自动驾驶

Branch-Stochastic Model Predictive Control for Motion Planning under Multi-Modal Uncertainty with Scenario Clustering

基于分支随机优化的运动规划在多模态不确定性下的场景聚类

Zekun Xing, Ramkrishna Chaudhari, Marion Leibold, Dirk Wollherr, Martin Buss

机构 * Chair of Automatic Control Engineering（自动控制工程教授会） 纠错

专题命中 规划控制 ：autonomous driving(abstract)；分类 cs.RO

AI总结 本文提出一种结合随机模型预测控制与分支结构的方法，用于在多模态不确定性下进行运动规划，通过场景聚类提高实时计算性能并减少保守性。

Comments This work has been accepted for presentation at IFAC World Congress 2026

DriveMA: Rethinking Language Interfaces in Driving VLAs with One-Step Meta-Actions

DriveMA: 重新思考驾驶VLAs中的语言接口以单步元动作

Weicheng Zheng, Yixin Huang, Qiao Sun, Derun Li, Hang zhao

专题命中 规划控制 ：end-to-end driving(abstract)；分类 cs.CV

AI总结 本文提出DriveMA，通过单步元动作替代传统的自然语言推理，解决了驾驶VLAs中语言接口的三个瓶颈问题，实现了更高效的端到端规划。

Comments We withdraw this submission because the current version contains a mismatch between the paper title, conceptual framing, and the intended contribution of the work. To avoid potential misunderstanding by readers, the authors have decided to withdraw this version and substantially revise the title, organization, and presentation before any future submission

Parallel OctoMapping: A Scalable Framework for Enhanced Path Planning in Autonomous Navigation

并行八叉树映射：一种用于自主导航中路径规划增强的可扩展框架

Yihui Mao, Tian Tan, Xuehui Shen, Warren E. Dixon, Rushikesh Kamalapurkar

机构 * Department of Mechanical and Aerospace Engineering, University of Florida（佛罗里达大学机械与航空航天工程系） ； Department of Electrical and Systems Engineering, University of Pennsylvania（宾夕法尼亚大学电气与系统工程系） 纠错

专题命中 规划控制 ：occupancy(abstract)；分类 cs.RO

AI总结 本文提出并行八叉树映射（POMP），一种高效的基于八叉树的映射技术，通过在固定占用网格分辨率下优化自由空间表示，提升路径规划效率和成功率，特别是在复杂环境中。