# COOL：用于个性化机器人辅助的好奇心驱动物体所有权学习

> 原文链接：https://robotics.hellobike.com/papers/2026-10-07-2610-09358
> 频道：papers | 更新日期：2026-10-07 | 标签：arxiv, cs.RO
> 核心摘要：COOL通过日常观察自主学习物体所有权并维护长期空间记忆，在离线与真实世界评估中验证个性化辅助。

# COOL：用于个性化机器人辅助的好奇心驱动物体所有权学习

::: v-pre
> 原文链接：[COOL: Curiosity-Driven Object Ownership Learning for Personalized Robotic Assistance](https://arxiv.org/abs/2610.09358)（arXiv 2610.09358 · 发布日期：2026-10-07）

**COOL: Curiosity-Driven Object Ownership Learning for Personalized Robotic Assistance**

COOL通过日常观察自主学习物体所有权并维护长期空间记忆，在离线与真实世界评估中验证个性化辅助。

## 核心要点

- 面向日常个性化服务，机器人需理解“我的物品”等指令并推理物体、人员、位置与所有权。
- COOL从长期人-物交互行为证据中自主推断所有权，并维护环境的长期空间记忆。
- 采用基于智能体的好奇心驱动数据收集，引导机器人前往有信息价值的位置并刷新过时观察。
- 离线实验、消融研究和真实世界评估表明，COOL可用于所有权条件导航与任务执行。

**验证范围与局限**：验证包括离线实验、消融研究和真实世界评估，任务范围为所有权推断、所有权条件导航与任务执行。

## 英文摘要 (Abstract)

Robots are increasingly expected to provide personalized services in everyday environments. To do so, they must ground natural-language commands such as "Where is my backpack?" or "Find my bottle" and execute them by reasoning about object instances, people, locations, and ownership. This is challenging because ownership is rarely labeled explicitly and must be inferred from long-term, behavioral evidence of human-object interactions. To address this, we present COOL, a novel robotic framework for autonomously learning object ownership from everyday observations and maintaining a long-term spatial memory of its environment. To keep its memory current, COOL uses an agent-based curiosity-driven data collection strategy that guides the robot toward the most promising locations to gain information and refresh stale observations. Offline experiments, ablation studies, and real-world evaluations show that COOL can infer ownership relations from real-world interactions and use this knowledge for ownership-conditioned navigation and task execution.

作者：Samira Huber, Ruben Hammele, Sören Pirk
:::

## 原始参考来源 (Sources)

- [arXiv 2610.09358：COOL: Curiosity-Driven Object Ownership Learning for Personalized Robotic Assistance](https://arxiv.org/abs/2610.09358)