Autoregressive End-to-End Planning with Time-Invariant Spatial Alignment and Multi-Objective Policy Refinement
This work addresses a key limitation in autonomous driving planning, offering incremental improvements for more consistent and feasible trajectory generation.
The paper tackles the spatio-temporal misalignment problem in autoregressive end-to-end planning for autonomous driving by proposing a Time-Invariant Spatial Alignment module and multi-objective policy refinement, achieving a state-of-the-art 89.8 PDMS on the NAVSIM dataset.
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/.