FOM-Nav: Frontier-Object Maps for Object Goal Navigation

TL;DR

FOM-Nav introduces a novel frontier-object map framework combined with vision-language models to improve object goal navigation efficiency, achieving state-of-the-art results in benchmarks and demonstrating real-world applicability.

Contribution

The paper presents FOM-Nav, a modular system that integrates frontier-object maps and multimodal scene understanding for enhanced navigation in unknown environments.

Findings

Achieves state-of-the-art SPL on MP3D and HM3D benchmarks.

Constructs large-scale datasets from real-world environments.

Demonstrates effective real-robot navigation performance.

Abstract

This paper addresses the Object Goal Navigation problem, where a robot must efficiently find a target object in an unknown environment. Existing implicit memory-based methods struggle with long-term memory retention and planning, while explicit map-based approaches lack rich semantic information. To address these challenges, we propose FOM-Nav, a modular framework that enhances exploration efficiency through Frontier-Object Maps and vision-language models. Our Frontier-Object Maps are built online and jointly encode spatial frontiers and fine-grained object information. Using this representation, a vision-language model performs multimodal scene understanding and high-level goal prediction, which is executed by a low-level planner for efficient trajectory generation. To train FOM-Nav, we automatically construct large-scale navigation datasets from real-world scanned environments. Extensive experiments validate the effectiveness of our model design and constructed dataset. FOM-Nav achieves state-of-the-art performance on the MP3D and HM3D benchmarks, particularly in navigation efficiency metric SPL, and yields promising results on a real robot.