Structured Observation Language for Efficient and Generalizable Vision-Language Navigation

TL;DR

SOL-Nav introduces a structured language approach to vision-language navigation, converting visual observations into compact descriptions to improve efficiency and generalization across environments.

Contribution

The paper presents SOL-Nav, a novel framework that transforms visual data into structured language, reducing model size and enhancing generalization in VLN tasks.

Findings

Achieves strong generalization to unseen environments.

Reduces model size and training data dependency.

Performs well on standard VLN benchmarks and real-world deployments.

Abstract

Vision-Language Navigation (VLN) requires an embodied agent to navigate complex environments by following natural language instructions, which typically demands tight fusion of visual and language modalities. Existing VLN methods often convert raw images into visual tokens or implicit features, requiring large-scale visual pre-training and suffering from poor generalization under environmental variations (e.g., lighting, texture). To address these issues, we propose SOL-Nav (Structured Observation Language for Navigation), a novel framework that translates egocentric visual observations into compact structured language descriptions for efficient and generalizable navigation. Specifically, we divide RGB-D images into a N*N grid, extract representative semantic, color, and depth information for each grid cell to form structured text, and concatenate this with the language instruction as pure language input to a pre-trained language model (PLM). Experimental results on standard VLN benchmarks (R2R, RxR) and real-world deployments demonstrate that SOL-Nav significantly reduces the model size and training data dependency, fully leverages the reasoning and representation capabilities of PLMs, and achieves strong generalization to unseen environments.