为居家养老设计更公平的机器人助手，聚焦日常活动困难。
Towards Equitable Robotic Furnishing Agents for Aging-in-Place: ADL-Grounded Design Exploration
- 基于老年人日常活动访谈，设计能理解需求的家居机器人。
- 实测显示用户重视操作确认、可调速度与多模态反馈。
- 强调公平与透明，适合关注老年福祉的智能系统研究者。
在居家养老场景中，日常生活活动（ADL）中的微小困难会累积，导致疲劳、焦虑、自主性下降和安全风险，影响老年人福祉。本文主张，服务老年人的机器人应超越便利功能，转向公平、正义与责任的核心。我们对四位70至80岁老人进行了基于ADL的半结构化访谈，识别出找物/整理物品、服药、搬运物品等常见挑战，并提炼出减轻认知-体力负担的需求。据此提出一种家庭内机器家具代理概念，融合计算机视觉、生成式AI与大语言模型，实现自然语言交互、情境感知提醒、安全执行及用户中心的透明性。随后通过视频引导的后续访谈，发现参与者偏好操作前确认、行为可预测性、速度与自主性可调，以及多模态反馈，并提出公平性相关关切。最后，本文提出关于如何在真实家庭中评估与部署公平机器人福祉系统的开放问题。
原文摘要 · Abstract (English)
In aging-in-place contexts, small difficulties in Activities of Daily Living (ADL) can accumulate, affecting well-being through fatigue, anxiety, reduced autonomy, and safety risks. This position paper argues that robotics for older adult wellbeing must move beyond "convenience features" and centre equity, justice, and responsibility. We conducted ADL-grounded semi-structured interviews with four adults in their 70s-80s, identifying recurrent challenges (finding/ organising items, taking medication, and transporting objects) and deriving requirements to reduce compounded cognitive-physical burden. Based on these insights, we propose an in-home robotic furnishing-agent concept leveraging computer vision and generative AI and LLMs for natural-language interaction, context-aware reminders, safe actuation, and user-centred transparency. We then report video-stimulated follow-up interviews with the same participants, highlighting preferences for confirmation before actuation, predictability, adjustable speed/autonomy, and multimodal feedback, as well as equity-related concerns. We conclude with open questions on evaluating and deploying equitable robotic wellbeing systems in real homes.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。