Robotic based simulation algorithm to predict optimized compensatory motion of transradial prosthesis users

Poor performance while using a prosthesis can stress the joints excessively and lead to chronic injuries or rejection of the prosthesis. In order to improve upper limb prosthetic performance, compensatory motion of amputees is often compared with the movements of able-bodied persons. This technique does not consider the limitations of the degrees of freedom (DoF) of a prosthesis. This paper presents a Robotic Human Body Model (RHBM) simulation algorithm that demonstrates an optimized amputee performance with an upper limb prosthesis given its restrictions. Seven males, amputated below the left elbow participated in the study. Each subject used a body powered prosthesis to perform five activities of daily living (ADL) and motion was collected and analyzed using a marker based motion capture system. The motion capture (MoCap) data were compared with a normalized simulation based on the weighted least norm (WLN) solution to identify excessive motion. The weights were calculated from four optimal users minimizing the root mean square (RMS) error and compared to the motion capture data of all subjects demonstrating the accuracy and robustness of the RHBM. Excessive motion in one subject due to the tight fitting of the device was identified. Smaller deviations of the other subjects from the simulation were also identified. Using the RHBM simulation, excessive movements were located which may have clinical implications regarding prosthesis fitting and training.