Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.07181 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910048365576192 |
|---|---|
| author | Zhou, Jiaxu Wang, Shaobo Yang, Zhiyuan Yu, Zhenjun Li, Tao |
| author_facet | Zhou, Jiaxu Wang, Shaobo Yang, Zhiyuan Yu, Zhenjun Li, Tao |
| contents | Vision-Language Navigation aims to enable agents to understand natural language instructions and carry out appropriate navigation actions in real-world environments. Most work focuses on indoor settings, with little research in complex outdoor scenes. Current UAV Vision-and-Language Navigation models typically act as black boxes without explicit reasoning. We introduce FreeFly-thinking, an end-to-end VLN framework that converts the UAV agent's egocentric images and language instructions into a series of actions, inspired by environment of urban architecture proposed by OpenFly. We first construct a UAV dataset for navigation task, and then performing natural language chain of thought. We adopt a two-stage training strategy: Supervised fine-tuning and Reinforcement fine-tuning. Experiments on unseen test demonstrate a strong performance, presenting robustness and efficiency in UAV navigation issue. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_07181 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FreeFly-Thinking : Aligning Chain-of-Thought Reasoning with Continuous UAV Navigation Zhou, Jiaxu Wang, Shaobo Yang, Zhiyuan Yu, Zhenjun Li, Tao Computer Vision and Pattern Recognition Vision-Language Navigation aims to enable agents to understand natural language instructions and carry out appropriate navigation actions in real-world environments. Most work focuses on indoor settings, with little research in complex outdoor scenes. Current UAV Vision-and-Language Navigation models typically act as black boxes without explicit reasoning. We introduce FreeFly-thinking, an end-to-end VLN framework that converts the UAV agent's egocentric images and language instructions into a series of actions, inspired by environment of urban architecture proposed by OpenFly. We first construct a UAV dataset for navigation task, and then performing natural language chain of thought. We adopt a two-stage training strategy: Supervised fine-tuning and Reinforcement fine-tuning. Experiments on unseen test demonstrate a strong performance, presenting robustness and efficiency in UAV navigation issue. |
| title | FreeFly-Thinking : Aligning Chain-of-Thought Reasoning with Continuous UAV Navigation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.07181 |