GMM-based Single-joint Angle Estimation using EMG signals

In this paper we explored the possibility to use Electromyography (EMG) to train a Gaussian Mixture Model (GMM) in order to estimate the bending angle of a single human joint. In particular, EMG signals from eight leg muscles and the knee joint angle were acquired during a kick task from three different subjects. GMM was validated on new unseen data and the classification performances were compared with respect to the number of EMG channels and the number of collected trials used during the training phase. Results showed that our framework is able to achieve high performances even using few EMG channels (normalized mean square error: 0.96, 0.98, 0.98 for the three subjects, respectively) and with a small training dataset, opening new and interesting perspectives for applications to humanoid robots and exoskeletons.