ATLAS Navigator: Active Task-driven LAnguage-embedded Gaussian Splatting

TL;DR

The paper introduces ATLAS Navigator, a hierarchical, language-embedded Gaussian splatting approach for real-time, task-driven robot navigation in unstructured environments, enabling semantic planning and dense geometric mapping for natural language tasks.

Contribution

It presents a novel hierarchical map representation combining language-embedded Gaussian splatting with semantic and geometric data for improved navigation.

Findings

Achieved about 60% success rate compared to privileged baselines.

Validated effectiveness in both indoor and outdoor real-world environments.

Demonstrated real-time operation with rich, task-relevant maps.

Abstract

We address the challenge of task-oriented navigation in unstructured and unknown environments, where robots must incrementally build and reason on rich, metric-semantic maps in real time. Since tasks may require clarification or re-specification, it is necessary for the information in the map to be rich enough to enable generalization across a wide range of tasks. To effectively execute tasks specified in natural language, we propose a hierarchical representation built on language-embedded Gaussian splatting that enables both sparse semantic planning that lends itself to online operation and dense geometric representation for collision-free navigation. We validate the effectiveness of our method through real-world robot experiments conducted in both cluttered indoor and kilometer-scale outdoor environments, with a competitive ratio of about 60% against privileged baselines. Experiment videos and more details can be found on our project page: https://atlasnav.github.io