Robotic cloth manipulation for clothing assistance task using Dynamic Movement Primitives

摘要

The need of robotic clothing assistance in the field of assistive robotics is growing, as it is one of the most basic and essential assistance activities in daily life of elderly and disabled people. In this study we are investigating the applicability of using Dynamic Movement Primitives (DMP) as a task parameterization model for performing clothing assistance task. Robotic cloth manipulation task deals with putting a clothing article on both the arms. Robot trajectory varies significantly for various postures and also there can be various failure scenarios while doing cooperative manipulation with non-rigid and highly deformable clothing article. We have performed experiments on soft mannequin instead of human. Result shows that DMPs are able to generalize movement trajectory for modified posture.