Robotic Teaching By Exploiting the Nervous System's Adaptive Mechanisms

James L. Patton, Ferdinando A. Mussa-Ivaldi

ASSISTIVE TECHNOLOGY RESEARCH SERIES(2001)

引用 28|浏览4
暂无评分
摘要
We tested the idea of teaching arm movements implicitly, based on recent studies of adaptive control. Eight subjects were repeatedly exposed to predetermined training forces during movement so they would develop an expectation and cancel them out. Forces were unexpectedly removed to cause a predictable after-effect. Forces were designed so that a "desired" movement would result as this after-effect. This method does not require any explicit instructions about the desired movement. Results showed a significant shift of trajectories toward the desired, although the after-effect washed out following the removal of the forces in about 75 movements. This approach may be effective for teaching healthy movement patterns to brain injured patients because their movement alterations can be perceived as improvements.
更多
查看译文
关键词
adaptive control
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要