Dynamic force/torque equilibrium for stable grasping by a triple robotic fingers system

This paper proposes a stable object grasping method to realize dynamic force/torque equilibrium by using a triple robotic fingers system with soft and deformable hemispherical fingertips. In the authors´ previous works, ¿Blind Grasping¿ control scheme, which realizes stable object grasping without use of any external sensing such as vision, force, or tactile sensing in the case of using a pair of robot fingers, has been proposed. This control methodology is based on a unique configuration of human hand, called ¿Fingers-Thumb Opposability¿. In this paper, a ternary finger in addition to a pair of fingers is introduced not only to expand a stable region of grasping, but also to enhance dexterity and versatility of the multi-fingered robotic hand system. To this end, a ¿Blind Grasping¿ manner is modified in order to install it in the triple fingers system. First, dynamics of the triple robotic fingers system and a grasped object with considering rolling constraints is modeled, and a control input based on the blind grasping manner is designed. Next, the closed-loop dynamics is derived and a stability analysis is shown briefly. Finally, its usefulness is discussed through numerical simulation results.