Relationship-Aware Hierarchical 3D Scene Graph for Task Reasoning

2026-02-02

AI Summary

This work addresses the limitation of existing SLAM and semantic mapping approaches in capturing high-level abstractions and reasoning about object relationships, which hinders task-level understanding and decision-making for agents in 3D environments. To overcome this, the authors propose a hierarchical 3D scene graph representation that integrates open-vocabulary semantics, leveraging visual language models (VLMs) and large language models (LLMs) in a synergistic manner for scene graph construction and task-oriented reasoning. This is the first approach to enable open-set comprehension of object semantics and their relational context through joint VLM–LLM collaboration. The method is validated on a quadrupedal robot platform across diverse environments and tasks, demonstrating significant improvements in the agent’s high-level semantic understanding and interactive capabilities within complex 3D scenes.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Visual Reasoning & Symbolic Representations
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

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.

Problem

• 3D scene graph
• relational reasoning
• task reasoning
• hierarchical representation
• autonomous agents

Innovation

• 3D scene graph
• open-vocabulary representation
• Vision Language Model (VLM)
• Large Language Model (LLM)
• task reasoning