Flexible control of a grasping object with posture error

Proposes adjusting methods of a regrasping strategy according to position and posture errors of a grasping object. We have already derived a generation algorithm for the regrasping strategy by evolutionary programming. The regrasping strategy however becomes inapplicable when the relative position and posture between a hand and an object is not same as those in numerical simulation. We therefore propose some adjusting methods which can change the intermediate position and posture of a grasping object. This method can avoid the limitation of joint work area and fingertip's slip so that it could extend the tolerable area of the position and posture error. We show its effectiveness for the regrasping motion with numerical simulations and experiments.