SAGE-Nav: Leveraging LLM Planning and Alignment Fusion for Hierarchical Scene Graph-Guided Navigation

2026-06-24

AI Summary

This work addresses the challenges of long-horizon reasoning and poor generalization to novel environments in object goal navigation by proposing a hierarchical navigation framework that decouples global planning from local control through the integration of large language model–driven semantic planning and dynamic scene graphs. The approach introduces a novel hierarchical scene graph encoder and a goal-aware alignment fusion network, which jointly leverage relational graph convolution, adaptive gating, and multimodal alignment mechanisms to enable efficient coordination between semantic reasoning and perceptual inputs. Evaluated on iTHOR and RoboTHOR, the method significantly improves navigation efficiency and zero-shot generalization while meeting low-latency deployment requirements.

Abstract

Object-Goal Navigation (ObjNav) requires embodied agents to autonomously locate specified targets using only egocentric visual observations. Existing monolithic methods struggle with long-horizon reasoning and generalize poorly to novel environments. To address these limitations, we propose SAGE-Nav, a novel hierarchical framework that integrates the reasoning capabilities of Large Language Models (LLMs) with dynamic scene graphs. Crucially, it decouples asynchronous global semantic planning from the high-frequency reactive control loop. The LLM serves as a global planner, decomposing abstract instructions into a sequence of semantically grounded waypoints. To translate these plans into dense multi-modal guidance, we design a Hierarchical Scene Graph Encoder (HSGE) that leverages relational graph convolutions to produce structure-aware embeddings preserving both semantic and spatial topology. Furthermore, we develop the Goal-aware Alignment-Fusion Network (GAFN) to dynamically fuse real-time perception with these structural priors. Using an adaptive gating mechanism with an explicit inductive bias, GAFN ensures robust visual-topological alignment for the low-level policy. Extensive evaluations in the i-THOR and RoboTHOR environments demonstrate that SAGE-Nav achieves state-of-the-art performance, delivering substantial gains in navigation efficiency and zero-shot generalization while maintaining the low control latency required for physical robotic deployment.

Problem

• Object-Goal Navigation
• long-horizon reasoning
• zero-shot generalization
• embodied agents
• novel environments

Innovation

• Hierarchical Navigation
• Large Language Models
• Scene Graph
• Alignment Fusion
• Zero-shot Generalization