Adaptive Manipulation of Constrained Objects by Soft-Fingered Robot Hands

The purpose of this paper is to report on our on-going project for developing multi-fingered robotic hand with soft and thin skin which is capable of performing various skillful grasping and manipulation tasks. Our hand is basically controlled by a position-based control algorithm. However, it can realize desired grasping force due to the soft skin while maintaining stability of the system. For an experimental validation of this concept, we conducted several manipulation tasks, such as manipulating of a cubic object and turning the crank of a small fishing rod. These results show that this two-fingered hand can perform various manipulation tasks of objects in three dimensional space stably and firmly. We will also present an adaptive control scheme for manipulation of constrained objects whose orientation is not well known beforehand. Adaptation is performed through coordination of hand and arm portions like humans do.