A DNN Framework for Learning Lagrangian Drift With Uncertainty - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Applied Intelligence Année : 2023

A DNN Framework for Learning Lagrangian Drift With Uncertainty

Résumé

Reconstructions of Lagrangian drift, for example for objects lost at sea, are often uncertain due to unresolved physical phenomena within the data. Uncertainty is usually overcome by introducing stochasticity into the drift, but this approach requires specific assumptions for modelling uncertainty. We remove this constraint by presenting a purely data-driven framework for modelling probabilistic drift in flexible environments. Using ocean circulation model simulations, we generate probabilistic trajectories of object location by simulating uncertainty in the initial object position. We train an emulator of probabilistic drift over one day given perfectly known velocities and observe good agreement with numerical simulations. Several loss functions are tested. Then, we strain our framework by training models where the input information is imperfect. On these harder scenarios, we observe reasonable predictions although the effects of data drift become noticeable when evaluating the models against unseen flow scenarios.
Fichier principal
Vignette du fichier
Journal_paper.pdf (1.55 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04516763 , version 1 (26-03-2024)

Identifiants

Citer

Joseph R. Jenkins, Adeline Paiement, Yann Ourmières, Julien Le Sommer, Jacques Verron, et al.. A DNN Framework for Learning Lagrangian Drift With Uncertainty. Applied Intelligence, 2023, 53 (20), pp.23729-23739. ⟨10.1007/s10489-023-04625-1⟩. ⟨hal-04516763⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More