arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

IEEE T-RO

IEEE Transactions on Robotics · 期刊 · Robotics

2025-12-09 至 2025-12-09 共收录 1
2407.04216 2025-12-09 cs.RO

Safe MPC Alignment with Human Directional Feedback

安全MPC对齐与人类方向反馈

Zhixian Xie, Wenlong Zhang, Yi Ren, Zhaoran Wang, George J. Pappas, Wanxin Jin

机构 * Intelligent Robotics and Interactive Systems (IRIS) Lab in the School for Engineering of Matter, Transport and Energy(智能机器人与交互系统实验室(IRIS)) Polytechnic School(理工学院) School for Engineering of Matter, Transport and Energy(工程物质、运输与能量学院) Departments of Industrial Engineering & Management Sciences(工业工程与管理科学系) Department of Electrical and Systems Engineering(电气与系统工程系)

AI总结 本文提出了一种基于人类方向反馈的可证明安全约束学习方法,通过少量修正即可使机器人有效学习安全约束。

Comments 20 pages, pre-print, submitted to TRO

详情

展开后加载摘要…

URL PDF HTML 收藏