Memory-Augmented Reinforcement Learning for Image-Goal Navigation - INRIA - Institut National de Recherche en Informatique et en Automatique Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2021

Memory-Augmented Reinforcement Learning for Image-Goal Navigation

Résumé

In this work, we address the problem of image-goal navigation in the context of visually-realistic 3D environments. This task involves navigating to a location indicated by a target image in a previously unseen environment. Earlier attempts, including RL-based and SLAM-based approaches, have either shown poor generalization performance, or are heavily-reliant on pose/depth sensors. We present a novel method that leverages a cross-episode memory to learn to navigate. We first train a state-embedding network in a self-supervised fashion, and then use it to embed previously-visited states into a memory. In order to avoid overfitting, we propose to use data augmentation on the RGB input during training. We validate our approach through extensive evaluations, showing that our data-augmented memory-based model establishes a new state of the art on the image-goal navigation task in the challenging Gibson dataset. We obtain this competitive performance from RGB input only, without access to additional sensors such as position or depth.
Fichier principal
Vignette du fichier
arxiv_final.pdf (2.3 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03110875 , version 1 (14-01-2021)
hal-03110875 , version 2 (02-03-2022)
hal-03110875 , version 3 (12-09-2022)

Identifiants

Citer

Lina Mezghani, Sainbayar Sukhbaatar, Thibaut Lavril, Oleksandr Maksymets, Dhruv Batra, et al.. Memory-Augmented Reinforcement Learning for Image-Goal Navigation. 2021. ⟨hal-03110875v1⟩
295 Consultations
324 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More