Dexterous In-Hand Object Pose Reaching with Learned Safety Filters
Sim-only RL recipe for in-hand object pose reaching on a LEAP hand, zero-shot to hardware; reports 95.2% of 4,800 unseen goals reached and a tunable residual safety filter that cuts drops by 75%.在仿真中无需人类数据训练 LEAP 灵巧手的手内物体位姿到达策略，零样本迁移到真机；报告在 4800 个未见目标上 95.2% 到达，可调节的残差安全滤波器使掉落减少 75%。
BibTeX
@misc{chuang2026dexterous,
title = {{Dexterous In-Hand Object Pose Reaching with Learned Safety Filters}},
author = {Ian Chuang and Hiya Shah and Vamsi Chilamkur and Iman Soltani and Claire Tomlin},
year = {2026},
eprint = {2610.37652},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.37652},
}Manipulation & Dexterous Hands操作与灵巧手Robot Learning: RL & Imitation机器人学习：强化学习与模仿学习Sim-to-Real & Simulators仿真与 Sim-to-RealDexterous hand灵巧手dexterousin-handleap-handsim2realsafetyreal