TANGO: Humanoid Navigation in Cluttered Environments with a Whole-Body Vision-Language-Action Model

Forum topic · 小凯 · 2026-09-10

Summary

TANGO is a whole-body vision-language navigation framework for humanoid robots traversing cluttered indoor environments. Unlike conventional 2D path-planning approaches, TANGO addresses navigation as continuous, geometry-aware whole-body adaptation, coordinating arm placement, torso adjustment, and gait modulation for collision-free movement through complex 3D spaces. Given a natural-language instruction and egocentric RGB observations, the model directly predicts 29-DoF joint-space actions for downstream whole-body control. Training is done entirely in simulation by synthesizing diverse collision-free traversal behaviors through global path planning, kinematic whole-body motion generation, obstacle-aware motion editing, and reinforcement-learning-based tracking, providing dynamically feasible action supervision for learning language-conditioned whole-body policies. In extensive simulation experiments, TANGO achieves state-of-the-art vision-language navigation performance and outperforms strong modular baselines in scenes requiring obstacle negotiation. The framework was deployed zero-shot on a Unitree G1 humanoid robot, showing robust language-guided traversal in cluttered real-world scenes without any real-world navigation training data. Paper: arXiv 2609.09158, posted 2026-09-08, in cs.RO and cs.AI.

Overview

Research area: cs.RO, cs.AI
Authors: Anqi Li, Yuxin Chen, Zhaobo Li, Zhuo Cao, Junli Ren, Masayoshi Tomizuka, Dhruv Shah
Published: 2026-09-08
arXiv: 2609.09158

Abstract

We study the problem of navigating cluttered indoor environments with a humanoid robot. Unlike conventional methods that model navigation as a 2D path planning problem, humanoid traversal in cluttered environments requires continuous geometry-aware whole-body adaptation, including coordinated arm placement, torso adjustment, and gait modulation for collision-free movement through complex 3D spaces.

We introduce TANGO, the first whole-body vision-language navigation framework for language-conditioned humanoid traversal in cluttered environments. Given a natural-language instruction and egocentric RGB observations, TANGO directly predicts 29-DoF joint-space actions for downstream whole-body control.

We train TANGO entirely in simulation by synthesizing diverse collision-free traversal behaviors via:

global path planning,
kinematic whole-body motion generation,
obstacle-aware motion editing,
RL-based tracking.

This pipeline provides dynamically feasible action supervision for learning language-conditioned whole-body policies.

Results

In extensive simulation experiments, TANGO demonstrates state-of-the-art performance in vision-language navigation, outperforming strong modular baselines in challenging scenes requiring obstacle negotiation.
TANGO was deployed zero-shot on a Unitree G1 humanoid robot, achieving robust language-guided traversal in cluttered real-world scenes — without training on any real-world navigation data.

---

*Auto-collected on 2026-09-10 from arXiv:2609.09158.*