Interleaved POMDP Planning for Multi-Object Search in Unknown Multi-Room Household Environments

2026-07-11

AI Summary

This work addresses the challenge of multi-object search in unknown domestic environments, where object location uncertainty and occlusion by obstacles hinder performance. Traditional POMDP approaches struggle to scale to large scenes. To overcome this, the authors propose Inter-POMDP, a novel hierarchical POMDP planning framework that interleaves high- and low-level reasoning. The high-level planner leverages LLM-enhanced object distribution histogram beliefs within a POUCT algorithm, while the low-level module employs obstacle-aware particle filtering to model navigation uncertainty and provide feedback for refining high-level decisions. By integrating large language model priors with particle-based belief representations, the method achieves both high planning quality and computational efficiency. Experiments demonstrate significant improvements over baselines, with up to 63% fewer collisions, and reductions of up to 35% in navigation steps and 32% in detection actions.

Abstract

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

Problem

• multi-object search
• unknown environments
• POMDP planning
• household environments
• uncertainty

Innovation

• Interleaved POMDP
• Multi-object search
• LLM-informed belief
• Obstacle-aware particle filter
• Hierarchical planning