arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.05083cs.CVcs.AIcs.RO
收藏

在寄存器上驾驶

Driving on Registers

LIGM, ENPC, IP Paris, UGE, CNRS(LIGM,ENPC,IP Paris,UGE,CNRS);Sorbonne Université, CNRS, ISIR(Sorbonne Université,CNRS,ISIR)
Ellington Kirby, Alexandre Boulch, Yihong Xu, Yuan Yin, Gilles Puy, Éloi Zablocki, Andrei Bursuc, Spyros Gidaris, Renaud Marlet, Florent Bartoccioni, Anh-Quan C… 展开作者

Ellington Kirby, Alexandre Boulch, Yihong Xu, Yuan Yin, Gilles Puy, Éloi Zablocki, Andrei Bursuc, Spyros Gidaris, Renaud Marlet, Florent Bartoccioni, Anh-Quan Cao, Nermin Samet, Tuan-Hung VU, Matthieu Cord

AI总结:

DrivoR通过引入相机感知的寄存器标记,结合轻量级Transformer解码器,在保持准确性的同时高效实现端到端自动驾驶。

摘要:

我们提出了DrivoR,一种简单且高效的基于Transformer的架构,用于端到端的自动驾驶。我们的方法基于预训练的视觉Transformer(ViTs),并引入了相机感知的寄存器标记,将多相机特征压缩成紧凑的场景表示,显著减少下游计算而不牺牲准确性。这些标记驱动两个轻量级的Transformer解码器,生成并评分候选轨迹。评分解码器学会模仿一个理想情况,并预测可解释的子分数,代表诸如安全、舒适和效率等方面,使在推理过程中能够实现基于行为的驾驶。尽管其设计极为简洁,DrivoR在NAVSIM-v1、NAVSIM-v2以及逼真闭环HUGSIM基准测试中均优于或匹配了强大的当代基线。我们的结果表明,纯Transformer架构结合针对性的标记压缩,足以实现准确、高效且适应性强的端到端驾驶。代码和检查点将通过项目页面提供。

英文摘要:

We present DrivoR, a simple and efficient transformer-based architecture for end-to-end autonomous driving. Our approach builds on pretrained Vision Transformers (ViTs) and introduces camera-aware register tokens that compress multi-camera features into a compact scene representation, significantly reducing downstream computation without sacrificing accuracy. These tokens drive two lightweight transformer decoders that generate and then score candidate trajectories. The scoring decoder learns to mimic an oracle and predicts interpretable sub-scores representing aspects such as safety, comfort, and efficiency, enabling behavior-conditioned driving at inference. Despite its minimal design, DrivoR outperforms or matches strong contemporary baselines across NAVSIM-v1, NAVSIM-v2, and the photorealistic closed-loop HUGSIM benchmark. Our results show that a pure-transformer architecture, combined with targeted token compression, is sufficient for accurate, efficient, and adaptive end-to-end driving. Code and checkpoints will be made available via the project page.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑