CoorDex: Coordinating Body and Hand Priors for Continuous Dexterous Humanoid Loco-Manipulation

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Sikai Li, Shuning Li, Zhenyu Wei, Yunchao Yao, Chenran Li, Mingyu Ding
• Venue: arXiv: cs.RO
• Published: 2026-06-22
• Collected: 2026-06-23
• URL: http://arxiv.org/abs/2606.23680v1
• PDF: https://arxiv.org/pdf/2606.23680v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers introduce CoorDex, a learning pipeline that enables high-DoF dexterous loco-manipulation on the move for humanoid robots. The pipeline converts high-dimensional body and dexterous hand control into coordinated latent residual control, trained from simulated whole-body and hand demonstrations.

What the Document Says

• The paper introduces CoorDex, a learning pipeline for high-DoF dexterous loco-manipulation on moving humanoid robots.
• CoorDex converts body and dexterous hand control into coordinated latent residual control.
• Training is based on simulated whole-body and hand demonstrations.
• The pipeline uses privileged motion tracking teachers for the humanoid body and dexterous hand.

Key Concepts Mentioned

• CoorDex
• high-DoF dexterous loco-manipulation
• latent residual control
• privileged motion tracking teachers
• simulated whole-body and hand demonstrations

Claims Made by the Document

• CoorDex enables high-DoF dexterous loco-manipulation on the move for humanoid robots.
• The pipeline improves finger-level contact reliability.

Evidence Used

• Demos of non-stop bottle grasping, carrying, fridge door opening on the move, and cube pick-and-turn are provided.
• Ablations on the walk-grasp-carry task show the effectiveness of CoorDex.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Humanoid loco-manipulation is often simplified into a stop-and-go process: walking to an object, stopping to manipulate it, and then resuming locomotion. It also commonly relies on low degree-of-freedom (DoF) end effectors that behave like an open-close grasp primitive. We introduce CoorDex, a learning pipeline that converts high-dimensional body and dexterous hand control into coordinated latent residual control, enabling high-DoF dexterous loco-manipulation on the move. Starting from simulated whole-body and hand demonstrations, CoorDex trains privileged motion tracking teachers for the humanoid body and dexterous hand, distills them into proprioception-conditioned latent priors, and uses the frozen priors as the action space for downstream residual reinforcement learning. A coordinated latent residual policy composes these priors through shared task context and separate body-hand residual heads, preserving natural whole-body motion while improving finger-level contact reliability. CoorDex enables a Unitree G1 humanoid with a 20-DoF WUJI hand to execute dexterous manipulation while in motion, including non-stop bottle grasping and carrying, fridge door opening on the move, and cube pick-and-turn. Ablations on the walk-grasp-carry task show that joint-space PPO, joint-space hand control, and monolithic latent prediction all fail under the same reward budget, while the latent-prior interface and coordinated residual structure make high-dimensional contact-rich loco-manipulation trainable. Project Page: https://skevinci.github.io/coordex/