EMG signal based knee joint torque estimation

Although Lower Limb Robotic Rehabilitation device exhibit a great prospect in the rehabilitation of impaired limb, yet it has not been widely applied to clinical rehabilitation of the patient with impairment. This is mostly due to insufficient bidirectional information interaction between exoskeleton and patient. The intended action data that can be extracted from surface electromyography (sEMG) signal may include the intended posture, intended torque, intended knee joint angle and intended desired impedance of the patient. Capturing intended knee joint torque from sEMG signal is one of the necessary parameter to achieve a smooth Human Machine Interaction force in a multilayer control mechanism. In this paper, a new technique to estimate Knee joint torque using SVM has been proposed that has used wavelet feature. The estimator is able to estimate required knee joint torque to lift 5kg, 12kg and 19kg weight using extensor and flexor muscles. Based on weight-torque relationship, greater the weight, greater the torque is required to lift the weight. The estimator has classified required torque of 5kg, 12kg and 19kg with accuracy of 98.7296%, 86.0254% and 95.6443% respectively. The estimator can also be used to estimate torque about knee joint at different joint angle.