Rethinking Imitation-based Planner for Autonomous Driving
This work addresses the problem of evaluating and improving imitation-based planners for autonomous driving, offering a strong baseline and insights for researchers, though it is incremental in nature.
The authors tackled the lack of standardized benchmarks for imitation-based driving planners by using the nuPlan dataset to study key features and data augmentation, proposing PlanTF, which achieves competitive performance with state-of-the-art methods and better generalization in long-tail cases.
In recent years, imitation-based driving planners have reported considerable success. However, due to the absence of a standardized benchmark, the effectiveness of various designs remains unclear. The newly released nuPlan addresses this issue by offering a large-scale real-world dataset and a standardized closed-loop benchmark for equitable comparisons. Utilizing this platform, we conduct a comprehensive study on two fundamental yet underexplored aspects of imitation-based planners: the essential features for ego planning and the effective data augmentation techniques to reduce compounding errors. Furthermore, we highlight an imitation gap that has been overlooked by current learning systems. Finally, integrating our findings, we propose a strong baseline model-PlanTF. Our results demonstrate that a well-designed, purely imitation-based planner can achieve highly competitive performance compared to state-of-the-art methods involving hand-crafted rules and exhibit superior generalization capabilities in long-tail cases. Our models and benchmarks are publicly available. Project website https://jchengai.github.io/planTF.