aiqbench analysis
Navi-Agent advances embodied AI by enabling zero-shot navigation in continuous environments without relying on geometric localization or global coordinates. By constructing a coordinate-free spatial topology from visual inputs, the system achieves state-of-the-art performance in place confirmation and recovery tasks while remaining competitive with traditional geometry-based methods. This approach represents a significant step toward more robust, generalizable robotic agents capable of operating in unknown settings using only vision and motion history.
Navi-Agent: Unlocalized Monocular Navigation Agent
arXiv:2609.20388v1 Announce Type: new Abstract: Vision-Language Navigation in Continuous Environments (VLN-CE) requires an embodied agent to execute long-horizon instructions in unknown environments. Existing zero-shot VLN-CE systems typically maintain spatial states through geometric localization or coordinate-based representations. Recent geometry-constrained navigation removes depth and globally consistent coordinates, but maintaining persistent spatial awareness for place confirmation, progress verification, and recovery remains challenging. We present Navi-Agent, a zero-shot VLN-CE agent that constructs a coordinate-free spatial state from visual observations and executed motion histories. Navi-Agent organizes this state as a navigation topology, where nodes represent visual places and edges represent motion transitions. This representation enables observation-based approximate self-localization, task progress verification, and visual revisitation-based recovery. Navi-Agent performs closed-loop navigation by decomposing instructions into sub-goals, executing local visual navigation, and verifying visited places through the constructed spatial state. Experiments on zero-shot VLN-CE benchmark and real-world robot platforms show that Navi-Agent achieves state-of-the-art performance among geometry-constrained methods while remaining competitive with approaches relying on geometric localization.
Abstract:Vision-Language Navigation in Continuous Environments (VLN-CE) requires an embodied agent to execute long-horizon instructions in unknown environments. Existing zero-shot VLN-CE systems typically maintain spatial states through geometric localization or coordinate-based representations. Recent geometry-constrained navigation removes depth and globally consistent coordinates, but maintaining persistent spatial awareness for place confirmation, progress verification, and recovery remains challenging. We present Navi-Agent, a zero-shot VLN-CE agent that constructs a coordinate-free spatial state from visual observations and executed motion histories. Navi-Agent organizes this state as a navigation topology, where nodes represent visual places and edges represent motion transitions. This representation enables observation-based approximate self-localization, task progress verification, and visual revisitation-based recovery. Navi-Agent performs closed-loop navigation by decomposing instructions into sub-goals, executing local visual navigation, and verifying visited places through the constructed spatial state. Experiments on zero-shot VLN-CE benchmark and real-world robot platforms show that Navi-Agent achieves state-of-the-art performance among geometry-constrained methods while remaining competitive with approaches relying on geometric localization.
8 pages, 7 figures. Submitted to 2027 IEEE International Conference on Robotics & Automation (ICRA)
Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2609.20388 [cs.RO]
(or arXiv:2609.20388v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.20388 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 268 words
Related coverage
Embodied Agents Take Control: Minimal-Interface Zero-Shot Agents Rival Industrial-Scale Policies in Vision-and-Language Navigation
arXiv · Jul 30, 2026
ASENA: Self-evolving Agents for Embodied Navigation
arXiv · Oct 1, 2026
LG-VLN: A Zero-Shot Vision-and-Language Navigation Framework with LangGraph State Orchestration
arXiv · Sep 15, 2026
Mind the Style: Impact of Communication Style on Human-Chatbot Interaction
arXiv · Aug 24, 2026