arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

IEEE RA-L

IEEE Robotics and Automation Letters · 期刊 · Robotics

2026-05-12 至 2026-05-12 共收录 3
2605.10063 2026-05-12 cs.RO

EFGCL: Learning Dynamic Motion through Spotting-Inspired External Force Guided Curriculum Learning

EFGCL:通过受启发于体操的外部力引导课程学习学习动态运动

Keita Yoneda, Kento Kawaharazuka, Kei Okada

机构 * Department of Mechano-Informatics, Graduate School of Information Science and Technology, The University of Tokyo(机械信息学系,信息科学和技术研究生院,东京大学) AI Center, Graduate School of Information Science and Technology, The University of Tokyo(人工智能中心,信息科学和技术研究生院,东京大学)

AI总结 本文提出EFGCL,一种基于物理指导原理的强化学习方法,通过引入外部辅助力加速四足机器人学习跳跃等动态全身运动,克服传统RL方法的不足。

Comments Accepted at RA-L 2026, website - https://keitayoneda.github.io/kleiyn-efgcl/, YouTube - https://youtu.be/sFK00hm14No/

Journal ref IEEE Robotics and Automation Letters (RA-L) 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.01008 2026-05-12 cs.RO

DexWrist: A Robotic Wrist for Constrained and Dynamic Manipulation

DexWrist:一种用于受限和动态操作的机械腕

Martin Peticco, Gabriella Ulloa, John Marangola, Nitish Dashora, Pulkit Agrawal

机构 * Improbable AI Lab, Massachusetts Institute of Technology(Improbable AI实验室,麻省理工学院)

AI总结 DexWrist通过结合准直接驱动和解耦并行运动机制,在紧凑设计中实现高扭矩和动态接触任务,提升了受限环境中的操作性能。

Comments 9 pages, 8 figures. Submitted to RA-L 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.09153 2026-05-12 cs.RO cs.AI

Beyond Self-Play: Hierarchical Reasoning for Continuous Motion in Closed-Loop Traffic Simulation

超越自我博弈:闭合环路交通模拟中的层次化推理

Weifan Zhang, Xiaofeng Zhao, Adel Bazzi, Mingrui Li, Yifan Wei, Dengfeng Sun

机构 * School of Aeronautics and Astronautics, Purdue University(普渡大学航空航天学院)

AI总结 本文提出一种层次化架构,结合高层多智能体交互推理与底层连续轨迹生成,以提升闭合环路交通模拟中代理的可扩展性和行为真实性,实验表明其在控制平滑性和安全性方面优于自我博弈和被动模仿基线。

Comments Submitted to IEEE Robotics and Automation Letters (RA-L)

详情

展开后加载摘要…

URL PDF HTML 收藏