Relationship-Aware Hierarchical 3D Scene Graph for Task Reasoning

TL;DR

This paper presents a hierarchical 3D scene graph framework that integrates open-vocabulary features and relational reasoning, enabling autonomous agents to better understand and interact with complex environments.

Contribution

It introduces an enhanced scene graph that combines VLM and LLM for semantic and relational reasoning across multiple abstraction levels.

Findings

Effective scene understanding in diverse environments

Improved task reasoning capabilities for robots

Successful deployment on quadruped robot

Abstract

Representing and understanding 3D environments in a structured manner is crucial for autonomous agents to navigate and reason about their surroundings. While traditional Simultaneous Localization and Mapping (SLAM) methods generate metric reconstructions and can be extended to metric-semantic mapping, they lack a higher level of abstraction and relational reasoning. To address this gap, 3D scene graphs have emerged as a powerful representation for capturing hierarchical structures and object relationships. In this work, we propose an enhanced hierarchical 3D scene graph that integrates open-vocabulary features across multiple abstraction levels and supports object-relational reasoning. Our approach leverages a Vision Language Model (VLM) to infer semantic relationships. Notably, we introduce a task reasoning module that combines Large Language Models (LLM) and a VLM to interpret the scene graph's semantic and relational information, enabling agents to reason about tasks and interact with their environment more intelligently. We validate our method by deploying it on a quadruped robot in multiple environments and tasks, highlighting its ability to reason about them.