On the Feasibility of Learning Finger-gaiting In-hand Manipulation with Intrinsic Sensing

Abstract-- Finger-gaiting manipulation is an important skill to achieve large-angle in-hand re-orientation of objects. However, achieving these gaits with arbitrary orientations of the hand is challenging due to the unstable nature of the task. In this work, we use model-free reinforcement learning (RL) to learn finger-gaiting only via precision grasps and demonstrate finger-gaiting for rotation about an axis purely using on-board proprioceptive and tactile feedback. To tackle the inherent Figure 1: A learned finger-gaiting policy that can continuously reorient instability of precision grasping, we propose the use of initial the target object about the hand z-axis. The policy only uses state distributions that enable effective exploration of the state sensing modalities intrinsic to the hand (such as touch and proprioception), space.