Receiver robot's motion for handing-over with a human

Abstract

Robots are required to assist our activities in daily life. In this paper, we focus on designing an algorithm to realize "handing-over" motion as a cooperative task between a human and a robot. We plan human-like robot motion to receive an object handed by a human. First, we analyze the handing-over motion performed by two humans. From the experimental results, we extract some characteristics in hand trajectory, velocity profile, and arm posture. We then confirm that human-like motion can be produced using these characteristics. Finally, we plan the robot motion with an instantaneous optimal control method, which evaluates the error (regarding the relative position, velocity and direction of the tip of hand) and the energy cost (joint torque) at each sampling step. In this method, we adopt a time-varying weight matrix that relates to how to reduce the error (or how the robot approaches the human) in order to give the robot's motion human-like characteristics. Simulation results show the validity of the proposed method.