Saved in:
Bibliographic Details
Main Authors: Shen, Yutong, Liu, Hangxu, Liu, Penghui, Luo, Jiashuo, Zhang, Yongkang, Morvley, Rex, Jiang, Chen, Zhang, Jianwei, Zhang, Lei
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.08572
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918380202622976
author Shen, Yutong
Liu, Hangxu
Liu, Penghui
Luo, Jiashuo
Zhang, Yongkang
Morvley, Rex
Jiang, Chen
Zhang, Jianwei
Zhang, Lei
author_facet Shen, Yutong
Liu, Hangxu
Liu, Penghui
Luo, Jiashuo
Zhang, Yongkang
Morvley, Rex
Jiang, Chen
Zhang, Jianwei
Zhang, Lei
contents Learning natural, stable, and compositionally generalizable whole-body control policies for humanoid robots performing simultaneous locomotion and manipulation (loco-manipulation) remains a fundamental challenge in robotics. Existing reinforcement learning approaches typically rely on a single monolithic policy to acquire multiple skills, which often leads to cross-skill gradient interference and motion pattern conflicts in high-degree-of-freedom systems. As a result, generated behaviors frequently exhibit unnatural movements, limited stability, and poor generalization to complex task compositions. To address these limitations, we propose MetaWorld-X, a hierarchical world model framework for humanoid control. Guided by a divide-and-conquer principle, our method decomposes complex control problems into a set of specialized expert policies (Specialized Expert Policies, SEP). Each expert is trained under human motion priors through imitation-constrained reinforcement learning, introducing biomechanically consistent inductive biases that ensure natural and physically plausible motion generation. Building upon this foundation, we further develop an Intelligent Routing Mechanism (IRM) supervised by a Vision-Language Model (VLM), enabling semantic-driven expert composition. The VLM-guided router dynamically integrates expert policies according to high-level task semantics, facilitating compositional generalization and adaptive execution in multi-stage loco-manipulation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2603_08572
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MetaWorld-X: Hierarchical World Modeling via VLM-Orchestrated Experts for Humanoid Loco-Manipulation
Shen, Yutong
Liu, Hangxu
Liu, Penghui
Luo, Jiashuo
Zhang, Yongkang
Morvley, Rex
Jiang, Chen
Zhang, Jianwei
Zhang, Lei
Robotics
Artificial Intelligence
Learning natural, stable, and compositionally generalizable whole-body control policies for humanoid robots performing simultaneous locomotion and manipulation (loco-manipulation) remains a fundamental challenge in robotics. Existing reinforcement learning approaches typically rely on a single monolithic policy to acquire multiple skills, which often leads to cross-skill gradient interference and motion pattern conflicts in high-degree-of-freedom systems. As a result, generated behaviors frequently exhibit unnatural movements, limited stability, and poor generalization to complex task compositions. To address these limitations, we propose MetaWorld-X, a hierarchical world model framework for humanoid control. Guided by a divide-and-conquer principle, our method decomposes complex control problems into a set of specialized expert policies (Specialized Expert Policies, SEP). Each expert is trained under human motion priors through imitation-constrained reinforcement learning, introducing biomechanically consistent inductive biases that ensure natural and physically plausible motion generation. Building upon this foundation, we further develop an Intelligent Routing Mechanism (IRM) supervised by a Vision-Language Model (VLM), enabling semantic-driven expert composition. The VLM-guided router dynamically integrates expert policies according to high-level task semantics, facilitating compositional generalization and adaptive execution in multi-stage loco-manipulation tasks.
title MetaWorld-X: Hierarchical World Modeling via VLM-Orchestrated Experts for Humanoid Loco-Manipulation
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2603.08572