UC Irvine

This dissertation presents a kinematic synthesis method developed to achieve a mechanical system that guides a natural ankle trajectory for a human walking gait. This methodology was the result of exploring hybrid task position optimization for a Watt I six-bar linkage, optimization of a four-bar linkage for 9 point path synthesis, and finally a homotopy directed optimization for a Stephenson III six-bar path generator. The new homotopy directed optimization technique was applied to 205 data points that defined the human ankle trajectory. The data was interpolated using B-splines, and an objective function, obtained from the six-bar linkage loop equations, evaluated the distance between the desired trajectory and the linkage trajectory. The result was 148 designs for 23 trajectories. A clustering algorithm was used to show that these designs are effectively the same. A complete solid model, together with a cam mechanism to control the foot orientation angle is presented. This resulted in both a new six-bar linkage synthesis methodology, as well as a unique linkage system to support natural movement of the human lower limb. The similarity of the linkage designs was exploited to introduce an adjustment that allows changes of the ankle trajectory over its natural variation.