Nav-$R^2$ Dual-Relation Reasoning for Generalizable Open-Vocabulary Object-Goal Navigation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xiang, Wentao, Zhang, Haokang, Yang, Tianhang, Chu, Zedong, Chu, Ruihang, Xie, Shichao, Yuan, Yujian, Sun, Jian, Gu, Zhining, Wang, Junjie, Wu, Xiaolong, Xu, Mu, Yang, Yujiu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914177321271296
author Xiang, Wentao
Zhang, Haokang
Yang, Tianhang
Chu, Zedong
Chu, Ruihang
Xie, Shichao
Yuan, Yujian
Sun, Jian
Gu, Zhining
Wang, Junjie
Wu, Xiaolong
Xu, Mu
Yang, Yujiu
author_facet Xiang, Wentao
Zhang, Haokang
Yang, Tianhang
Chu, Zedong
Chu, Ruihang
Xie, Shichao
Yuan, Yujian
Sun, Jian
Gu, Zhining
Wang, Junjie
Wu, Xiaolong
Xu, Mu
Yang, Yujiu
contents Object-goal navigation in open-vocabulary settings requires agents to locate novel objects in unseen environments, yet existing approaches suffer from opaque decision-making processes and low success rate on locating unseen objects. To address these challenges, we propose Nav-$R^2$, a framework that explicitly models two critical types of relationships, target-environment modeling and environment-action planning, through structured Chain-of-Thought (CoT) reasoning coupled with a Similarity-Aware Memory. We construct a Nav$R^2$-CoT dataset that teaches the model to perceive the environment, focus on target-related objects in the surrounding context and finally make future action plans. Our SA-Mem preserves the most target-relevant and current observation-relevant features from both temporal and semantic perspectives by compressing video frames and fusing historical observations, while introducing no additional parameters. Compared to previous methods, Nav-R^2 achieves state-of-the-art performance in localizing unseen objects through a streamlined and efficient pipeline, avoiding overfitting to seen object categories while maintaining real-time inference at 2Hz. Resources will be made publicly available at \href{https://github.com/AMAP-EAI/Nav-R2}{github link}.
format Preprint
id arxiv_https___arxiv_org_abs_2512_02400
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Nav-$R^2$ Dual-Relation Reasoning for Generalizable Open-Vocabulary Object-Goal Navigation
Xiang, Wentao
Zhang, Haokang
Yang, Tianhang
Chu, Zedong
Chu, Ruihang
Xie, Shichao
Yuan, Yujian
Sun, Jian
Gu, Zhining
Wang, Junjie
Wu, Xiaolong
Xu, Mu
Yang, Yujiu
Computer Vision and Pattern Recognition
Object-goal navigation in open-vocabulary settings requires agents to locate novel objects in unseen environments, yet existing approaches suffer from opaque decision-making processes and low success rate on locating unseen objects. To address these challenges, we propose Nav-$R^2$, a framework that explicitly models two critical types of relationships, target-environment modeling and environment-action planning, through structured Chain-of-Thought (CoT) reasoning coupled with a Similarity-Aware Memory. We construct a Nav$R^2$-CoT dataset that teaches the model to perceive the environment, focus on target-related objects in the surrounding context and finally make future action plans. Our SA-Mem preserves the most target-relevant and current observation-relevant features from both temporal and semantic perspectives by compressing video frames and fusing historical observations, while introducing no additional parameters. Compared to previous methods, Nav-R^2 achieves state-of-the-art performance in localizing unseen objects through a streamlined and efficient pipeline, avoiding overfitting to seen object categories while maintaining real-time inference at 2Hz. Resources will be made publicly available at \href{https://github.com/AMAP-EAI/Nav-R2}{github link}.
title Nav-$R^2$ Dual-Relation Reasoning for Generalizable Open-Vocabulary Object-Goal Navigation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.02400