CarPlanner: Consistent Auto-regressive Trajectory Planning for Large-scale Reinforcement Learning in Autonomous Driving
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_ | 1866915211009589248 |
|---|---|
| author | Zhang, Dongkun Liang, Jiaming Guo, Ke Lu, Sha Wang, Qi Xiong, Rong Miao, Zhenwei Wang, Yue |
| author_facet | Zhang, Dongkun Liang, Jiaming Guo, Ke Lu, Sha Wang, Qi Xiong, Rong Miao, Zhenwei Wang, Yue |
| contents | Trajectory planning is vital for autonomous driving, ensuring safe and efficient navigation in complex environments. While recent learning-based methods, particularly reinforcement learning (RL), have shown promise in specific scenarios, RL planners struggle with training inefficiencies and managing large-scale, real-world driving scenarios. In this paper, we introduce \textbf{CarPlanner}, a \textbf{C}onsistent \textbf{a}uto-\textbf{r}egressive \textbf{Planner} that uses RL to generate multi-modal trajectories. The auto-regressive structure enables efficient large-scale RL training, while the incorporation of consistency ensures stable policy learning by maintaining coherent temporal consistency across time steps. Moreover, CarPlanner employs a generation-selection framework with an expert-guided reward function and an invariant-view module, simplifying RL training and enhancing policy performance. Extensive analysis demonstrates that our proposed RL framework effectively addresses the challenges of training efficiency and performance enhancement, positioning CarPlanner as a promising solution for trajectory planning in autonomous driving. To the best of our knowledge, we are the first to demonstrate that the RL-based planner can surpass both IL- and rule-based state-of-the-arts (SOTAs) on the challenging large-scale real-world dataset nuPlan. Our proposed CarPlanner surpasses RL-, IL-, and rule-based SOTA approaches within this demanding dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_19908 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CarPlanner: Consistent Auto-regressive Trajectory Planning for Large-scale Reinforcement Learning in Autonomous Driving Zhang, Dongkun Liang, Jiaming Guo, Ke Lu, Sha Wang, Qi Xiong, Rong Miao, Zhenwei Wang, Yue Robotics Computer Vision and Pattern Recognition Machine Learning Trajectory planning is vital for autonomous driving, ensuring safe and efficient navigation in complex environments. While recent learning-based methods, particularly reinforcement learning (RL), have shown promise in specific scenarios, RL planners struggle with training inefficiencies and managing large-scale, real-world driving scenarios. In this paper, we introduce \textbf{CarPlanner}, a \textbf{C}onsistent \textbf{a}uto-\textbf{r}egressive \textbf{Planner} that uses RL to generate multi-modal trajectories. The auto-regressive structure enables efficient large-scale RL training, while the incorporation of consistency ensures stable policy learning by maintaining coherent temporal consistency across time steps. Moreover, CarPlanner employs a generation-selection framework with an expert-guided reward function and an invariant-view module, simplifying RL training and enhancing policy performance. Extensive analysis demonstrates that our proposed RL framework effectively addresses the challenges of training efficiency and performance enhancement, positioning CarPlanner as a promising solution for trajectory planning in autonomous driving. To the best of our knowledge, we are the first to demonstrate that the RL-based planner can surpass both IL- and rule-based state-of-the-arts (SOTAs) on the challenging large-scale real-world dataset nuPlan. Our proposed CarPlanner surpasses RL-, IL-, and rule-based SOTA approaches within this demanding dataset. |
| title | CarPlanner: Consistent Auto-regressive Trajectory Planning for Large-scale Reinforcement Learning in Autonomous Driving |
| topic | Robotics Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2502.19908 |