Leveraging Large Language Model-based Room-Object Relationships Knowledge for Enhancing Multimodal-Input Object Goal Navigation

TL;DR

This paper introduces a modular approach that leverages large language model-derived object-to-room knowledge to improve object-goal navigation efficiency in simulated and real environments.

Contribution

It presents a novel data-driven method integrating LLM-based knowledge with multimodal inputs for enhanced navigation performance.

Findings

Outperforms baseline by 10.6% in SPL metric

Effective in both simulated and real-world environments

Utilizes multi-channel Swin-Unet for multi-task learning

Abstract

Object-goal navigation is a crucial engineering task for the community of embodied navigation; it involves navigating to an instance of a specified object category within unseen environments. Although extensive investigations have been conducted on both end-to-end and modular-based, data-driven approaches, fully enabling an agent to comprehend the environment through perceptual knowledge and perform object-goal navigation as efficiently as humans remains a significant challenge. Recently, large language models have shown potential in this task, thanks to their powerful capabilities for knowledge extraction and integration. In this study, we propose a data-driven, modular-based approach, trained on a dataset that incorporates common-sense knowledge of object-to-room relationships extracted from a large language model. We utilize the multi-channel Swin-Unet architecture to conduct multi-task learning incorporating with multimodal inputs. The results in the Habitat simulator demonstrate that our framework outperforms the baseline by an average of 10.6% in the efficiency metric, Success weighted by Path Length (SPL). The real-world demonstration shows that the proposed approach can efficiently conduct this task by traversing several rooms. For more details and real-world demonstrations, please check our project webpage (https://sunleyuan.github.io/ObjectNav).