Simulation of Human Upright Standing Push-Recovery Based on OpenSim

Abstract

Investigating the human standing balance mechanisms under push-recovery task is of great importance to the study of biped robot balance control. Under human push-recovery mission, the passive stiffness, stretch reflex and short-range stiffness control mechanisms of human ankle joint are the main components in the internal mechanism of human body. To this end, this paper dedicates to evaluating the roles of the three aforementioned mechanisms during human upright standing push-recovery mission. Firstly, based on the simulation platform OpenSim4.0, this paper chooses a simplified lower-limb musculoskeletal model as the research object. Subsequently, this paper completes the design of the passive stiffness, stretch reflex and passive stiffness controller, and completes the static standing test and upright push-recovery simulation of the selected musculoskeletal model. Finally, in order to verify the effectiveness of the simulation, this paper uses electromyography, force plate and dynamic capture system to collect the relevant data of the human upright push-recovery. The experimental and simulation results reveal that the selected musculoskeletal model can basically simulate the process of human upright push-recovery under the joint actions of the three mechanisms noted above, which, to some degree, can reflect the effectiveness of the established method. Thus, the established method may provide some insights on the balance control of the bipedal robot.