Human like motion generation for ergonomic assessment - a muscle driven Digital Human Model using muscle synergies

In this paper, an approach for digital human modelling and an appropriate simulation environment is presented. The human body or parts of it are modeled as a multibody system with Hill-type muscle models as actuators, and human like motions are created with an optimal control (OC) framework. The focus is on inner (muscle) loads for ergonomic assessment and human like motion generation. A basic reaching test is set up in a motion lab where muscle activation signals via EMG and upper body trajectories are measured with a motion capture system when performing a multitude of different reaching tasks. The measured data is used for validation of the simulation results and additionally muscle synergies are extracted from the EMG signals. These synergies can be used as control parameters in the musculoskeletal model, whereby the number of actuators is reduced. This leads to computational speedup, reduction of anatomical redundancy and captures human muscle activation profiles.