2016 Volume 52 Issue 1 Pages 30-39
Kecskemethy et al showed that they can estimate the minute joint angle and joint displacement in the elbow joint from a pro-/supination angle by identifying three stiffness parameters of the forearm skeleton model they suggested. However, it is not so easy to estimate them because the identified stiffness parameters may frequently fall into a local solution by using the conventional least-squares method. In addition, although there exists the fact that the human joint system adjusts the joint stiffness with posture changes caused by some movement, the conventional method identifies the stiffness parameters as constants. Therefore, for estimating minute joint angle and joint displacement more accurately, this paper proposes a new method to identify stiffness parameters by use of a particle swarm optimization (PSO) method with taking into account such a human joint system characteristics. Consequently, it is shown that the proposed method can improve the fitting performance through the comparison between estimated values and experimental values obtained from CT images.