Contrastive Learning for Enhancing Robust Scene Transfer in Vision-based Agile Flight
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917600376651776 |
|---|---|
| author | Xing, Jiaxu Bauersfeld, Leonard Song, Yunlong Xing, Chunwei Scaramuzza, Davide |
| author_facet | Xing, Jiaxu Bauersfeld, Leonard Song, Yunlong Xing, Chunwei Scaramuzza, Davide |
| contents | Scene transfer for vision-based mobile robotics applications is a highly relevant and challenging problem. The utility of a robot greatly depends on its ability to perform a task in the real world, outside of a well-controlled lab environment. Existing scene transfer end-to-end policy learning approaches often suffer from poor sample efficiency or limited generalization capabilities, making them unsuitable for mobile robotics applications. This work proposes an adaptive multi-pair contrastive learning strategy for visual representation learning that enables zero-shot scene transfer and real-world deployment. Control policies relying on the embedding are able to operate in unseen environments without the need for finetuning in the deployment environment. We demonstrate the performance of our approach on the task of agile, vision-based quadrotor flight. Extensive simulation and real-world experiments demonstrate that our approach successfully generalizes beyond the training domain and outperforms all baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2309_09865 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Contrastive Learning for Enhancing Robust Scene Transfer in Vision-based Agile Flight Xing, Jiaxu Bauersfeld, Leonard Song, Yunlong Xing, Chunwei Scaramuzza, Davide Robotics Computer Vision and Pattern Recognition Scene transfer for vision-based mobile robotics applications is a highly relevant and challenging problem. The utility of a robot greatly depends on its ability to perform a task in the real world, outside of a well-controlled lab environment. Existing scene transfer end-to-end policy learning approaches often suffer from poor sample efficiency or limited generalization capabilities, making them unsuitable for mobile robotics applications. This work proposes an adaptive multi-pair contrastive learning strategy for visual representation learning that enables zero-shot scene transfer and real-world deployment. Control policies relying on the embedding are able to operate in unseen environments without the need for finetuning in the deployment environment. We demonstrate the performance of our approach on the task of agile, vision-based quadrotor flight. Extensive simulation and real-world experiments demonstrate that our approach successfully generalizes beyond the training domain and outperforms all baselines. |
| title | Contrastive Learning for Enhancing Robust Scene Transfer in Vision-based Agile Flight |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2309.09865 |