Font Size:
Prediction of Human Elbow Joint Torque Based on Improved BP Neural Network
Last modified: 2015-06-28
Abstract
Pathological tremor brings too much inconvenience to patients in life and work. For better tremor suppression, a suitable biomechanical model must be established. Based on the Hill skeleton-muscle model, quantitative relations between EMG and static torque of elbow joint can be identified with improved neural network. The weights of improved neural network are adjusted according to the need, and muscle activation grade is confirmed. Through this method, a biomechanical model is established. Using OpenSim software we can simulate the drive of skeleton model by EMG signals and the validity of the model is tested by experiment.
Keywords
skeleton-muscle model, EMG(Electromyographic signal) , joint torque, improved neural network(NN)
An account with this site is required in order to view papers. Click here to create an account.