Evaluating reinforcement learning-based neural controllers for quadcopter navigation in windy conditions

Vista sencilla de ítem
dc.contributor.authorAndrés Fernández, Alain
dc.contributor.authorMartínez, Aritz D.
dc.contributor.authorTunçay, Sümer
dc.contributor.authorCarlucho, Ignacio
dc.date.accessioned2025-09-30T09:44:43Z
dc.date.available2025-09-30T09:44:43Z
dc.date.issued2025-09-04
dc.date.updated2025-09-30T09:44:43Z
dc.description.abstractAccurate quadcopter navigation under windy conditions remains challenging for traditional control methods, especially in the presence of unpredictable wind gusts and strict navigational constraints. This paper evaluates Deep Reinforcement Learning (DRL) based controllers under such conditions, analysing the impact of wind domain randomisation, multi-goal training, enhanced state representations with explicit wind information, and the use of temporal data to capture affecting dynamics over time. Experiments in the AirSim simulator across four trajectories — evaluated under both no-wind and windy conditions — demonstrate that DRL-based controllers outperform classical methods, particularly under stochastic wind disturbances. Moreover, we show that training a DRL agent with domain randomisation improves robustness against wind but reduces efficiency in no-wind scenarios. However, incorporating wind information into the agent's state space enhances robustness without sacrificing performance in wind-free settings. Furthermore, training with stricter waypoint constraints emerges as the most effective strategy, leading to precise trajectories and improved generalisation to wind disturbances. To further interpret the learned policies, we apply Shapley Additive explanations analysis, revealing how different training configurations influence the agent's feature importance. These findings underscore the potential of DRL-based neural controllers for resilient autonomous aerial systems, highlighting the importance of structured training strategies, informed state representations, and explainability for real-world deployment.en
dc.description.sponsorshipAlain Andres and Aritz D. Martinez acknowledge support from FaRADAI project (ref. 101103386) funded by the European Commission under the European Defence Fund (EDF-2021-DIGIT-R)en
dc.identifier.citationAndres, A., Martinez, A. D., Tunçay, S., & Carlucho, I. (2025). Evaluating reinforcement learning-based neural controllers for quadcopter navigation in windy conditions. Engineering Applications of Artificial Intelligence, 161. https://doi.org/10.1016/J.ENGAPPAI.2025.112090
dc.identifier.doi10.1016/J.ENGAPPAI.2025.112090
dc.identifier.issn0952-1976
dc.identifier.urihttps://hdl.handle.net/20.500.14454/3784
dc.language.isoeng
dc.publisherElsevier Ltd
dc.rights© 2025 The Authors
dc.subject.otherDeep reinforcement learning
dc.subject.otherExplainability
dc.subject.otherProximal policy optimisation
dc.subject.otherShapley values
dc.subject.otherUnmanned aerial vehicle
dc.subject.otherWind
dc.titleEvaluating reinforcement learning-based neural controllers for quadcopter navigation in windy conditionsen
dc.typejournal article
dcterms.accessRightsopen access
oaire.citation.titleEngineering Applications of Artificial Intelligence
oaire.citation.volume161
oaire.licenseConditionhttps://creativecommons.org/licenses/by/4.0/
oaire.versionVoR
Archivos
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
andres_evaluating_2025.pdf
Tamaño:
11.95 MB
Formato:
Adobe Portable Document Format
Descargar
Colecciones
Artículos