Learning to Drive Anywhere with Model-Based Reannotation
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_ | 1866914166863822848 |
|---|---|
| author | Hirose, Noriaki Ignatova, Lydia Stachowicz, Kyle Glossop, Catherine Levine, Sergey Shah, Dhruv |
| author_facet | Hirose, Noriaki Ignatova, Lydia Stachowicz, Kyle Glossop, Catherine Levine, Sergey Shah, Dhruv |
| contents | Developing broadly generalizable visual navigation policies for robots is a significant challenge, primarily constrained by the availability of large-scale, diverse training data. While curated datasets collected by researchers offer high quality, their limited size restricts policy generalization. To overcome this, we explore leveraging abundant, passively collected data sources, including large volumes of crowd-sourced teleoperation data and unlabeled YouTube videos, despite their potential for lower quality or missing action labels. We propose Model-Based ReAnnotation (MBRA), a framework that utilizes a learned short-horizon, model-based expert model to relabel or generate high-quality actions for these passive datasets. This relabeled data is then distilled into LogoNav, a long-horizon navigation policy conditioned on visual goals or GPS waypoints. We demonstrate that LogoNav, trained using MBRA-processed data, achieves state-of-the-art performance, enabling robust navigation over distances exceeding 300 meters in previously unseen indoor and outdoor environments. Our extensive real-world evaluations, conducted across a fleet of robots (including quadrupeds) in six cities on three continents, validate the policy's ability to generalize and navigate effectively even amidst pedestrians in crowded settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_05592 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning to Drive Anywhere with Model-Based Reannotation Hirose, Noriaki Ignatova, Lydia Stachowicz, Kyle Glossop, Catherine Levine, Sergey Shah, Dhruv Robotics Computer Vision and Pattern Recognition Machine Learning Systems and Control Developing broadly generalizable visual navigation policies for robots is a significant challenge, primarily constrained by the availability of large-scale, diverse training data. While curated datasets collected by researchers offer high quality, their limited size restricts policy generalization. To overcome this, we explore leveraging abundant, passively collected data sources, including large volumes of crowd-sourced teleoperation data and unlabeled YouTube videos, despite their potential for lower quality or missing action labels. We propose Model-Based ReAnnotation (MBRA), a framework that utilizes a learned short-horizon, model-based expert model to relabel or generate high-quality actions for these passive datasets. This relabeled data is then distilled into LogoNav, a long-horizon navigation policy conditioned on visual goals or GPS waypoints. We demonstrate that LogoNav, trained using MBRA-processed data, achieves state-of-the-art performance, enabling robust navigation over distances exceeding 300 meters in previously unseen indoor and outdoor environments. Our extensive real-world evaluations, conducted across a fleet of robots (including quadrupeds) in six cities on three continents, validate the policy's ability to generalize and navigate effectively even amidst pedestrians in crowded settings. |
| title | Learning to Drive Anywhere with Model-Based Reannotation |
| topic | Robotics Computer Vision and Pattern Recognition Machine Learning Systems and Control |
| url | https://arxiv.org/abs/2505.05592 |