aiqbench analysis
This research presents WAND, a reinforcement learning framework that enhances quadrotor navigation by explicitly estimating and compensating for time-varying wind disturbances using Temporal Convolutional Networks. By integrating disturbance estimation into the policy via a WindAdapter module, the system achieves significantly higher success rates in cluttered environments compared to standard feedforward methods. The study validates real-time onboard feasibility through both simulation and indoor fan-induced flight tests, addressing a critical gap in autonomous aerial robotics under partial observability.
WAND: Learning Robust Navigation under Complex Wind Disturbances and Dense Obstacles for Quadrotors
arXiv:2610.11809v1 Announce Type: new Abstract: Robust navigation in cluttered environments remains a fundamental challenge for quadrotors, particularly when strong wind disturbances arise, which perturb vehicle dynamics, limit control authority, and substantially increase collision risk. Existing learning-based navigation policies typically rely on obstacle perception and proprioceptive observations, requiring the policy to infer time-varying disturbance effects implicitly and thereby limiting robustness under partial observability. This paper proposes WAND (Wind-Aware Navigation with Disturbance Estimation), a reinforcement learning framework for navigation under time-varying wind disturbances in dense obstacle fields. Specifically, WAND estimates wind-induced disturbance acceleration from historical proprioceptive states using a Temporal Convolutional Network (TCN). This estimation is integrated into the policy via a zero-initialized residual module, \emph{WindAdapter}, while simultaneously providing feedforward compensation for low-level control. The dual use of the estimate couples disturbance-conditioned navigation with feedforward disturbance rejection. Across 12 wind-disturbed simulation settings, WAND improved the observed success rate by 8.3 percentage points on average relative to feedforward compensation alone. Controlled opposite-crosswind experiments further showed wind-direction-dependent trajectory adaptation. In indoor fan-induced flight tests, WAND succeeded in 18 of 20 trials, demonstrating the feasibility of real-time onboard navigation.
Abstract:Robust navigation in cluttered environments remains a fundamental challenge for quadrotors, particularly when strong wind disturbances arise, which perturb vehicle dynamics, limit control authority, and substantially increase collision risk. Existing learning-based navigation policies typically rely on obstacle perception and proprioceptive observations, requiring the policy to infer time-varying disturbance effects implicitly and thereby limiting robustness under partial observability. This paper proposes WAND (Wind-Aware Navigation with Disturbance Estimation), a reinforcement learning framework for navigation under time-varying wind disturbances in dense obstacle fields. Specifically, WAND estimates wind-induced disturbance acceleration from historical proprioceptive states using a Temporal Convolutional Network (TCN). This estimation is integrated into the policy via a zero-initialized residual module, \emph{WindAdapter}, while simultaneously providing feedforward compensation for low-level control. The dual use of the estimate couples disturbance-conditioned navigation with feedforward disturbance rejection. Across 12 wind-disturbed simulation settings, WAND improved the observed success rate by 8.3 percentage points on average relative to feedforward compensation alone. Controlled opposite-crosswind experiments further showed wind-direction-dependent trajectory adaptation. In indoor fan-induced flight tests, WAND succeeded in 18 of 20 trials, demonstrating the feasibility of real-time onboard navigation.
Robotics (cs.RO)
arXiv:2610.11809 [cs.RO]
(or arXiv:2610.11809v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.11809 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
IEEE Robotics and Automation Letters, vol. 11, no. 11, pp. 12392-12399, Nov. 2026
https://doi.org/10.1109/LRA.2026.3730214 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 295 words
Related coverage
Cooperative Multi-UAV Navigation in Complex Environments via Systematic Multi-Agent Deep Reinforcement Learning
arXiv · Jul 29, 2026
Nonlinear Model Predictive Control for Trajectory Tracking of Differentially Flat Fixed-Wing Aerial Systems
arXiv · Aug 24, 2026
Robust Co-design Optimisation for Agile Fixed-Wing UAVs
arXiv · Jul 22, 2026
MagServo: Uncertainty-Resilient Hierarchical Magnetic Servoing via Learned Latent Representations
arXiv · Oct 6, 2026