DORAEMON: Decentralized Ontology-aware Reliable Agent with Enhanced Memory Oriented Navigation

2025-05-28

AI Summary

Zero-shot adaptive navigation for domestic service robots in unknown environments faces challenges including insufficient coordination between path planning and scene understanding, unstructured memory representations, and weak task comprehension. This paper proposes Ventra-Dorsal, a biologically inspired dual-stream cognitive architecture that integrates vision-language models (VLMs) with embodied perceptual memory modeling to enable reliable map-free, pretraining-free navigation. We introduce a novel semantic-spatial hierarchical fusion mechanism and a dynamic topological mapping method, design the Nav-Ensurance safety assurance module, and propose AORI—a new metric for navigation intelligence. Evaluated on HM3D, MP3D, and GOAT datasets, our approach achieves state-of-the-art success rate (SR) and success-weighted path length (SPL), significantly outperforming existing zero-shot methods. Results demonstrate the framework’s effectiveness and robustness under zero-shot, map-free, and pretraining-free conditions.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Natural Language Processing: Safety and Robustness
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Adaptive navigation in unfamiliar environments is crucial for household service robots but remains challenging due to the need for both low-level path planning and high-level scene understanding. While recent vision-language model (VLM) based zero-shot approaches reduce dependence on prior maps and scene-specific training data, they face significant limitations: spatiotemporal discontinuity from discrete observations, unstructured memory representations, and insufficient task understanding leading to navigation failures. We propose DORAEMON (Decentralized Ontology-aware Reliable Agent with Enhanced Memory Oriented Navigation), a novel cognitive-inspired framework consisting of Ventral and Dorsal Streams that mimics human navigation capabilities. The Dorsal Stream implements the Hierarchical Semantic-Spatial Fusion and Topology Map to handle spatiotemporal discontinuities, while the Ventral Stream combines RAG-VLM and Policy-VLM to improve decision-making. Our approach also develops Nav-Ensurance to ensure navigation safety and efficiency. We evaluate DORAEMON on the HM3D, MP3D, and GOAT datasets, where it achieves state-of-the-art performance on both success rate (SR) and success weighted by path length (SPL) metrics, significantly outperforming existing methods. We also introduce a new evaluation metric (AORI) to assess navigation intelligence better. Comprehensive experiments demonstrate DORAEMON's effectiveness in zero-shot autonomous navigation without requiring prior map building or pre-training.

Problem

• Enables adaptive navigation in unfamiliar environments for robots
• Addresses spatiotemporal discontinuity and unstructured memory in navigation
• Improves decision-making and safety in zero-shot autonomous navigation

Innovation

• Hierarchical Semantic-Spatial Fusion for navigation
• RAG-VLM and Policy-VLM for decision-making
• Nav-Ensurance ensures safety and efficiency