Planning and Control of Anthropomorphic Robotic Hands using Synergies

In this work we present a method for the generation of hand postural synergies for a series of precision grasp types to be used in dextrous robot hands. Our method records the robot hand motions while teleoperated by human subjects via a dataglove, doing different grasp types on a series of objects. This exploits the fact that humans automatically compensate for calibration errors on the glove to robot mapping. The method is applied to the Shadow Robot Hand and to the iCub Hand. Despite the hands having a significantly different number of actuators, our analysis demonstrates that the effective number of degrees-of-freedom of the tested precision grasps is lower than the number of actuators in both hands. Therefore, the existing actuators are enough to drive the hands with realistic human-like postures and in-hand movements. In line with previous works on general grasp synergies, our work confirms that human precision grasps also lie on lowdimensional spaces.