HTTP/2 200 server: GitHub.com content-type: text/html; charset=utf-8 last-modified: Sat, 19 Jul 2025 01:30:16 GMT access-control-allow-origin: * strict-transport-security: max-age=31556952 etag: W/"687af528-7d3f" expires: Sun, 28 Dec 2025 23:12:18 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: 1831:21D6A4:823962:920EDB:6951B6FA accept-ranges: bytes date: Sun, 28 Dec 2025 23:02:18 GMT via: 1.1 varnish age: 0 x-served-by: cache-bom-vanm7210075-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1766962938.339302,VS0,VE209 vary: Accept-Encoding x-fastly-request-id: 70c5632f3f5d404d365e0b28a5d409d513f84df7 content-length: 5259

NaVILA: Legged Robot Vision-Language-Action Model for Navigation

RSS 2025

An-Chieh Cheng^*, Yandong Ji^*, Zhaojing Yang^*, Zaitian Gongye, Xueyan Zou,

Jan Kautz, Erdem Bıyık, Hongxu Yin^†, Sifei Liu^† Xiaolong Wang^†,

UC San Diego, USC, NVIDIA

^*Equal Contribution, ordered alphabetically
^† Equal Advising

pdf arxiv Code (VLA) Code (Low-level) Benchmark

TL;DR

NaVILA is a two-level framework that combines VLAs with locomotion skills for navigation. It generates high-level language-based commands, while a real-time locomotion policy ensures obstacle avoidance.

🔥 Highlights

1. Strongly Generalizable VLA Model. We tamed a VLM model into a VLA model and trained it on diverse datasets, including real-world human video data, simulated indoor navigation data, and question-answering tasks. NaVILA outperform all methods that do not rely on simulator pre-trained waypoint predictors, even when those methods leverage additional sensors.
2. Vision-Based Legged Robot Policy. We introduce an end-to-end vsion-based locomotion policy trained without teacher-student distillation. By directly interacting with the environment using LiDAR during training, our approach significantly reduces the sim-to-real gap. Our policy ensures safety in challenging environments, such as near transparency surfaces, and excels at traversing rough terrain.
3. VLN-CE-Isaac Benchmark. We introduce a high-fidelity, physics-realistic benchmark for low-level robotic control. It provides a safe, scalable platform to evaluate navigation across diverse robots and scenarios, reducing real-world testing costs and risks.
4. Real-World Deployment. NaVILA demonstrates strong performance in challenging real-world environments with quadruped and humanoid robots, showcasing its generalization capabilities and robustness.
   

Real-world Results: Booster T1

Real-world Results: Unitree Go2

Real-world Results: Unitree G1

Learning from YouTube Human Touring Videos

Results: VLN-CE-Isaac NaVILA-Go2-Vision

Results: VLN-CE-Issac NaVILA-H1-Vision

Results: VLN-CE-Issac Vision vs Blind Policy

Results: R2R-CE (Habitat)

Results: RxR-CE (Habitat)

Citation

@inproceedings{cheng2024navila,
title = {NaVILA: Legged Robot Vision-Language-Action Model for Navigation},
    author = {Cheng, An-Chieh and Ji, Yandong and Yang, Zhaojing and Zou, Xueyan and Kautz, Jan and Biyik, Erdem and Yin,
    Hongxu and Liu, Sifei and Wang, Xiaolong},
    booktitle = {RSS},
    year = {2025},
}
  

Acknowledgement

We sincerely thank Chengjing Yuan for their support with hardware setup and 3D modeling. We also thank Xuxin Cheng and Jialong Li for their help in setting up the G1 robot, as well as Jiazhao Zhang and Yukang Chen for their valuable discussions.