Human Arm Motion Analysis Based on the Inverse Optimization Approach

Abstract

This study proposes a new inverse optimal control (IOC) framework for the optimal modelling of human arm motion. The proposed IOC is based on parameterized Lagrangians and is applied to a 3D arm motion. For the investigated arm motion, we suppose that the control strategy employed by the human motor control is to minimize a weighted sum of objective functions. In this context, we propose to recover from experimental data the weights of eight kinematics and dynamics biomechanical objective functions gathered into a basis of objective functions. Moreover, we extend the basis from 8 to 28 objective functions and assign their weight values by making use of the proposed IOC technique. The exhibited results for both bases are consistent with the literature and are obtained in only a couple of minutes and not hours as reported in the literature. Finally, this study gives clues on which basis of objective functions can be relevant for analyzing a human arm motion and/or can be extended to the use of more complex models with more degrees-of-freedom.