VLFM: Vision-Language Frontier Maps for Zero-Shot Semantic Navigation

TL;DR

VLFM introduces a zero-shot semantic navigation method using vision-language models to identify and reach unseen objects in novel environments, achieving state-of-the-art results in simulated datasets and successful real-world deployment.

Contribution

The paper presents VLFM, a novel approach combining occupancy maps and vision-language models for zero-shot semantic navigation in unseen environments.

Findings

Achieves state-of-the-art SPL on Gibson, HM3D, and MP3D datasets.

Successfully deployed on real-world robot (Boston Dynamics Spot).

Demonstrates effective zero-shot navigation without prior environment knowledge.

Abstract

Understanding how humans leverage semantic knowledge to navigate unfamiliar environments and decide where to explore next is pivotal for developing robots capable of human-like search behaviors. We introduce a zero-shot navigation approach, Vision-Language Frontier Maps (VLFM), which is inspired by human reasoning and designed to navigate towards unseen semantic objects in novel environments. VLFM builds occupancy maps from depth observations to identify frontiers, and leverages RGB observations and a pre-trained vision-language model to generate a language-grounded value map. VLFM then uses this map to identify the most promising frontier to explore for finding an instance of a given target object category. We evaluate VLFM in photo-realistic environments from the Gibson, Habitat-Matterport 3D (HM3D), and Matterport 3D (MP3D) datasets within the Habitat simulator. Remarkably, VLFM achieves state-of-the-art results on all three datasets as measured by success weighted by path length (SPL) for the Object Goal Navigation task. Furthermore, we show that VLFM's zero-shot nature enables it to be readily deployed on real-world robots such as the Boston Dynamics Spot mobile manipulation platform. We deploy VLFM on Spot and demonstrate its capability to efficiently navigate to target objects within an office building in the real world, without any prior knowledge of the environment. The accomplishments of VLFM underscore the promising potential of vision-language models in advancing the field of semantic navigation. Videos of real-world deployment can be viewed at naoki.io/vlfm.