PulseAugur
中
实时 14:06:02
English(EN) GraphWorld: Long-Horizon Planning with World Models for End-to-End Autonomous Driving

新AI模型通过远期规划增强自动驾驶能力

作者 PulseAugur 编辑部 · [3 个来源] ·

研究人员开发了两种新框架 Metis 和 GraphWorld，旨在改进自动驾驶和城市导航系统。Metis 使用 Transformer 混合架构解耦视频生成和动作预测，提高了效率和泛化能力。GraphWorld 通过引入以自我为中心的交互图来建模代理关系并指导轨迹规划，专注于远期规划。这两种方法在各种基准测试中都展示了最先进的性能，降低了碰撞率并提高了复杂场景下的规划能力。 AI

影响 这些模型提高了自动驾驶系统的远期规划能力和效率，有可能在复杂场景下提高安全性和泛化能力。

排序理由 该集群包含两篇在 arXiv 上发表的研究论文，详细介绍了用于自动驾驶的新型 AI 模型。

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →

报道来源 [3]

  1. arXiv cs.CV TIER_1 English(EN) · Jingyu Li, Zhe Liu, Dongnan Hu, Junjie Wu, Zipei Ma, Wenxiao Wu, Chao Han, Zhihui Hao, Zhikang Liu, Kun Zhan, Jiankang Deng, Xiatian Zhu, Li Zhang ·

    Metis: A Generalizable and Efficient World-Action Model for Autonomous Driving and Urban Navigation

    arXiv:2606.15869v1 Announce Type: new Abstract: World action models~(WAMs) have shown great promise for autonomous driving and urban navigation. Built upon Vision-Language-Action models or video generation models, existing approaches suffer key limitations: (1) High inference lat…

  2. arXiv cs.CV TIER_1 English(EN) · Ziying Song, Caiyan Jia, Lin Liu, Lei Yang, Shengkai Zhang, Feiyang Jia, Fengda Zhao, Peiliang Wu, Shaoqing Xu, Chen Lv, Yadan Luo ·

    GraphWorld: Long-Horizon Planning with World Models for End-to-End Autonomous Driving

    arXiv:2606.16274v1 Announce Type: new Abstract: End-to-end autonomous driving has made significant progress by unifying perception, prediction, and planning within a single learning framework, achieving strong performance in short-horizon decision making. However, most existing E…

  3. arXiv cs.CV TIER_1 English(EN) · Yadan Luo ·

    GraphWorld: Long-Horizon Planning with World Models for End-to-End Autonomous Driving

    End-to-end autonomous driving has made significant progress by unifying perception, prediction, and planning within a single learning framework, achieving strong performance in short-horizon decision making. However, most existing E2E-AD methods remain confined to short-horizon p…

相关实体

相关话题