パラメータ空間表現によるロボット軌道のロバスト性評価とロバストなロボット軌道の導出法

概要

論文の詳細を見る
We have already proposed the method to transfer human skill to a robot through direct teaching. By using the method, we can automatically extract human skill including sensory feedback to adapt to environmental variation. However, we have not discussed what trajectory is robust against environmental variation. In this paper, we discuss the robustness condition of robot trajectories against environmental variation and how to derive the robust trajectory. We propose a method to determine whether a trajectory is robust or not. The method considers the probability of the environment variation uses the parameter space expression of trajectory in order to derive the robustness condition. We confirm the effectiveness of the proposed method through a case study on a typical assembly task, peg-in-hole.