Hydra-MDP: End-to-end Multimodal Planning with Multi-target Hydra-Distillation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912007984250880 |
|---|---|
| author | Li, Zhenxin Li, Kailin Wang, Shihao Lan, Shiyi Yu, Zhiding Ji, Yishen Li, Zhiqi Zhu, Ziyue Kautz, Jan Wu, Zuxuan Jiang, Yu-Gang Alvarez, Jose M. |
| author_facet | Li, Zhenxin Li, Kailin Wang, Shihao Lan, Shiyi Yu, Zhiding Ji, Yishen Li, Zhiqi Zhu, Ziyue Kautz, Jan Wu, Zuxuan Jiang, Yu-Gang Alvarez, Jose M. |
| contents | We propose Hydra-MDP, a novel paradigm employing multiple teachers in a teacher-student model. This approach uses knowledge distillation from both human and rule-based teachers to train the student model, which features a multi-head decoder to learn diverse trajectory candidates tailored to various evaluation metrics. With the knowledge of rule-based teachers, Hydra-MDP learns how the environment influences the planning in an end-to-end manner instead of resorting to non-differentiable post-processing. This method achieves the $1^{st}$ place in the Navsim challenge, demonstrating significant improvements in generalization across diverse driving environments and conditions. More details by visiting \url{https://github.com/NVlabs/Hydra-MDP}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_06978 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Hydra-MDP: End-to-end Multimodal Planning with Multi-target Hydra-Distillation Li, Zhenxin Li, Kailin Wang, Shihao Lan, Shiyi Yu, Zhiding Ji, Yishen Li, Zhiqi Zhu, Ziyue Kautz, Jan Wu, Zuxuan Jiang, Yu-Gang Alvarez, Jose M. Computer Vision and Pattern Recognition We propose Hydra-MDP, a novel paradigm employing multiple teachers in a teacher-student model. This approach uses knowledge distillation from both human and rule-based teachers to train the student model, which features a multi-head decoder to learn diverse trajectory candidates tailored to various evaluation metrics. With the knowledge of rule-based teachers, Hydra-MDP learns how the environment influences the planning in an end-to-end manner instead of resorting to non-differentiable post-processing. This method achieves the $1^{st}$ place in the Navsim challenge, demonstrating significant improvements in generalization across diverse driving environments and conditions. More details by visiting \url{https://github.com/NVlabs/Hydra-MDP}. |
| title | Hydra-MDP: End-to-end Multimodal Planning with Multi-target Hydra-Distillation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.06978 |