Embodied Navigation Foundation Model
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916952780308480 |
|---|---|
| author | Zhang, Jiazhao Li, Anqi Qi, Yunpeng Li, Minghan Liu, Jiahang Wang, Shaoan Liu, Haoran Zhou, Gengze Wu, Yuze Li, Xingxing Fan, Yuxin Li, Wenjun Chen, Zhibo Gao, Fei Wu, Qi Zhang, Zhizheng Wang, He |
| author_facet | Zhang, Jiazhao Li, Anqi Qi, Yunpeng Li, Minghan Liu, Jiahang Wang, Shaoan Liu, Haoran Zhou, Gengze Wu, Yuze Li, Xingxing Fan, Yuxin Li, Wenjun Chen, Zhibo Gao, Fei Wu, Qi Zhang, Zhizheng Wang, He |
| contents | Navigation is a fundamental capability in embodied AI, representing the intelligence required to perceive and interact within physical environments following language instructions. Despite significant progress in large Vision-Language Models (VLMs), which exhibit remarkable zero-shot performance on general vision-language tasks, their generalization ability in embodied navigation remains largely confined to narrow task settings and embodiment-specific architectures. In this work, we introduce a cross-embodiment and cross-task Navigation Foundation Model (NavFoM), trained on eight million navigation samples that encompass quadrupeds, drones, wheeled robots, and vehicles, and spanning diverse tasks such as vision-and-language navigation, object searching, target tracking, and autonomous driving. NavFoM employs a unified architecture that processes multimodal navigation inputs from varying camera configurations and navigation horizons. To accommodate diverse camera setups and temporal horizons, NavFoM incorporates identifier tokens that embed camera view information of embodiments and the temporal context of tasks. Furthermore, to meet the demands of real-world deployment, NavFoM controls all observation tokens using a dynamically adjusted sampling strategy under a limited token length budget. Extensive evaluations on public benchmarks demonstrate that our model achieves state-of-the-art or highly competitive performance across multiple navigation tasks and embodiments without requiring task-specific fine-tuning. Additional real-world experiments further confirm the strong generalization capability and practical applicability of our approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_12129 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Embodied Navigation Foundation Model Zhang, Jiazhao Li, Anqi Qi, Yunpeng Li, Minghan Liu, Jiahang Wang, Shaoan Liu, Haoran Zhou, Gengze Wu, Yuze Li, Xingxing Fan, Yuxin Li, Wenjun Chen, Zhibo Gao, Fei Wu, Qi Zhang, Zhizheng Wang, He Robotics Navigation is a fundamental capability in embodied AI, representing the intelligence required to perceive and interact within physical environments following language instructions. Despite significant progress in large Vision-Language Models (VLMs), which exhibit remarkable zero-shot performance on general vision-language tasks, their generalization ability in embodied navigation remains largely confined to narrow task settings and embodiment-specific architectures. In this work, we introduce a cross-embodiment and cross-task Navigation Foundation Model (NavFoM), trained on eight million navigation samples that encompass quadrupeds, drones, wheeled robots, and vehicles, and spanning diverse tasks such as vision-and-language navigation, object searching, target tracking, and autonomous driving. NavFoM employs a unified architecture that processes multimodal navigation inputs from varying camera configurations and navigation horizons. To accommodate diverse camera setups and temporal horizons, NavFoM incorporates identifier tokens that embed camera view information of embodiments and the temporal context of tasks. Furthermore, to meet the demands of real-world deployment, NavFoM controls all observation tokens using a dynamically adjusted sampling strategy under a limited token length budget. Extensive evaluations on public benchmarks demonstrate that our model achieves state-of-the-art or highly competitive performance across multiple navigation tasks and embodiments without requiring task-specific fine-tuning. Additional real-world experiments further confirm the strong generalization capability and practical applicability of our approach. |
| title | Embodied Navigation Foundation Model |
| topic | Robotics |
| url | https://arxiv.org/abs/2509.12129 |