Computationally Efficient Personalized EMG-Driven Musculoskeletal Model of Wrist Joint

Myoelectric control has gained much attention which translates the human intentions into commands for exoskeletons. The electromyogram (EMG)-driven musculoskeletal (MSK) model shows prominent performance given its ability to interpret underlying neuromechanical processes among MSK system. This model-based scheme contains inherent physiological parameters, e.g., isometric muscle force, tendon slack length, or optimal fiber need be tailored each individual via minimizing differences between experimental measurement and estimation. However, creation of personalized EMG-driven through evolutionary algorithms is time-consuming, hurdling use in practical scenarios. article proposes a computationally efficient optimization method estimate subject-specific parameters wrist based on direct collocation (DC) method. By constraining variables experimentally measured EMG signals introducing variables, fast achieved by identifying discretized at grid simultaneously. Experimental evaluations 12 healthy subjects are performed. Results demonstrate that proposed outperforms baseline used literature, including genetic algorithm, simulated annealing algorithm (SA), particle swarm (PSO) algorithm. DC possibility alleviate costly procedure facilitate applications.