CONSTRAINT-AWARE ZERO-SHOT VISION-LANGUAGE NAVIGATION IN CONTINUOUS ENVIRONMENTS

Kehan Chen, Dong An, Yan Huang, Rongtao Xu, Yifei Su, Yonggen Ling, Ian Reid, Liang Wang

Keywords: Instruction Navigation, training free, unexplored environment

Abstract:

We present Constraint-Aware Navigator (CA-Nav), a zero-shot approach for Vision-Language Navigation in Continuous Environments (VLN-CE). CA-Nav reframes the zero-shot VLN-CE task as a sequential constraint-aware sub-instruction completion process, continuously translating sub-instructions into navigation plans via a cross-modal value map. Central to our approach are two modules namely Constraint-aware Sub-instruction Manager (CSM) and Constraint-aware Value Mapper (CVM). CSM defines the completion criteria of decomposed sub-instructions as constraints and tracks navigation progress by switching sub-instructions in a constraint-aware manner. Based on the constraints identified by CSM, CVM builds a value map on-the-fly and refines it using superpixel clustering to enhance navigation stability. CA-Nav achieves the state-of-the-art performance on two VLN-CE benchmarks, surpassing the compared best method by 12% on R2R-CE and 13% on RxR-CE in terms of Success Rate on the validation unseen split. Furthermore, CA-Nav demonstrates its effectiveness in real-world robot deployments across diverse indoor scenes and instructions.

Primary Area: applications to robotics, autonomy, planning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 7158