DreamNav: A Trajectory-Based Imaginative Framework for Zero-Shot Vision-and-Language Navigation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Yunheng, Fang, Yuetong, Wang, Taowen, Feng, Yixiao, Tan, Yawen, Zhang, Shuning, Liu, Peiran, Ji, Yiding, Xu, Renjing
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909786737475584
author Wang, Yunheng
Fang, Yuetong
Wang, Taowen
Feng, Yixiao
Tan, Yawen
Zhang, Shuning
Liu, Peiran
Ji, Yiding
Xu, Renjing
author_facet Wang, Yunheng
Fang, Yuetong
Wang, Taowen
Feng, Yixiao
Tan, Yawen
Zhang, Shuning
Liu, Peiran
Ji, Yiding
Xu, Renjing
contents Vision-and-Language Navigation in Continuous Environments (VLN-CE), which links language instructions to perception and control in the real world, is a core capability of embodied robots. Recently, large-scale pretrained foundation models have been leveraged as shared priors for perception, reasoning, and action, enabling zero-shot VLN without task-specific training. However, existing zero-shot VLN methods depend on costly perception and passive scene understanding, collapsing control to point-level choices. As a result, they are expensive to deploy, misaligned in action semantics, and short-sighted in planning. To address these issues, we present DreamNav that focuses on the following three aspects: (1) for reducing sensory cost, our EgoView Corrector aligns viewpoints and stabilizes egocentric perception; (2) instead of point-level actions, our Trajectory Predictor favors global trajectory-level planning to better align with instruction semantics; and (3) to enable anticipatory and long-horizon planning, we propose an Imagination Predictor to endow the agent with proactive thinking capability. On VLN-CE and real-world tests, DreamNav sets a new zero-shot state-of-the-art (SOTA), outperforming the strongest egocentric baseline with extra information by up to 7.49\% and 18.15\% in terms of SR and SPL metrics. To our knowledge, this is the first zero-shot VLN method to unify trajectory-level planning and active imagination while using only egocentric inputs.
format Preprint
id arxiv_https___arxiv_org_abs_2509_11197
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DreamNav: A Trajectory-Based Imaginative Framework for Zero-Shot Vision-and-Language Navigation
Wang, Yunheng
Fang, Yuetong
Wang, Taowen
Feng, Yixiao
Tan, Yawen
Zhang, Shuning
Liu, Peiran
Ji, Yiding
Xu, Renjing
Robotics
Artificial Intelligence
Computation and Language
Computer Vision and Pattern Recognition
Vision-and-Language Navigation in Continuous Environments (VLN-CE), which links language instructions to perception and control in the real world, is a core capability of embodied robots. Recently, large-scale pretrained foundation models have been leveraged as shared priors for perception, reasoning, and action, enabling zero-shot VLN without task-specific training. However, existing zero-shot VLN methods depend on costly perception and passive scene understanding, collapsing control to point-level choices. As a result, they are expensive to deploy, misaligned in action semantics, and short-sighted in planning. To address these issues, we present DreamNav that focuses on the following three aspects: (1) for reducing sensory cost, our EgoView Corrector aligns viewpoints and stabilizes egocentric perception; (2) instead of point-level actions, our Trajectory Predictor favors global trajectory-level planning to better align with instruction semantics; and (3) to enable anticipatory and long-horizon planning, we propose an Imagination Predictor to endow the agent with proactive thinking capability. On VLN-CE and real-world tests, DreamNav sets a new zero-shot state-of-the-art (SOTA), outperforming the strongest egocentric baseline with extra information by up to 7.49\% and 18.15\% in terms of SR and SPL metrics. To our knowledge, this is the first zero-shot VLN method to unify trajectory-level planning and active imagination while using only egocentric inputs.
title DreamNav: A Trajectory-Based Imaginative Framework for Zero-Shot Vision-and-Language Navigation
topic Robotics
Artificial Intelligence
Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.11197