Stride recognition in the control concept of trans-femoral prosthesis

In order to improve the gait of disable people, a control concept of above-knee prosthesis was presented. The surface electromyography signal extracted from leg muscles was applied to recognize the phase of stride, and translated into on-off signal of self-lock control. A hybrid neural network-genetic algorithm was applied to describe the relation between surface electromyography signals and every phase of stride. The result of this study indicates that this algorithm can predict the highly nonlinear relation between different phase of stride and surface electromyography signals with higher identification rate.