Learning Reflexive Behavior for Contact-Rich Manipulation

1. 12:00
   arXiv · Robotics

   Learning Reflexive Behavior for Contact-Rich Manipulation

   研究者在仿真中基于弹簧、平面和导轨三种交互基元训练了一个本体感知反射策略，将任务空间命令映射为关节位置目标，无需直接力或几何测量。策略作为冻结执行层置于高层控制器之下，在双臂箱体提升中使力保持在阈值以下，在粗糙表面跟随中低于 10 N 参考值，在 0.02 mm 轴孔插入中将平均估计接触力降至基线的一半以下，硬件成功率从至多 22% 提升至 36-58%。

Current heat 7·Comparable peak 10（Oct 5）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.