Estimation of Wrist Joint Moment by Fusing Musculoskeletal Model and Muscle Synergy for Neuromuscular Interface

The joint moment provides specific information of human motion. It plays an important role as an advanced interfacing technology in robot assistant systems for elderly and disabled people. The surface electromyography (sEMG) signals are usually affected by the adjacent muscles. And muscle tendon units in the same muscle show different activation characteristics with different movement patterns. It is significant to calculate the contribution degree of signals from multi-channels to different movements. In this paper, the wrist joint moment, in particular the flexion and extension of wrist (WFE), is estimated by a novel approach that combines muscle synergy theory with musculoskeletal model. sEMG signal and joint angle of WFE were collected and input to the estimation model to calculate the joint moment. Experiments on five healthy subjects have demonstrated that, the estimation result of the proposed approach is more accurate with higher average correlation coefficient (CC) and lower normalized root-mean-square error (NRMSE) between estimated moment and reference moment.