Bridging the Indoor-Outdoor Gap: Vision-Centric Instruction-Guided Embodied Navigation for the Last Meters

2026-02-06

AI Summary

This work addresses the challenge of achieving seamless, instruction-driven navigation from outdoor to indoor environments without reliance on precise prior information—a capability lacking in existing embodied navigation methods. We formally define, for the first time, an outdoor-to-indoor vision-and-language navigation task that operates without external priors and propose a vision-centric, end-to-end framework that relies solely on egocentric visual inputs and natural language instructions to perform continuous navigation. Key contributions include the creation of the first open-source dataset for this task, the introduction of trajectory-conditioned video synthesis for scalable data generation, and the design of an image-prompt-based decision-making mechanism. Experimental results demonstrate that our approach significantly outperforms state-of-the-art baselines in terms of both success rate and path efficiency.

Technology Category

• Computer Vision: Language and Vision
• Planning, Routing, and Scheduling: Planning with Language Models
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Embodied navigation holds significant promise for real-world applications such as last-mile delivery. However, most existing approaches are confined to either indoor or outdoor environments and rely heavily on strong assumptions, such as access to precise coordinate systems. While current outdoor methods can guide agents to the vicinity of a target using coarse-grained localization, they fail to enable fine-grained entry through specific building entrances, critically limiting their utility in practical deployment scenarios that require seamless outdoor-to-indoor transitions. To bridge this gap, we introduce a novel task: out-to-in prior-free instruction-driven embodied navigation. This formulation explicitly eliminates reliance on accurate external priors, requiring agents to navigate solely based on egocentric visual observations guided by instructions. To tackle this task, we propose a vision-centric embodied navigation framework that leverages image-based prompts to drive decision-making. Additionally, we present the first open-source dataset for this task, featuring a pipeline that integrates trajectory-conditioned video synthesis into the data generation process. Through extensive experiments, we demonstrate that our proposed method consistently outperforms state-of-the-art baselines across key metrics including success rate and path efficiency.

Problem

• embodied navigation
• indoor-outdoor transition
• instruction-guided
• last-meter delivery
• vision-centric

Innovation

• vision-centric navigation
• instruction-guided embodied navigation
• out-to-in navigation
• prior-free navigation
• trajectory-conditioned video synthesis