AstraNav-World: World Model for Foresight Control and Consistency
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_ | 1866908946917228544 |
|---|---|
| author | Chen, Jintao Hu, Junjun Bai, Haochen Luo, Minghua Xue, Xinda Ren, Botao Bai, Chengyu Xie, Shichao Chen, Ziyi Liu, Fei Chu, Zedong Wu, Xiaolong Xu, Mu Zhang, Shanghang |
| author_facet | Chen, Jintao Hu, Junjun Bai, Haochen Luo, Minghua Xue, Xinda Ren, Botao Bai, Chengyu Xie, Shichao Chen, Ziyi Liu, Fei Chu, Zedong Wu, Xiaolong Xu, Mu Zhang, Shanghang |
| contents | Embodied navigation in open, dynamic environments demands accurate foresight of how the world will evolve and how actions will unfold over time. We propose AstraNav-World, an end-to-end world model that jointly reasons about future visual states and action sequences within a unified probabilistic framework. Our framework integrates a diffusion-based video generator with a vision-language policy, enabling synchronized rollouts where predicted scenes and planned actions are updated simultaneously. Training optimizes two complementary objectives: generating action-conditioned multi-step visual predictions and deriving trajectories conditioned on those predicted visuals. This bidirectional constraint makes visual predictions executable and keeps decisions grounded in physically consistent, task-relevant futures, mitigating cumulative errors common in decoupled "envision-then-plan" pipelines. Experiments across diverse embodied navigation benchmarks show improved trajectory accuracy and higher success rates. Ablations confirm the necessity of tight vision-action coupling and unified training, with either branch removal degrading both prediction quality and policy reliability. In real-world testing, AstraNav-World demonstrated exceptional zero-shot capabilities, adapting to previously unseen scenarios without any real-world fine-tuning. These results suggest that AstraNav-World captures transferable spatial understanding and planning-relevant navigation dynamics, rather than merely overfitting to simulation-specific data distribution. Overall, by unifying foresight vision and control within a single generative model, we move closer to reliable, interpretable, and general-purpose embodied agents that operate robustly in open-ended real-world settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_21714 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AstraNav-World: World Model for Foresight Control and Consistency Chen, Jintao Hu, Junjun Bai, Haochen Luo, Minghua Xue, Xinda Ren, Botao Bai, Chengyu Xie, Shichao Chen, Ziyi Liu, Fei Chu, Zedong Wu, Xiaolong Xu, Mu Zhang, Shanghang Computer Vision and Pattern Recognition Embodied navigation in open, dynamic environments demands accurate foresight of how the world will evolve and how actions will unfold over time. We propose AstraNav-World, an end-to-end world model that jointly reasons about future visual states and action sequences within a unified probabilistic framework. Our framework integrates a diffusion-based video generator with a vision-language policy, enabling synchronized rollouts where predicted scenes and planned actions are updated simultaneously. Training optimizes two complementary objectives: generating action-conditioned multi-step visual predictions and deriving trajectories conditioned on those predicted visuals. This bidirectional constraint makes visual predictions executable and keeps decisions grounded in physically consistent, task-relevant futures, mitigating cumulative errors common in decoupled "envision-then-plan" pipelines. Experiments across diverse embodied navigation benchmarks show improved trajectory accuracy and higher success rates. Ablations confirm the necessity of tight vision-action coupling and unified training, with either branch removal degrading both prediction quality and policy reliability. In real-world testing, AstraNav-World demonstrated exceptional zero-shot capabilities, adapting to previously unseen scenarios without any real-world fine-tuning. These results suggest that AstraNav-World captures transferable spatial understanding and planning-relevant navigation dynamics, rather than merely overfitting to simulation-specific data distribution. Overall, by unifying foresight vision and control within a single generative model, we move closer to reliable, interpretable, and general-purpose embodied agents that operate robustly in open-ended real-world settings. |
| title | AstraNav-World: World Model for Foresight Control and Consistency |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.21714 |