用时空对齐与多目标优化，提升自回归规划模型的驾驶表现
Autoregressive End-to-End Planning with Time-Invariant Spatial Alignment and Multi-Objective Policy Refinement
- 引入时不变空间对齐模块，统一未来每一步的视角
- 在NAVSIM上达89.8 PDMS，为自回归模型最优
- 通过行为级偏好优化，实现更精细的驾驶策略学习
自回归模型固有的序列建模能力使其成为自动驾驶端到端规划的强大基线。然而，其性能受限于时空错位问题：规划器需基于过去感知数据预测未来动作，导致代理视角不一致，制约了该方法的上限。为此，我们提出时不变空间对齐（TISA）模块，将初始环境特征投影到每个未来时刻的一致本体坐标系中，有效修正代理的视角，无需显式预测未来场景。此外，采用运动学动作预测头（加速度与航向角速率），确保轨迹物理可行性。最后，引入基于直接偏好优化（DPO）的多目标后训练阶段，突破纯模仿学习局限。该方法针对特定驾驶行为提供定向反馈，相比标准DPO的单一全局目标，提供了更细粒度的学习信号。模型在NAVSIM数据集上达到自回归模型中的最新纪录89.8 PDMS。视频演示见 https://tisa-dpo-e2e.github.io/。
原文摘要 · Abstract (English)
The inherent sequential modeling capabilities of autoregressive models make them a formidable baseline for end-to-end planning in autonomous driving. Nevertheless, their performance is constrained by a spatio-temporal misalignment, as the planner must condition future actions on past sensory data. This creates an inconsistent worldview, limiting the upper bound of performance for an otherwise powerful approach. To address this, we propose a Time-Invariant Spatial Alignment (TISA) module that learns to project initial environmental features into a consistent ego-centric frame for each future time step, effectively correcting the agent's worldview without explicit future scene prediction. In addition, we employ a kinematic action prediction head (i.e., acceleration and yaw rate) to ensure physically feasible trajectories. Finally, we introduce a multi-objective post-training stage using Direct Preference Optimization (DPO) to move beyond pure imitation. Our approach provides targeted feedback on specific driving behaviors, offering a more fine-grained learning signal than the single, overall objective used in standard DPO. Our model achieves a state-of-the-art 89.8 PDMS on the NAVSIM dataset among autoregressive models. The video document is available at https://tisa-dpo-e2e.github.io/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。