Stumble mode identification of prosthesis based on the Dempster-Shafer evidential theory

In order to guarantee the safety of the prosthesis users, effective identification of stumble situation in the walking process is the key point. In this paper, the movement information of the human body is collected by using accelerator and gyroscope installed on the prosthesis. And the correlation coefficients of different signals are obtained via the correlation analysis. The correlation coefficients of the two kinds of sensors are fused by Dempster-Shafer evidential theory. The experiments show that the recognition of prosthesis stumble based on D-S evidential theory method is effective and accurate. It has great significance for the control of prosthesis.