Computationally Efficient Personalized EMG-Driven Musculoskeletal Model of Wrist Joint

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

Automated summary

Myoelectric control is a promising method for commanding exoskeletons by interpreting human intentions. This article proposes a computationally efficient optimization method for personalized EMG-driven musculoskeletal model based on wrist motion. Experimental evaluations show that the proposed method outperforms other optimization algorithms, making the use of musculoskeletal model in practical applications more feasible.