发表机构
Georgia Institute of Technology; The Hong Kong University of Science and Technology; Ising AI; MIT; Notre Dame University（佐治亚理工学院; 香港科技大学; 伊辛人工智能公司; 麻省理工学院; 圣母大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究未知多房间家庭环境下多目标搜索问题，提出Inter-POMDP算法，通过高级POUCT规划器与低级运动规划器相互作用，平衡规划质量与效率，相比基线方法减少碰撞、导航步数及检测次数。
AI 中文摘要
在未知的家庭环境中进行多目标搜索需要在广泛的不确定性下进行规划，从未知的物体位置到有未观察到的障碍物的杂乱空间。部分可观测马尔可夫决策过程（POMDP）为这类问题提供了一个有原则的框架，但在大领域中仍然难以处理。我们提出了Inter-POMDP，一种新颖的交错POMDP规划算法，将这一挑战分解为两个相互作用的层次：一个高级POUCT规划器使用基于大语言模型的直方图信念对物体分布进行推理，而一个低级运动规划器使用障碍物感知粒子信念作为领域知识来建模导航不确定性，以指导高级POUCT。这种交错设计在未知的多房间环境中的大搜索空间下平衡了规划质量和效率。模拟和实际实验均表明，与基线方法相比，我们的Inter-POMDP算法将碰撞次数最多减少63%，导航步数最多减少35%，检测次数最多减少32%。完整视频见此https链接。
英文摘要
Multi-object search in unknown household environments requires planning under extensive uncertainty - from unknown object locations to cluttered spaces with unobserved obstacles. POMDPs offer a principled framework for such problems but remain intractable in large domains. We propose Inter-POMDP, a novel interleaved POMDP planning algorithm that decomposes this challenge into two interacting levels: a high-level POUCT planner reasons over object distributions using LLM-informed histogram beliefs, while a low-level motion planner models navigation uncertainty with obstacle-aware particle beliefs as domain knowledge to guide high-level POUCT. This interleaved design balances planning quality and efficiency despite the large search space across unknown multi-room environments. Both simulation and real-world experiments show that our Inter-POMDP algorithm reduces collision counts by up to 63%, navigation steps by up to 35%, and detection counts by up to 32% compared with baseline methods. Full videos are https://sites.google.com/view/inter-pomdp
