# 面向稀疏人类指导的真实世界移动操作的分层残差策略学习

> 原文链接：https://robotics.hellobike.com/papers/2026-10-14-2610-20607
> 频道：papers | 更新日期：2026-10-14 | 标签：arxiv, cs.RO
> 核心摘要：HiRe-MoMa通过分层残差微调冻结基策略，并以稀疏人类干预对齐价值，在TIAGo++真机移动操作实验中验证。

# 面向稀疏人类指导的真实世界移动操作的分层残差策略学习

::: v-pre
> 原文链接：[Hierarchical Residual Policy Learning for Real-World Mobile Manipulation with Sparse Human Guidance](https://arxiv.org/abs/2610.20607)（arXiv 2610.20607 · 发布日期：2026-10-14）

**Hierarchical Residual Policy Learning for Real-World Mobile Manipulation with Sparse Human Guidance**

HiRe-MoMa通过分层残差微调冻结基策略，并以稀疏人类干预对齐价值，在TIAGo++真机移动操作实验中验证。

## 核心要点

- 真实世界MoMa微调受高维动作空间和异构模块复杂耦合影响，稳定性与样本效率具挑战。
- HiRe-MoMa在冻结基策略上学习策略无关的加性残差，并将残差分为移动底盘与机械臂头。
- 使用覆盖不足20%的稀疏人类干预重塑critic价值景观，引导自主探索并提升样本效率。
- TIAGo++真实实验显示，约30分钟在线交互即可持续提升任务表现，且不依赖基策略架构。

**验证范围与局限**：验证范围为TIAGo++平台上的真实世界移动操作实验。

## 英文摘要 (Abstract)

Mobile Manipulation (MoMa) is a cornerstone of autonomous robotic systems, yet real-world fine-tuning remains a formidable challenge because of high-dimensional action spaces and intricate coupling between heterogeneous modules. In this paper, we propose HiRe-MoMa (Hierarchical Residual Mobile Manipulation), a systematic framework for stable and sample-efficient real-world fine-tuning of complex MoMa systems. At its core, HiRe-MoMa adopts a Hierarchical Residual Architecture that fine-tunes a frozen base policy. A Policy-Agnostic Residual learns additive corrections atop the base policy, while Hierarchical Decoupling splits the residual into mobile-base and manipulator-arm heads with unidirectional conditioning, mitigating gradient interference and aligning corrections with the platform's physical hierarchy. To improve sample efficiency in data-expensive real-world settings, we propose Expert-Guided Value Alignment, which leverages sparse human interventions with under 20% training coverage to reshape the critic's value landscape, effectively guiding autonomous exploration through expert priors. Extensive real-world experiments on the TIAGo++ platform demonstrate that HiRe-MoMa consistently enhances task performance with only around 30 minutes of online interaction, regardless of the base policy architecture. To our knowledge, this is among the first frameworks to enable efficient on-robot fine-tuning for mobile manipulation in the real world, providing a practical paradigm for deploying complex robotic systems in dynamic, human-centric environments. Further details can be found on our project website: https://hire-moma.github.io/

作者：Zhefei Gong, Hefeifei Jiang, Zehao Wang, Jiachen Li
:::

## 原始参考来源 (Sources)

- [arXiv 2610.20607：Hierarchical Residual Policy Learning for Real-World Mobile Manipulation with Sparse Human Guidance](https://arxiv.org/abs/2610.20607)