Generalizing Motion Planners with Mixture of Experts for Autonomous Driving
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_ | 1866913566905335808 |
|---|---|
| author | Sun, Qiao Wang, Huimin Zhan, Jiahao Nie, Fan Wen, Xin Xu, Leimeng Zhan, Kun Jia, Peng Lang, Xianpeng Zhao, Hang |
| author_facet | Sun, Qiao Wang, Huimin Zhan, Jiahao Nie, Fan Wen, Xin Xu, Leimeng Zhan, Kun Jia, Peng Lang, Xianpeng Zhao, Hang |
| contents | Large real-world driving datasets have sparked significant research into various aspects of data-driven motion planners for autonomous driving. These include data augmentation, model architecture, reward design, training strategies, and planner pipelines. These planners promise better generalizations on complicated and few-shot cases than previous methods. However, experiment results show that many of these approaches produce limited generalization abilities in planning performance due to overly complex designs or training paradigms. In this paper, we review and benchmark previous methods focusing on generalizations. The experimental results indicate that as models are appropriately scaled, many design elements become redundant. We introduce StateTransformer-2 (STR2), a scalable, decoder-only motion planner that uses a Vision Transformer (ViT) encoder and a mixture-of-experts (MoE) causal Transformer architecture. The MoE backbone addresses modality collapse and reward balancing by expert routing during training. Extensive experiments on the NuPlan dataset show that our method generalizes better than previous approaches across different test sets and closed-loop simulations. Furthermore, we assess its scalability on billions of real-world urban driving scenarios, demonstrating consistent accuracy improvements as both data and model size grow. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_15774 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Generalizing Motion Planners with Mixture of Experts for Autonomous Driving Sun, Qiao Wang, Huimin Zhan, Jiahao Nie, Fan Wen, Xin Xu, Leimeng Zhan, Kun Jia, Peng Lang, Xianpeng Zhao, Hang Robotics Computer Vision and Pattern Recognition Large real-world driving datasets have sparked significant research into various aspects of data-driven motion planners for autonomous driving. These include data augmentation, model architecture, reward design, training strategies, and planner pipelines. These planners promise better generalizations on complicated and few-shot cases than previous methods. However, experiment results show that many of these approaches produce limited generalization abilities in planning performance due to overly complex designs or training paradigms. In this paper, we review and benchmark previous methods focusing on generalizations. The experimental results indicate that as models are appropriately scaled, many design elements become redundant. We introduce StateTransformer-2 (STR2), a scalable, decoder-only motion planner that uses a Vision Transformer (ViT) encoder and a mixture-of-experts (MoE) causal Transformer architecture. The MoE backbone addresses modality collapse and reward balancing by expert routing during training. Extensive experiments on the NuPlan dataset show that our method generalizes better than previous approaches across different test sets and closed-loop simulations. Furthermore, we assess its scalability on billions of real-world urban driving scenarios, demonstrating consistent accuracy improvements as both data and model size grow. |
| title | Generalizing Motion Planners with Mixture of Experts for Autonomous Driving |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.15774 |