Embracing Bulky Objects with Humanoid Robots: Whole-Body Manipulation with Reinforcement Learning

2025-09-16

AI Summary

To address the insufficient end-effector grasping stability and load capacity of humanoid robots when holding large, heavy objects, this paper proposes a reinforcement learning–based whole-body manipulation (WBM) framework. Methodologically, it integrates pretrained human motion priors with neural signed distance fields (NSDFs) to enable continuous geometric perception–driven multi-contact coordination planning; additionally, a teacher–student architecture transfers large-scale human motion data to support long-horizon, high-degree-of-freedom whole-body dynamic control. Evaluated in both simulation and on a physical humanoid robot, the method achieves stable, robust抱持 of diverse object shapes and sizes, significantly improving operational robustness and payload capacity. It further demonstrates strong sim-to-real transferability and cross-object generalization capability.

Technology Category

• Intelligent Robots: Manipulation
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Learning with Manifolds

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Whole-body manipulation (WBM) for humanoid robots presents a promising approach for executing embracing tasks involving bulky objects, where traditional grasping relying on end-effectors only remains limited in such scenarios due to inherent stability and payload constraints. This paper introduces a reinforcement learning framework that integrates a pre-trained human motion prior with a neural signed distance field (NSDF) representation to achieve robust whole-body embracing. Our method leverages a teacher-student architecture to distill large-scale human motion data, generating kinematically natural and physically feasible whole-body motion patterns. This facilitates coordinated control across the arms and torso, enabling stable multi-contact interactions that enhance the robustness in manipulation and also the load capacity. The embedded NSDF further provides accurate and continuous geometric perception, improving contact awareness throughout long-horizon tasks. We thoroughly evaluate the approach through comprehensive simulations and real-world experiments. The results demonstrate improved adaptability to diverse shapes and sizes of objects and also successful sim-to-real transfer. These indicate that the proposed framework offers an effective and practical solution for multi-contact and long-horizon WBM tasks of humanoid robots.

Problem

• Develop reinforcement learning for humanoid robot whole-body manipulation
• Enable stable multi-contact embracing of bulky objects
• Improve geometric perception and adaptability for diverse objects

Innovation

• Reinforcement learning with human motion prior
• Neural signed distance field for geometric perception
• Teacher-student architecture for motion distillation