论文/arXiv 2610.11722·20 天前

从单智能体到多智能体集群：面向可组合多智能体人-物交互的层级框架

针对单智能体人-物交互策略难以拓展至多智能体协作任务的问题，提出层级框架将单智能体策略转化为可复用技能，在多类HOI任务上验证了有效性。

AI 整理

• 本体 SDK 与控制

原文链接：From Solo to Ensemble: A Hierarchical Framework for Composable Multi-Agent Human-Object Interaction（arXiv 2610.11722 · 发布日期：2026-10-08）

From Solo to Ensemble: A Hierarchical Framework for Composable Multi-Agent Human-Object Interaction

针对单智能体人-物交互策略难以拓展至多智能体协作任务的问题，提出层级框架将单智能体策略转化为可复用技能，在多类HOI任务上验证了有效性。

核心要点

• 现有多智能体人-物交互方法依赖特定任务微调，高低层逻辑耦合，难以跨不同场景复用。
• 提出层级框架，将单智能体HOI策略蒸馏为可复用的面向对象运动技能，降低任务适配难度。
• 下游任务中冻结底层技能，高层策略基于共享对象等信息生成动作，协调多智能体执行。
• 实验验证该框架支持稳定的代理运动执行，可跨多类场景实现可组合策略学习。

验证范围与局限：仅在仿真多类人-物交互任务上完成验证，未开展真机实验。

英文摘要 (Abstract)

Physics-based human-object interaction has achieved robust single-agent manipulation skills, yet extending them to multi-agent cooperative tasks remains challenging. Existing approaches typically adapt interaction policies through task-specific fine-tuning, which entangles low-level contact-rich execution with high-level coordination and limits reuse across object geometries, interaction types, and team sizes. We propose a hierarchical framework that converts a single-agent HOI policy into a reusable Object-oriented Motion Skill. Specifically, we reinterpret teacher rollouts as object-oriented action supervision by extracting short-horizon object-proxy motions from executed trajectories, and distill task-specific teachers into a low-level skill operating in an Object-oriented Action Space. For downstream tasks, the distilled skill is frozen as a reusable executor, while a high-level policy coordinates multiple agents by generating region-wise object-oriented actions conditioned on the shared object, task goal, agent states, and local manipulation regions. This formulation shifts multi-agent HOI learning from direct contact-rich full-body control to compact object-level proxy-motion coordination. Experiments on diverse HOI tasks show that the distilled Object-oriented Motion Skill supports robust proxy-motion execution and enables composable policy learning across different interaction types, object geometries, and team sizes.

作者：Zekai Deng, Kangyi Chen, Ye Shi, Jingya Wang

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.11722：From Solo to Ensemble: A Hierarchical Framework for Composable Multi-Agent Human-Object Interaction访问 2026-10-21