GeoLoco: Leveraging 3D Geometric Priors from Visual Foundation Model for Robust RGB-Only Humanoid Locomotion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Yufei, Chen, Xieyuanli, Pan, Hainan, Shi, Chenghao, Chen, Yanjie, Huang, Kaihong, Zeng, Zhiwen, Lu, Huimin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908871945093120
author Liu, Yufei
Chen, Xieyuanli
Pan, Hainan
Shi, Chenghao
Chen, Yanjie
Huang, Kaihong
Zeng, Zhiwen
Lu, Huimin
author_facet Liu, Yufei
Chen, Xieyuanli
Pan, Hainan
Shi, Chenghao
Chen, Yanjie
Huang, Kaihong
Zeng, Zhiwen
Lu, Huimin
contents The prevailing paradigm of perceptive humanoid locomotion relies heavily on active depth sensors. However, this depth-centric approach fundamentally discards the rich semantic and dense appearance cues of the visual world, severing low-level control from the high-level reasoning essential for general embodied intelligence. While monocular RGB offers a ubiquitous, information-dense alternative, end-to-end reinforcement learning from raw 2D pixels suffers from extreme sample inefficiency and catastrophic sim-to-real collapse due to the inherent loss of geometric scale. To break this deadlock, we propose GeoLoco, a purely RGB-driven locomotion framework that conceptualizes monocular images as high-dimensional 3D latent representations by harnessing the powerful geometric priors of a frozen, scale-aware Visual Foundation Model (VFM). Rather than naive feature concatenation, we design a proprioceptive-query multi-head cross-attention mechanism that dynamically attends to task-critical topological features conditioned on the robot's real-time gait phase. Crucially, to prevent the policy from overfitting to superficial textures, we introduce a dual-head auxiliary learning scheme. This explicit regularization forces the high-dimensional latent space to strictly align with the physical terrain geometry, ensuring robust zero-shot sim-to-real transfer. Trained exclusively in simulation, GeoLoco achieves robust zero-shot transfer to the Unitree G1 humanoid and successfully negotiates challenging terrains.
format Preprint
id arxiv_https___arxiv_org_abs_2603_07624
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle GeoLoco: Leveraging 3D Geometric Priors from Visual Foundation Model for Robust RGB-Only Humanoid Locomotion
Liu, Yufei
Chen, Xieyuanli
Pan, Hainan
Shi, Chenghao
Chen, Yanjie
Huang, Kaihong
Zeng, Zhiwen
Lu, Huimin
Robotics
The prevailing paradigm of perceptive humanoid locomotion relies heavily on active depth sensors. However, this depth-centric approach fundamentally discards the rich semantic and dense appearance cues of the visual world, severing low-level control from the high-level reasoning essential for general embodied intelligence. While monocular RGB offers a ubiquitous, information-dense alternative, end-to-end reinforcement learning from raw 2D pixels suffers from extreme sample inefficiency and catastrophic sim-to-real collapse due to the inherent loss of geometric scale. To break this deadlock, we propose GeoLoco, a purely RGB-driven locomotion framework that conceptualizes monocular images as high-dimensional 3D latent representations by harnessing the powerful geometric priors of a frozen, scale-aware Visual Foundation Model (VFM). Rather than naive feature concatenation, we design a proprioceptive-query multi-head cross-attention mechanism that dynamically attends to task-critical topological features conditioned on the robot's real-time gait phase. Crucially, to prevent the policy from overfitting to superficial textures, we introduce a dual-head auxiliary learning scheme. This explicit regularization forces the high-dimensional latent space to strictly align with the physical terrain geometry, ensuring robust zero-shot sim-to-real transfer. Trained exclusively in simulation, GeoLoco achieves robust zero-shot transfer to the Unitree G1 humanoid and successfully negotiates challenging terrains.
title GeoLoco: Leveraging 3D Geometric Priors from Visual Foundation Model for Robust RGB-Only Humanoid Locomotion
topic Robotics
url https://arxiv.org/abs/2603.07624