新的自动驾驶模型使用世界建模实现更安全、更鲁棒的规划 · 跟踪2个来源

报道来源 [4]

1. arXiv cs.CV TIER_1 English(EN) · Daniele De Martini · 2026-06-30 15:36

   PriorEye: Geospatial Visual Priors for End-to-End Autonomous Driving

   Most end-to-end autonomous driving methods rely solely on instantaneous sensor observations, limiting them to reactive behavior without the anticipatory foresight human drivers employ through prior experience. We introduce geospatial visual priors, street-level visual context anc…

2. arXiv cs.CV TIER_1 English(EN) · Junjie Cheng, Ruiqi Song, Ye Wu, Nanxing Zeng, Ximiao Li, Yunfeng Ai · 2026-06-30 04:00

   OWMDrive：通过4D占用世界模型实现因果感知端到端自动驾驶

   arXiv:2606.30421v1 Announce Type: new Abstract: Autonomous driving systems are steadily moving toward end-to-end paradigms to mitigate the limited adaptability of rule-based pipelines in complex traffic environments. However, most existing learning-based methods still make decisi…

3. arXiv cs.CV TIER_1 English(EN) · Zihao Sheng, Xin Ye, Jingru Luo, Sikai Chen, Liu Ren · 2026-06-30 04:00

   ExploreVLA：端到端自动驾驶的密集世界建模与探索

   arXiv:2604.02714v2 Announce Type: replace Abstract: End-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning in…

4. arXiv cs.CV TIER_1 English(EN) · Yunfeng Ai · 2026-06-29 15:05

   OWMDrive：通过4D占用世界模型实现因果感知端到端自动驾驶

   Autonomous driving systems are steadily moving toward end-to-end paradigms to mitigate the limited adaptability of rule-based pipelines in complex traffic environments. However, most existing learning-based methods still make decisions from static representations of the current s…