Skip to content
AI News HubLIVE
Original source2 min read

Bridging vision-language reasoning and learned control for legged navigation in the wild

Summary

ROAM combines a vision-language model's world knowledge with reinforcement-learned locomotion and planning policies, enabling legged robots to navigate kilometer-scale alpine and urban environments, interpret signs, and assess risks, though VLM reliability and spatial reasoning remain challenges.

SourcearXiv RoboticsAuthor: Takahiro Miki, Robin Schmid, Manthan Patel, Turcan Tuna, Julia Richter, Julian Nubert, Joonho Lee, Matthias M\"uller, Marco Hutter
Bridging vision-language reasoning and learned control for legged navigation in the wild
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 18 Nov 2026]

Title:Bridging vision-language reasoning and learned control for legged navigation in the wild

View a PDF of the paper titled Bridging vision-language reasoning and learned control for legged navigation in the wild, by Takahiro Miki and 8 other authors

View PDF HTML (experimental)

Abstract:Autonomous navigation in diverse real-world environments remains one of the greatest challenges in robotics. It demands both robust low-level motion control to handle diverse terrain and high-level reasoning to interpret environmental cues. We present ROAM (Robust Open-world Autonomous Mobility), a system that enables legged robots to perform kilometer-scale navigation in complex alpine and urban settings. ROAM combines the world knowledge of a Vision-Language Model (VLM), pre-trained on internet-scale data, with robust locomotion and local planning policies learned in simulation via reinforcement learning (RL). The VLM provides common-sense reasoning and semantic interpretation of the scene, assessing risks, interpreting signs, and producing semantic labels with associated navigation costs. An open-vocabulary segmentation module converts these labels into a dense, real-time risk map, which conditions two pre-trained RL policies responsible for agile locomotion and local navigation, effectively coupling high-level VLM reasoning with low-level physical control. Real-world experiments showed the system's ability to autonomously navigate a wide range of complex real-world environments. The robot completes kilometer-long hikes with minimal intervention, interprets signs to avoid danger, and evaluates its environment to identify safe areas on which to step. This work represents a step toward the reliable deployment of systems in unstructured environments. However, it also underscores the challenges that remain in VLM reliability and spatial reasoning, indicating that foundational models are a crucial component but not yet a complete solution for navigation in the wild.

Comments: 42 pages, 6 figures, 15 supplementary figures. Author's version of the article published in Science Robotics. Project page: this https URL

Subjects:

Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)

Cite as: arXiv:2611.25215 [cs.RO]

(or arXiv:2611.25215v1 [cs.RO] for this version)

https://doi.org/10.48550/arXiv.2611.25215

arXiv-issued DOI via DataCite (pending registration)

Journal reference: Science Robotics (2026)

Related DOI:

https://doi.org/10.1126/scirobotics.aee1549

DOI(s) linking to related resources

Submission history

From: Takahiro Miki [view email] [v1] Wed, 18 Nov 2026 21:40:00 UTC (14,630 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Bridging vision-language reasoning and learned control for legged navigation in the wild, by Takahiro Miki and 8 other authors

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.RO

new | recent | 2026-11

Change to browse by:

cs cs.AI cs.CV

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

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.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • ROAM couples VLM semantic reasoning with RL-trained locomotion and local planning policies.
  • An open-vocabulary segmentation module converts VLM labels into a dense real-time risk map that conditions two pre-trained RL policies.
  • Real-world tests show kilometer-long hikes, sign interpretation for danger avoidance, and safe footstep selection.
  • The work highlights VLM reliability and spatial reasoning as key remaining bottlenecks for foundation models in wild navigation.

Highlights and analysis are generated automatically and may contain errors. Check the original source.