TANGO: Whole-Body VLA Maps RGB to 29-DoF Actions for Humanoid Navigation, Trained Fully in Simulation
berkeley_ai · x · 2026-09-24
TANGO, a CoRL 2026 work, is a whole-body VLA for humanoid navigation that directly predicts 29-DoF joint actions from RGB observations, enabling coordinated arm, torso, and gait control to traverse cluttered 3D spaces. Trained entirely in simulation, it transfers zero-shot to the real world. Project page and paper are public.
Related event: Berkeley's TANGO Enables Whole-Body Humanoid Navigation(2 posts)→
More from Embodied
- RealSense and NVIDIA demo a local VLM describing a room 10x per second on Jetson Thor — chrismatthieu · 2026-09-24
- DAVIO combines Depth Anything 3 and IMU for real-time dense metric SLAM, code released — zhenjun_zhao · 2026-09-24
- Kairos extends 3D scene graphs to 4D to forecast pedestrian presence and directional flow — zhenjun_zhao · 2026-09-24
- VivixLabs demo shows real-time full-body live avatars driven by conversation — umesh_ai · 2026-09-24
- Meta Glasses' hearing-aid angle could shield them from bans via the ADA — RachelVT42 · 2026-09-24
- Mapping the UK robotics scene: from SLAM startups to Dexory's $165M rounds — lukas_m_ziegler · 2026-09-24