Simulation of the Human Arm Dynamics and Motor Control in Unstable Environments: A Comparison with Experimental Data

Abstract

The simulation of the human body dynamics offers a powerful way to investigate the interaction between humans and systems. This so-called physical human-machine interaction (HMI) is useful to consider during the development of new products like tools, cars or aircrafts, in order to increase comfort and safety. In this work, the Modelica Human Body Library is presented first. It is the only digital human model (DHM) for physical HMI available in the multi-domain modeling language Modelica so far. The model considers the human body dynamics as well as the motor control of the limbs. A special feature of the model is the human inspired control scheme, which takes into account effects such as human reaction time and limb stiffness adaption. After an introduction to the DHM, its capabilities are evaluated by comparing simulation results with experimental data from a experimental study. In the study, different unstable force fields were applied to the hands of subjects as they approached a target. The resulting hand trajectories were measured. As many control tasks for humans, especially while using tools, are characterized by unstable behavior, this study is well suited to evaluate a DHM. Comparison with the experimental data shows that the resulting hand trajectories in the simulation are comparable to the ones the subjects produced due to the force fields. This applies for both, the initial tries with big deviations as well as the tries after the subjects adapted their arm stiffness to the force field.