SSTG-Nav: Metric-Grounded Spatial-Semantic Topological Graphs for Reusable Object Navigation

2026-08-01

AI Summary

This work addresses the challenge of long-term, repeated semantic navigation for service robots, where reusable memory is lacking and mapping semantic goals to reachable locations remains difficult. To overcome this, the authors propose a metric-semantic memory system that constructs a task-agnostic spatial-semantic topological graph through a single environmental survey. This graph integrates multi-view semantic evidence while preserving spatially discriminative recovery poses. By combining metric anchoring with source-aware fusion mechanisms, the system enables efficient and robust repeated navigation and failure recovery. Experiments on HM3D-v2 demonstrate a geometric success rate of 99.4%, with SR/SPL scores of 0.926/0.586, and Top-3 success rate and SPL@3 reaching 0.975 and 0.601, respectively, validating the effectiveness of the pre-exploration paradigm for repeated semantic navigation.

Abstract

Service robots operating for months in the same homes, offices, and facilities should become more reliable with experience instead of searching familiar space from scratch for every request. Yet ObjectNav is predominantly formulated as one-shot exploration, leaving a central deployment challenge unresolved: recognizing an object does not identify a reachable place to stop, and one confident map error can terminate the task. We introduce SSTG-Nav, a reusable metric-semantic memory that turns a one-time survey into actionable object goals, consolidates evidence across viewpoints, and retains spatially distinct recovery standoffs. On 1,000 HM3D-v2 episodes across 36 scenes, our goal-independent topology achieves a 99.4% geometric success ceiling. Holding semantic responses fixed, metric grounding raises SR/SPL from 0.835/0.560 to 0.920/0.603, and source-aware fusion reaches 0.926/0.586. Fusion-aware Top-3 recovery raises Success@1/2/3 to 0.928/0.965/0.975 and reaches 0.601 SPL@3. Model, field-of-view, density, and corruption controls identify where these gains originate, and a ROS2/Nav2 realization demonstrates the complete reusable query-to-execution pipeline. Together, the results establish pre-exploration as a powerful practical regime for dependable, repeated semantic navigation.

Problem

• Object Navigation
• Reusable Memory
• Metric-Semantic Mapping
• Topological Graph
• Service Robots

Innovation

• Metric Grounding
• Spatial-Semantic Topological Graph
• Reusable Navigation Memory
• Source-Aware Fusion
• Object Goal Navigation