Unsupervised Domain Adaptation for Visual Navigation

TL;DR

This paper introduces an unsupervised domain adaptation technique that translates target domain images to source domain style, enabling effective transfer of visual navigation policies from simulation to real-world environments.

Contribution

It presents a novel image translation method aligned with learned navigation representations to facilitate simulation-to-real-world transfer without supervision.

Findings

Outperforms baseline methods in simulation tasks

Enables successful real-world policy transfer

Improves navigation performance across diverse environments

Abstract

Advances in visual navigation methods have led to intelligent embodied navigation agents capable of learning meaningful representations from raw RGB images and perform a wide variety of tasks involving structural and semantic reasoning. However, most learning-based navigation policies are trained and tested in simulation environments. In order for these policies to be practically useful, they need to be transferred to the real-world. In this paper, we propose an unsupervised domain adaptation method for visual navigation. Our method translates the images in the target domain to the source domain such that the translation is consistent with the representations learned by the navigation policy. The proposed method outperforms several baselines across two different navigation tasks in simulation. We further show that our method can be used to transfer the navigation policies learned in simulation to the real world.