu, Xianlei Long, Mingyan Li, Chao Chen, and Fuqiang Gu. 2026. Disturbance-Aware Hybrid Learning for Robust and Adaptive UAV Flight in Extreme Winds. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 7592–7600. https://doi.org/10.24963/ijcai.2026/844

@inproceedings{ijcai2026p844,
  title     = {Disturbance-Aware Hybrid Learning for Robust and Adaptive UAV Flight in Extreme Winds},
  author    = {Liu, Huidong and Dou, Jiarui and Ai, Jiangshan and Hu, Enwen and Long, Xianlei and Li, Mingyan and Chen, Chao and Gu, Fuqiang},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {7592--7600},
  year      = {2026},
  month     = {8},
  note      = {AI and Robotics},
  doi       = {10.24963/ijcai.2026/844},
  url       = {https://doi.org/10.24963/ijcai.2026/844},
}

TY  - CPAPER
TI  - Disturbance-Aware Hybrid Learning for Robust and Adaptive UAV Flight in Extreme Winds
AU  - Liu, Huidong
AU  - Dou, Jiarui
AU  - Ai, Jiangshan
AU  - Hu, Enwen
AU  - Long, Xianlei
AU  - Li, Mingyan
AU  - Chen, Chao
AU  - Gu, Fuqiang
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 7592
EP  - 7600
DO  - 10.24963/ijcai.2026/844
UR  - https://doi.org/10.24963/ijcai.2026/844
ER  -