Minimum jerk for human catching movements in 3D

Abstract

To investigate fast human reaching movements in 3D, we asked 11 right-handed persons to catch a tennis ball while we tracked the movements of their arms. To ensure consistent trajectories of the ball, we used a catapult to throw the ball from three different positions. Tangential velocity profiles of the hand were in general bell-shaped and hand movements in 3D coincided with well known results for 2D point-to-point movements such as minimum jerk theory or the 2/3rd power law. Furthermore, two phases, consisting of fast reaching and slower fine movements at the end of hand placement could clearly be seen. The aim of this study was to find a way to generate human-like (catching) trajectories for a humanoid robot.